Assessment and management of post-COVID-19 pulmonary complications: a rapid review

Abstract

The rising global prevalence of post-COVID-19 condition (PCC) underscores the substantial and ongoing burden faced by individuals following severe acute respiratory syndrome coronavirus 2 infection. The volume of emerging evidence regarding pulmonary-related PCC complications highlights the urgent need for current, evidence-informed guidelines to ensure timely assessment and effective treatment for those affected by PCC. Thus, the aim of this review was to synthesise existing research on the management and treatment of pulmonary complications in individuals with PCC. A rapid review of published and grey literature focused on pulmonary-related PCC complications was completed in November 2023 and updated in June 2025, in accordance with PRISMA (preferred reporting items for systematic reviews and meta-analyses) guidelines. We identified 73 unique articles, including 12 guidance documents, 24 secondary studies (including 11 systematic reviews with meta-analyses, eight systematic reviews and three scoping reviews) and 37 primary research studies (13 randomised controlled trials) and narratively synthesised their findings. Guidance documents addressed workup and management for pulmonary-related PCC complications, recommending the use of pulmonary function testing with diffusing capacity and the importance of ruling out other conditions. Although evidence regarding the use of medical and pharmacological interventions for treatment of pulmonary-related PCC complications were limited and inconclusive, the current evidence base suggested potential effectiveness of a multidisciplinary rehabilitation approach for pulmonary-related PCC treatment, involving specialist consultations and tailored rehabilitation programmes. The heterogeneity in study quality and risk of bias warrants cautious interpretation of the findings. The current evidence and evolving healthcare landscape suggest the need for updated, evidence-informed clinical guidance.

Shareable abstract

For pulmonary-related post-COVID-19 condition thorough assessment is recommended to rule out other conditions. Growing evidence supports multidisciplinary rehabilitation for management, while evidence for pharmacological treatments remains limited. https://bit.ly/4pOj57D

Introduction

To date, more than 1.4 million Canadians have reported experiencing post-COVID-19 condition (PCC) [1], which is the continuation or development of new symptoms 3 months following initial severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, lasting at least 2 months, which cannot be explained by alternate reasons [2]. PCC can have profound multi-system impact(s), with specific symptoms ranging from shortness of breath, fatigue, cognitive disturbances, headache, myalgia, anxiety or depression, and decreased functional capacity [3, 4]. Notably, pulmonary complications are commonly seen in PCC [5, 6], and the United States Centers for Disease Control reports pulmonary complications to be twice as prevalent in survivors of SARS-CoV-2 infection than in those who were not infected [7]. For example, in an international survey (comprising mostly of US (41%) and UK (35%) respondents), dyspnoea and cough were reported in 40% and 20% of people with PCC, respectively [8], and in Canada, a 2023 report by Statistics Canada identified dyspnoea as the third most common reported symptom (in 28% of people with PCC) [9]. PCC-related dyspnoea has been linked to poorer sleep, mood, health-related quality of life (HRQoL) and impaired daily activities compared to individuals with PCC but without dyspnoea [10]. Pulmonary complications reported in people with PCC include lung perfusion abnormalities, airway epithelial damage, air trapping, impaired lung function and systemic inflammation [11].

As understanding of the pathophysiology and epidemiology of pulmonary complications among those with PCC rapidly evolves, so does the evidence base with respect to therapies and clinical best practice guidance. At the time of initial drafting of this work, the most recent guideline document for the management of pulmonary complications in PCC was published in 2022 by Yelin et al. [12], which included evidence from 27 publications concerning clinical management, pulmonary rehabilitation (PR), PCC fibrotic changes, persistent cough and dyspnoea. Unfortunately, the evidence base at the time of publication only allowed authors to provide conditional guidance on the assessment of PCC, without any treatment recommendations [12]. Since the publication of this early guidance document [12], there have been numerous primary studies, systematic reviews and guidelines published on the assessment, workup and/or management of pulmonary complications in PCC, including the Canadian Guidelines for PCC summarised by Falcone et al. [13].

Given the rapidly evolving literature on PCC and its pulmonary complications, as well as its growing clinical burden, there is a pressing need to revise, summarise and update current guidance to inform evidence-based recommendations that can guide patient care.

Research objective and questions

The objective of this rapid literature review was to identify, appraise and summarise the literature concerning the management and treatment of pulmonary complications for those with PCC. Specifically, the following research questions were addressed:

• 1) What are the expert recommendations/guidelines for the work up and or management of pulmonary complications in PCC including: a) post-COVID-19 organising pneumonia; b) post-COVID-19 fibrotic changes; c) post-COVID-19 pulmonary embolism; d) post-COVID-19 exercise limitation and/or desaturation; e) post-COVID-19 airways hyperreactivity; f) persistent dyspnoea associated with PCC; g) persistent cough (with or without sputum) associated with PCC; and h) persistent noncardiac chest pain associated with PCC.

• 2) What are the existing observational, case–control/series or randomised controlled trial (RCT) data on medical interventions for any of the above complications?

• 3) What are the existing observational, case–control/series or RCT data on rehabilitation interventions for any of the above complications?

Methods

Study design

A rapid review of the published and grey literature for PCC and its pulmonary complications was completed, in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [14]. This review was registered on PROSPERO (CRD42024471757).

Search strategy

An initial electronic database search of the literature published from 1 January 2020 to 16 November 2023 was completed using the Medline, Embase and Cochrane Central Register of Controlled Trials databases and was updated in June 2025. The search strategies were developed by two information specialists and are provided in appendix 1. To identify literature on PCC and its pulmonary complications and treatment, a comprehensive set of keyword terms and Medical Subject Headings (MeSH) were utilised. The first component searched for “covid-19” and was combined with a comprehensive set of keyword terms and MeSH terms for “persistent symptoms”. The second component used a comprehensive set of keyword terms and MeSH terms for “long covid” and was combined with a comprehensive set of keyword terms and MeSH terms for specific types of respiratory symptoms (e.g. organising pneumonia and pulmonary embolism) and treatments (e.g. respiratory muscle training). The search strategy did not include keywords or MeSH terms to limit the type of study designs retrieved in the search. The search was limited to the English language; no other limitations were applied.

Grey literature search

To supplement the evidence from the published literature, a grey literature search was completed. A Google search for relevant grey literature was performed by an information specialist (Deanna Townsend, Knowledge Resource Services, Alberta Health Services, Edmonton, Alberta)), including purposive searches of the websites of relevant organisations (e.g. CADTH COVID-19 Grey Literature Resources) and International and Canadian health authorities (e.g. Alberta Health Services, British Columbia Provincial Health Services Authority). The comprehensive grey literature search is provided in appendix 2. The list was then screened by one reviewer (W. Michaelchuk) using the inclusion and exclusion criteria outlined in appendix 3 to identify eligible records.

Study selection

All titles and abstracts were screened independently by one author (D. Chiarieri-Hirsch, E. Giroux, J. Shatto or W. Michaelchuk) using Covidence online software (Covidence, Melbourne, Australia, 2024). The first 80 titles and abstracts were screened in quadruplicate by reviewers (D. Chiarieri-Hirsch, E. Giroux, J. Shatto, W. Michaelchuk and L.J.J. Soril) to ensure consistent application of the inclusion and exclusion criteria (appendix 3). Studies were included if 1) they featured adult (≥18 years of age) patients with PCC (persistent symptoms ≥3 months post-acute infection, lasting ≥2 months that cannot be explained by an alternate diagnosis) experiencing any pulmonary complications and 2) medical or rehabilitative interventions were involved in the workup, treatment and/or management of their pulmonary complications. Guidance documents, including expert (e.g. certified health professionals, panels, review boards) guidelines, recommendations or position statements were eligible for inclusion. Eligible secondary research study designs included umbrella reviews, systematic reviews and meta-analyses, scoping reviews, and rapid reviews. The reference lists of systematic reviews were also hand-searched for relevant articles. Eligible primary research study designs included RCTs, quasi-experimental, observational cohort studies (prospective or retrospective), case–control studies and case series/reports of 20 participants or more. While relevant pre-print and grey literature records meeting all other inclusion criteria were also eligible for inclusion, commentaries, opinion pieces, narrative review articles and letters to the editor were not eligible for inclusion. To minimise the impact of duplication of findings, any primary studies identified in our literature search that were already represented within identified systematic, scoping, umbrella or rapid reviews were also excluded. Abstracts included by any reviewer proceeded to full-text review. Full-text articles were also screened independently by one author (D. Chiarieri-Hirsch, E. Giroux, J. Shatto or W. Michaelchuk) and included based on the inclusion and exclusion criteria outlined in appendix 3. Any uncertainty was resolved through discussion or through consultation with L.J.J. Soril.

Risk of bias and quality assessments

Aligning with the evidence-informed guidance provided by the Cochrane Rapid Reviews Methods Group [15], risk of bias and quality assessments were performed for all included studies. The Evidence Project risk of bias tool [16], Risk of Bias Assessment Tool for Systematic Review (ROBIS) [17], and the Appraisal of Guidelines for Research and Evaluation II Instrument (AGREE II) [18], were used to assess risk of bias for primary studies, reviews and guidance documents, respectively. All risk of bias and quality assessments were also synthesised in tables and summarised narratively.

Data extraction and synthesis

Relevant data from the included published and grey literature were extracted independently by W. Michaelchuk, D. Chiarieri-Hirsch, E. Giroux or J. Shatto using a standardised data extraction template. Jurisdiction or country, and publication type (e.g. primary study, secondary study or guidance document) were consistently extracted for each study. For primary studies, detailed study design, intervention characteristics (i.e. type, description, length, setting), presence of a comparison/control, PCC definition used, list and prevalence of PCC pulmonary complications, participant demographic characteristics (i.e. number of participants, sex, race, ethnicity, comorbidities, age, loss to follow-up), and outcome measures were extracted. For guidance documents, the type of document, professionals involved, study objective, pulmonary-related PCC complications discussed, recommendations made, and any specific considerations (e.g. implementation considerations, clinical nuances) were extracted. For secondary studies, the study design (e.g. rapid review, systematic review or scoping review), objective, number of studies, presence of comparisons (if applicable), intervention(s) discussed, PCC definition used, pulmonary-related PCC complications discussed, study outcomes and main findings were extracted. The collective information was synthesised in tables and described narratively below.

Results

Search results

Through the electronic database searches of the published literature (appendix 1), 9955 unique records were identified and reviewed for inclusion. Of these, 469 proceeded to full-text review and 69 articles were included in the final narrative synthesis. Reasons for exclusion reasons are detailed in the PRISMA diagram (supplementary figure S1), with the most common being an ineligible patient population (n=151), such as studies involving paediatric patients or those not meeting our definition of PCC. Grey literature searches (appendix 2) and handsearching identified 553 additional records, leading to 39 additional records being reviewed and four records being included. In total, 73 unique records were included, 12 answering research question 1 [12, 13, 19–27], 16 answering research question 2 [28–43] and 48 answering research question 3 [6, 38, 42–87]. There were three secondary studies, Ora et al. [38], Zeraatkar et al. [42] and Zheng et al. [43], that reviewed information relevant for both research questions 2 and 3. Thus, they are reflected in the counts for the number of articles addressing research questions 2 and 3. 17 relevant primary studies identified through this literature search were already included within the secondary studies (i.e. systematic, scoping and rapid reviews) and were therefore excluded from the primary study sections to avoid over-representing their findings [88–104]. As a result of this de-duplication, all primary studies found in tables 2–5 in this manuscript represent studies that have not previously been synthesised in existing reviews. Appendix 4 lists all primary studies identified in the included secondary research articles and indicates how many reviews each primary study appeared in. Supplementary figure S1 shows the flow diagram for study selection in alignment with the PRISMA guidelines [14] and supplementary figure S2 provides an outline of practical recommendations based on the findings from this review.

TABLE 2.

Characteristics of included studies for question 2

First author [ref.], year, country	Study design	Participants, n	Participant characteristics			Post-COVID-19 condition (PCC)		Summary of risk of bias or study quality
			Age, mean±sd years	Female, n (%)	Comorbidities, n (%)	Definition of PCC	Pulmonary complications, n (%)
Primary research (n=9)
Doehner [23]  2024  Germany	Prospective non-RCT	145 (n=70 intervention group; n=75 control group)	Intervention: 50.8±10.6 Control: 55.1±11.7	Intervention: 52 (74%) Control: 56 (75%)	IHHT group: CAD 5 (7.1%), heart failure 1 (1.4%), arterial hypertension 21 (30.0%), atrial fibrillation 4 (5.7%), diabetes mellitus 4 (5.7%), COPD 2 (2.9%), current smoker 5 (7.1%), hypothyroidism 7 (10.0%), depression 3 (4.3%) Control group: CAD 15 (20.0%), heart failure 3 (4.0%), arterial hypertension 31 (41.3%), atrial fibrillation 1 (1.3%), diabetes mellitus 6 (8.0%), COPD 10 (13.3%), current smoker 7 (9.3%), hypothyroidism 5 (6.7%), depression 1 (1.3%)	Persistent symptoms ≥12 weeks post-infection	Signs of respiratory insufficiency (Borg dyspnoea scale ≥3 or SpO2 ≤85%): 145 (100%)	See appendix 5
Geng [29] 2024  USA	RCT	155 (n=102 intervention group; n=53 control group)	Intervention (median (IQR)): 45 (35–56) Control: 41 (31–45)	Intervention: 61 (60%) Control: 31 (59%)	Depression: intervention 24 (23.5%), control 13 (24.5%) Allergies: intervention 17 (16.7%), control 12 (22.6%) Asthma: intervention 15 (14.7%), control 13 (24.5%) Anxiety: intervention 15 (14.7%), control 8 (15.1%) GORD: intervention 15 (14.7%), control 6 (11.3%)	Persistent symptoms ≥90 days post-infection	Shortness of breath: intervention 46.1%, control 52.8%	See appendix 5
He [30] 2024  China	RCT	73 (n=35 intervention group; n=38 control group)	Intervention (median (IQR)): 71 (67.3–74.5) Control: 68 (62.5–72.0)	Intervention: 9 (26%) Control: 17 (45%)	Hypertension: intervention 20 (57%), control 19 (50%) Diabetes mellitus: intervention 14 (40%), control 7 (18%) Coronary heart disease: intervention 8 (23%), control 8 (21%) Heart failure: intervention 2 (6%), control 1 (3%) Cerebrovascular disease: intervention 2 (6%), control 4 (11%) COPD: intervention 7 (20%), control 7 (18%) ILD: intervention 5 (14%), control 5 (13%) Chronic nephrosis: intervention 2 (6%), control 5 (13%)	Persistent symptoms ≥12 weeks post-infection	Cough: intervention 31 (89%), control 33 (87%) Chest pain: intervention 5 (14%), control 7 (18%) Chest tightness: intervention 20 (57%), control 21 (55%) Expectoration: intervention 23 (66%), control 30 (79%) Shortness of breath: intervention 17 (51%), control 19 (50%)	See appendix 5
Kjellberg [31]  2025  Sweden	RCT	80 (n=39 intervention group; n=41 control group)	Intervention: 41.1±10.1 Control: 41.4±8.2	Intervention: 32 (80%) Control: 33 (83%)	NR	Persistent symptoms ≥12 weeks post-infection	Cough or dyspnoea: intervention 28 (70%), control 27 (68%)	See appendix 5
Ntiamoah [32]  2024  USA	Retrospective cohort study	75 (n=30 steroid group; n=45 nonsteroid group)	56±12 (steroid group: 56±12; nonsteroid group: 57±13)	47 (63%) (steroid group: 20 (67%); nonsteroid group 27 (60%))	Asthma 16 (21%), rheumatologic disease 9 (12%), history of pulmonary embolism 9 (12%), COPD 7 (7%), CAD 5 (7%), history of malignancy 6 (8%), history of any transplantation 3 (4%), ILD 2 (3%), heart failure 2 (3%), pulmonary hypertension 1 (1%)	Not explicitly reported; average of 121 days post-infection	Ground-glass opacities 67 (91%), consolidations 22 (29%), reticulations 17 (23%), fibrosis 4 (5%)	See appendix 5
Sipetic [33]  2023  Serbia	Cross-sectional prospective observational study	871	45±15	534 (62%)	NR#	Persistent symptoms ≥12 weeks post-infection	Dyspnoea 122 (14.0%), cough 298 (34.2%), chest pain (pressure) 119 (13.7%)	See appendix 5
Torres [34]  2024  USA	Retrospective observational study	28	57±11	11 (39%)	Hypertension 7 (25%), obesity 7 (25%), hypothyroidism 6 (21%), CAD 4 (14%), diabetes 5 (17%), connective tissue disease 3 (10%)	Persistent symptoms ≥12 weeks post-infection	Post-infection ILD: 28 (100%)	See appendix 5
Vagedes [35]  2025  Germany	Retrospective case series	46	50±13	40 (87%)	Secondary diagnoses: circulatory system 7 (15.2%), mental and behavioural disorders 6 (13.0%), endocrine, nutritional and metabolic 5 (10.9%) Tertiary diagnoses: endocrine, nutritional and metabolic 6 (13.0%), musculoskeletal system and connective tissue 5 (10.9%), circulatory system 4 (8.7%)	Symptoms that persist for ≥12 weeks after acute infection	Respiratory symptoms during acute infection: 32 (69.6%), respiratory symptoms on hospitalisation	See appendix 5
Zha [36] 2024  China	RCT	95 (n=47 intervention, n=48 control)	Intervention (median (IQR)): 57.5 (53.0–68.0) Control: 59.5 (53.0–70.0)	Intervention: 33 (70%), control: 29 (60%)	Hypertension: intervention 15 (31.9%), control 13 (27.1%) Diabetes mellitus: intervention 4 (8.5%), control 3 (6.3%) Coronary heart disease: intervention 5 (10.6%), control 4 (8.3%) Heart failure: intervention 1 (2.1%), control 1 (2.1%) Cerebrovascular disease: intervention 2 (4.3%), control 2 (4.2%) COPD: intervention 3 (6.8%), control 7 (14.6%) Asthma: intervention 1 (2.1%), control 1 (2.1%) ILD: intervention 4 (8.5%), control 3 (6.3%)	Persistent symptoms ≥12 weeks post-infection	Dyspnoea 95 (100%)	See appendix 5
Secondary research (n=7)
Li [37] 2023  China	Systematic review	11 studies (respiratory relevant)	NR#	NR#	NR#	>4 weeks after onset of acute illness	Dyspnoea (NR)	High risk of bias
Ora [38] 2023¶  Italy	Systematic review	35 studies (nine relevant to medical interventions)	NR#	NR#	NR#	Persistent symptoms >4 weeks after an initial infection	Dyspnoea (NR), fibrosis (NR), unspecified pulmonary symptoms (NR), exercise limitation (NR)	High risk of bias
Seo [39] 2025  South Korea	Scoping review	Three studies (respiratory relevant)	NR#	NR#	NR#	Multiple definitions (i.e. >4 weeks after an initial infection and persistent symptoms ≥12 weeks post-acute infection)	Pulmonary fibrosis (NR)	High risk of bias
Tsang [40] 2025  China	Systematic review and meta-analysis	Five studies (respiratory relevant)	Range: 18–75	658 (59.4%)	NR#	≥3 months post-acute infection	Cardiopulmonary dysfunction: 200 (18.1%), “qi” deficiency in the lung: 331 (29.9%)	High risk of bias
Yong [41] 2023  Malaysia	Systematic and critical review	26 studies (n=4 completed, n=22 ongoing), 6689 participants (n=1484 completed, n=5205 ongoing)	NR for overall number Completed range: 45–60	NR for overall number Completed: 53%	NR#	Symptoms that persist for ≥12 weeks after acute infection	Fibrotic changes (NR), dyspnoea (NR)	Low risk of bias
Zeraatkar [42]  2024¶  Canada	Living systematic review	One study (pulmonary relevant)	NR#	NR#	NR#	≥3 months after initial infection	Pulmonary symptoms (NR)	Low risk of bias
Zheng [43]  2022¶  UK	Systematic review and meta-analysis	One study (relevant to medical interventions)	NR#	NR#	NR#	Ongoing symptoms lasting >4 weeks after acute infection	Dyspnoea (NR), reduced exercise capacity (NR)	Low risk of bias

CAD: coronary artery disease; GORD: gastro-oesophageal reflux disease; IHHT: intermittent hypoxia/hyperoxia training; ILD: interstitial lung disease; IQR: interquartile range; NR: not reported; RCT: randomised controlled trial; S_pO2: peripheral oxygen saturation. ^#: Numbers or proportions not reported. ^¶: Study also reported in table 4; this study provides data to support research questions 2 and 3.

TABLE 5.

Summary of study findings for question 3

First author [ref.], year, country	Characteristics of rehabilitation programme					Length of follow-up	Outcome measures	Summary of findings#
	Rehabilitation type	Setting	Components	Frequency and duration	Timing of initiation
Primary research (n=28)
Acharya [58]  2025  India	Multicomponent exercise rehabilitation	Outpatient	1) Pulmonary rehabilitation: supervised breathing exercises, aerobic training, and inspiratory muscle training 2) Neurological rehabilitation consisted of cognitive therapy sessions and memory exercises 3) Psychological support involved weekly counselling, mindfulness-based cognitive therapy and pharmacological management	12-week programme; duration and frequency not reported	NR	12 weeks	1) Lung function 2) Quality of life (EQ-5D-5L)	1) FEV1 % pred (mean±sd) improved significantly following intervention (baseline: 78.4± 6.5, post: 85.7± 5.9, p<0.001); FVC % pred improved significantly following intervention (baseline: 80.3±7.2, post: 86.9±6.1, p<0.001); DLCO % pred improved significantly following intervention (baseline: 65.3±8.7, post: 76.4±7.2, p<0.001) 2) EQ-5D-5L index improved significantly following intervention (baseline: 0.64±0.15, post: 0.81±0.12, p<0.001)
Altmann [60]  2023  Germany	Specialised long-COVID rehabilitation	Inpatient (cardiac rehabilitation facility)	1) Endurance training (bicycle ergometer exercise training, aqua training and terrain training with self-management using pre-tested or calculated heart rate or perceived exertion using the Borg scale) 2) Meditative group sessions 3) Self-managed and device-based muscular training therapy, educational programmes and 4) Oxygen therapy 5) Psychological assistance 6) Attention to social needs post-discharge 7) Comorbidity management as needed 8) Education 9) Intermittent hypoxia–hyperoxia therapy	4–5 weeks duration Device-based inspiratory breathing therapy (daily), oxygen therapy (as needed), intermittent hypoxia–hyperoxia therapy (three times per week for 45 min)	Time (mean (min–max)) since SARS-CoV-2 infection incidence was 2.0 (1–3) months for COVID-19 and 10.1 (4–20) months for long COVID	NR	1) Vital capacity 2) Inspiratory muscle strength 3) Minimum peripheral oxygen saturation 4) Bicycle exercise performance 5) Fit for work 6) Return to work in 6 months 7) Unfit for work in general	Vital capacity improved more markedly (from 2292 to 3164 mL) in acute COVID-19 patients than in long Covid patients (from 2414 to 2609 mL) Acute COVID-19 patients reached higher inspiratory muscle strength compared to long COVID patients at discharge (maximal resistance 30 cmH20 and 21 cmH20, respectively) Lower minimal oxygen saturation in acute COVID-19 patients compared to long COVID patients (93.7% and 93.4%, respectively) Comparable improvement in bicycle exercise performance in both groups (change from 39.0 to 73.8 W in acute COVID-19 and from 46.9 to 74.0 W in long COVID) 33% of long COVID patients were fit for work at discharge versus 5% of acute COVID-19 patients 52% of long COVID patients were considered able to return to work in 6 months versus 95% of acute COVID-19 patients at discharge 29% of long COVID patients were considered unfit for work in general versus 5% of acute COVID-19 patients at discharge
Armstrong [85]  2024  UK	Personalised health behaviour support	Virtual, home-based	30-min semi-structured interview to explore individual experiences and motivations for change Seven weekly 30-min one-on-one support sessions Daily symptom and physical activity diaries, which informed personalised feedback (used behaviour change techniques such as goal setting, action planning and self-monitoring) Note: a symptom-contingent pacing approach was emphasised, encouraging participants to adjust activity levels based on symptom severity	8-week programme with duration varying by component (see column to the left)	NR	8 weeks	Daily steps Sedentary time MVPA time ISWT Handgrip strength 30 s STS CAT mMRC dyspnoea EQ-5D-5L	Note: Not powered for statistical testing; trends reported Greater increase in daily steps in the intervention group compared to usual care (1240 (95% CI 555–1926) steps) Small reduction in daily sedentary time in the intervention group compared to usual care (−10 (95% CI −83–64) mins) Small increase in daily MVPA time in the intervention group compared to usual care (9 (95% CI −29–47) min) Small increase in ISWT distance in the intervention group compared to usual care (21 (95% CI −39–78) m) Greater improvement in handgrip strength in the intervention group compared to usual care (1.2 (95% CI 0.5–3.2) kg) No between-group difference in 30 s STS (95% CI 0 (−1–1) repetitions) Greater reduction in CAT scores in the intervention group compared to usual care (−4 (95% CI −6–−1) units) No between-group difference in mMRC dyspnoea (0 (95% CI 0–1) units) No between-group difference in EQ-5D-5L scores (95% CI 0.1 (−0.1–0.3) units)
Bargaje [59]  2024  India	Multicomponent exercise rehabilitation	Outpatient, home-based virtual or hybrid	A personalised PR programme Patient and caregiver education Breathing and postural exercises Aerobic training (10–15 min using ergometers) Strength training targeting major upper and lower limb muscles Stretching exercises Exercise intensity and progression were based on patient tolerance, vital signs and perceived exertion using the modified Borg scale	8 weeks, 45–60 min per session, number of sessions not specified	NR	8 weeks	6MWT distance EQ-5D-DL	Mean±sd distance covered in metres pre: 209.22±159.12, post: 287.97±125.78, p<0.0001 Significant increase in the number of patients reporting to have “no problems” at baseline and the end of the programme in all the domains of QoL (mobility (χ2=68.26, p<0.0001), self-care (χ2=83.112, p<0.0001), usual activity (χ2=24.2, p<0.0001), pain and discomfort (χ2=10.27, p=0.001), anxiety and depression (χ2=12.08, p=0.0005), sleep (χ2=5.6, p=0.018))
Bileviciute-  Ljungar  [61] 2024  Sweden	Multicomponent exercise rehabilitation	Virtual, home-based	Multidisciplinary telerehabilitation programme delivered via Microsoft Teams. It focused on three main goals: regulating the autonomic nervous system (through breathing, mindfulness and ACT), restoring physical function (via tailored exercises using the ExorLive Go app) and creating individualised rehabilitation plans The programme included physical (e.g. yoga, Qigong), psychological, and educational components, and participants retained access to the exercise app post-programme	8-week programme with group sessions three times per week (2 h each), self-guided physical activity (3 h·week−1) and six one-on-one sessions with team members	51-week symptom duration for the telerehabilitation group, 58-week duration for controls	6 months	Breathing function questionnaire mMRC dyspnoea	In both groups, breathing functions were found to be significantly improved from a median score of 2 to 1 after 8 weeks Participants on the waiting list scored more improvements in breathing functions as compared to those in the intervention group (Mann–Whitney test, p=0.034) Both groups also saw improvements in breathing function at 6 months mMRC dyspnoea was improved for both groups at 8 weeks (p<0.05) and at 6 months (p<0.05) Clinical COPD Questionnaire score improved significantly in the intervention group (p<0.05) but not the control group (p>0.05) at 8 weeks and 6 months
Campos [86]  2024  Brazil	Multicomponent exercise rehabilitation	Outpatient	Supervised sessions which included aerobic exercise on a treadmill at moderate intensity, beginning at 75% of the speed achieved during the ISWT and progressing based on 60–75% of heart rate reserve and a perceived exertion of 4–6 on the modified Borg scale Resistance training was also included, performed at 80% of 10-repetition maximum for major muscle groups, with three sets of 10 repetitions and 1–2 min rest intervals Session concluded with stretching	Twice weekly sessions for 8 weeks, each lasting ∼80 min	NR	8 weeks	Modified Pulmonary Function Status Dyspnoea Questionnaire ISWT Lung function	Dyspnoea in activities of daily living decreased significantly in the intervention group only (Δ=−5.0, p<0.05) Both groups had a significant reduction in overall score (p<0.05) Both groups had a significant increase in ISWT (intervention: Δ=100.8, p<0.05, control Δ=65.9, p<0.05) PEF (L·s–1) increased significantly (p=0.028) for the face to-face rehabilitation group only (Δ=0.61, p<0.05)
Campos [87]  2024  Brazil	Multicomponent exercise rehabilitation	Outpatient	Patients who had undergone PR completed individualised resistance, strength and inspiratory muscle training, under the supervision of physicians and physical therapists	8 weeks, 2–3 times a week for 3–4 h each	Average±sd time since diagnosis of COVID-19 was 7.1±1.8 months (intervention) and 8.8±1.9 (control) months	8 weeks	Lung function (spirometry, impulse oscillometry and lung ultrasound) 6MWT distance	FEV1/FVC were normal in both groups but significantly higher in the intervention group (p<0.05) Fres and AX were significantly lower in the intervention group (p<0.05) No differences between groups in lung ultrasound measures were identified 6MWT distance was significantly higher in the intervention group (437±92 m versus 361±79 m)
Colas [62]  2023  France	Multicomponent exercise rehabilitation	Outpatient	Aerobic and resistance exercise Therapeutic education session on symptom management (e.g. fatigue, sleep, nutrition and exercise)	Overall: three sessions per week (2 h per session) for 4 weeks Aerobic and resistance exercise: 2×90 min weekly Therapeutic education: once weekly	NR	NR	1) V′O2peak 2) V′O2 at VT1 3) Maximal watts achieved during a cardiopulmonary exercise test	V′O2 significantly improved (pre: 23.0±5.92, post: 27.0±7.22, p<0.001). VT1 significantly improved (pre: 15.6±5.03, post: 17.3±4.74, p<0.001) Maximal watts significantly improved (pre: 133±50.9, post: 154±55.7, p<0.001)
Daynes [63]  2025  UK	Multicomponent exercise rehabilitation	Outpatient and virtual	Face-to-face rehabilitation programme led by a multidisciplinary team (PTs, OTs, nurses, exercise physiologists, support staff) Included symptom-titrated aerobic and resistance training, interactive educational discussions and tailored self-management strategies (e.g. pacing, planning, symptom management, vocational advice when appropriate Participants also received 1:1 support and individualised home exercise programmes (three additional sessions per week) recorded in a symptom/activity diary Peer support was an additional benefit of the group-based format Remote rehabilitation used a virtual platform Participants completed self-directed symptom-titrated aerobic and resistance training and accessed tailored education and symptom management tools online No equipment was provided, participants received phone support and additional guidance through a secure messaging system Healthcare professionals monitored symptom progression and exercise responses via the platform, adjusting support as needed	Both programmes spanned 8 weeks Face-to-face programme: two supervised sessions per week (90–120 mins each) plus three home sessions Remote programme: self-directed model with professional support every 2 weeks	Days±sd since hospitalisation: face-to-face rehab 578±176, virtual rehab 542±219, control 519±232	8 weeks	ISWT EQ-5D-5L PHQ-9 Dyspnea-12	1) Statistically significant improvement in ISWT between the face-to-face rehabilitation and usual-care group with a mean difference of 52 (95% CI 19–85) m in favour of the intervention (p=0.002); statistically significant difference between the remote rehabilitation and usual-care group with a mean difference of 34 (95% CI 1–66) m in favour of the intervention (p=0.047) 2–4) There were no differences between groups for HRQoL or self-reported symptoms measured by the EQ-5D-5L, Dyspnoea-12 or PHQ-9
del Corral [64]  2025  Spain	Multicomponent exercise rehabilitation	Outpatient and home-based	Home-based RMT and supervised aerobic exercise RMT was performed using the Orygen-Dual valve Training intensity started at 50% of MIP/MEP and increased by 10% every 2 weeks Evening sessions were supervised remotely by a PT Aerobic exercise included warm-up, interval cycling at 60–75% of max heart rate, cool-down and stretching Exercise intensity was guided by a perceived exertion rating of 4–6/10	8-week intervention with RMT performed three times per week (split into two 20 min segments) Aerobic endurance training conducted twice weekly, 50 min per session	Days±sd post diagnosis: intervention 673.3±277.0, control 673.8±284.1	8 weeks	EQ-5D-5L V′O2peak (CPET) Respiratory muscle strength Lung function	Both intervention and control significantly improved EQ-5D-5L No between-group difference was identified (Δ=0.06 (95% CI −0.01–0.13), d=0.3) Both intervention and control significantly improved V′O2peak No between-group difference was identified (Δ=0.4 (95% CI −0.5–1.3), d=0.2) Intervention group exhibited statistically significant improvements in respiratory muscle strength and IME compared to control (MIP: p<0.001, d=1.2; MEP: p<0.001, d=1.3; IME: p=0.004, d=0.7) DLCO and PEF exhibited significant within-group improvements in the intervention group only PEF displayed a significant increase compared with controls (p=0.044, d=0.4) No other significant between-group differences were identified (p>0.05)
Frisk [65] 2023  Norway	Micro-choice rehabilitation programme	Outpatient	Phase 1: physical examinations, education about intervention Phase 2: concentrated micro-choice-based rehabilitation (shift in focus from targeting and monitoring symptoms to micro-choices to facilitate increased levels of physical activity and functioning) Phase 3: integrating the changes into everyday living (answering two questions digitally regarding strategies for handling symptoms)	Phase 2: delivered during 3 consecutive days Phase 3: daily questions for the first 3 weeks post-rehabilitation	Mean±sd time to rehabilitation after confirmed COVID-19 (months) 10.2±4.8	3 months post-intervention	Dyspnoea: mMRC and Dyspnea-12 Lung function: FEV1, FVC, DLCO, SpO2, RER Exercise capacity: CPET, V′O2peak, 30 s STS test, SCT	Dyspnoea improved – 12% reduction in participants scoring ≥1 on mMRC (p=0.03), reduction in Dyspnea-12 score by 3.3 (p<0.001) Lung function was within normal values at baseline and 3-month follow-up Exercise capacity improved by 3.8 on the SCT (p<0.001) and 3.3 on the 30 s STS test (p<0.001), V′O2peak increased by 0.9 mL·kg−1·min−1 (p=0.002)
Frisk [65] 2023  Norway	Micro-choice rehabilitation programme	Outpatient	As above	As above	As above	Provides 12-month follow-up data, extending upon the original study	Dyspnea-12 Lung function: FEV1, FVC, DLCO, SpO2, RER Exercise capacity: CPET, V′O2peak, 30 s STS test, SCT	A further reduction in dyspnoea was identified at 12 months: reduction in Dyspnea-12 score by 4.8 (95% CI 3.2–6.4, p<0.001) compared to baseline Lung function remained within normal values at 12-month follow-up Further improvements in exercise capacity were identified: SCT improved by 6.0 (95% CI 4.4–7.7, p<0.001) and 30 s STS test improved by 4.6 (95% CI 3.5–5.7, p<0.001); relative V′O2peak was significantly increased at 3 months and this increase was maintained at 12 months at (1.1 mL·kg−1·min−1 (95% CI 0.1–2.0, p=0.023) (all comparisons relative to baseline)
Mammi [67]  2023  Italy	Multicomponent exercise rehabilitation	Outpatient (rehabilitation clinic)	Breathing pattern awareness, guided by diaphragmatic breathing training and exercises for thoracoabdominal coordination Observation of patient's performance and tolerance guided the progression in difficulty Endurance training (walking/ treadmill based) Home exercise sets Educational information accessible on website page for self-treatment	10×45-min individual sessions with physiotherapists (twice weekly for 5 weeks)	NR	NR	Symptom intensity of breathlessness: NRS (with 0=no symptoms and 10=worst possible intensity of symptoms) HRQoL: EQindex and EQVAS	Symptom intensity±sd of breathlessness reduced from 5.614±2.41 to 2.705±2.51 HRQoL (EQindex) increased from 0.652±0.21 to 0.756±0.20; HRQoL (EQVAS) increased from 60.223±17.42 to 79.442±16.48
Moine [68]  2024  France	Multicomponent exercise rehabilitation	Outpatient	Multidisciplinary, and individualised intervention delivered at two harmonised clinical sites including: Endurance training sessions (walking and cycling near the ventilation threshold) Resistance training sessions (four exercises per session targeting both upper and lower limbs, performed at a perceived exertion level of 7/10) Patient education Dyspnoea management Psychosocial counselling	Average of 26 endurance training sessions delivered over a 4-week period (6–7 days per week); 12 resistance training sessions (three per week) Other component featured eight sessions over the 4-week period (two per week)	Average 13 months post-acute infection (range 7–16)	4 weeks	Dyspnea-12 mMRC dyspnoea Hyperventilation (Nijmegen questionnaire) SF-12 physical and mental Muscular strength (% pred) 6MWT distance	Median Dyspnea-12 score improved by 7 units (−7 (95% CI −9–−2), p<0.001) Median mMRC dyspnoea impact score improved by 1 unit (−1 (95% CI −1–0), p<0.01) Mean Nijmegen hyperventilation score improved by 8 units (−8 (95% CI −11–−5), p<0.001) Mean SF-12 mental component score improved by 10 points (10 (95% CI 6–13), p<0.001), mean SF-12 physical component score improved by 9 points (9 (95% CI 6–12), p<0.001) No change in quadriceps muscle strength following intervention; mean QMVC change of 6 Nm (6 (95% CI −1–12), p=0.10, nonsignificant); mean QMVC % pred change of 5% (5 (95% CI −1–11), p=0.07, nonsignificant) Mean 6MWT distance improved by 61 m following intervention (61 (95% CI 28–103), p<0.001)
Nasrullah [69]  2024  USA	Multicomponent exercise rehabilitation	Outpatient, virtual or hybrid	The programme consisted of an orientation session, initial assessments by physical and respiratory therapists, and a risk assessment to develop a personalised care plan It emphasises a multidisciplinary approach, incorporating goals to improve activities of daily living, functional abilities, strength, endurance and self-management skills	Two to three times weekly for 3 months	Mean time=4 months following initial COVID-19 infection	3 months	mMRC dyspnoea SOBQ CAT 6MWT distance TUG STSs PHQ-9	1) mMRC dyspnoea significantly improved following the intervention (mean change −1.20 (95% CI −1.46–−2.47, p<0.0001) 2) SOBQ score significantly improved following the intervention (mean change −28.55 (95% CI −34.47–−22.62), p<0.0001) 3) CAT score significantly improved following the intervention (mean change −9.66 (95% CI −11.29–−8.03), p<0.0001) 4) 6MWT distance significantly improved following the intervention (mean change 405 (95% CI 333–477), p<0.0001) 5) TUG significantly improved following the intervention (mean change −3.50 (95% CI (−4.53–−2.47), p<0.0001) 6) STSs significantly improved following the intervention (mean change 4.31 (95% CI 3.71–4.90), p<0.0001) 7) PHQ-9 did not change significantly following the intervention (mean change 0.63 (95% CI −1.13–2.40), p=0.7402)
Nerli [70] 2024  Norway	Cognitive and behavioural rehabilitation programme	Outpatient	Physician-conducted clinical assessment to rule out other diagnoses, validate patient symptoms and provide psychoeducation based on the CATS, e.g. normalising the body's stress responses, explaining how infections can lead to temporary, reversible symptom patterns. Cognitive–behavioural therapy sessions led by physiotherapists, using techniques (e.g. nondirective communication, Socratic dialogue, guided discovery) Patients encouraged to safely engage in new physical and mental activities to build confidence, reduce symptom monitoring and foster positive expectations about recovery	Two to eight encounters; ∼2–6 weeks between encounters	NR	12 months	ΔmMRC dyspnoea ΔSF-36 physical function	Mean difference for mMRC of −0.20 (95% CI −0.5–−0.04) at end of intervention, p=0.10 Mean difference of −0.40 (95% CI −0.60–−0.20) favouring intervention group at 12 month follow-up, p<0.01 Mean difference in favour of intervention group for SF-36 physical function of 9.0 (95% CI 4.3–14.2) at end of intervention and 9.0 (95% CI 4.0–13.9) at 12 month follow-up, p<0.01 for both time-points
Onik [71] 2024  Poland	Spa rehabilitation	Outpatient (health resort/spa)	Balneotherapy Respiratory exercises (i.e. relaxation, extended exhalation, diaphragmatic breathing, lower ribcage activation) General exercises (endurance, balance, coordination, strength) Physical medicine modalities (i.e. massage, lymphatic drainage, whirlpool baths, mud therapy, low-level laser therapy, infrared light therapy, cryotherapy, electromagnetic field therapy, electrotherapy, ultrasound therapy) Health education (lifestyle, inhaler use, diet, physical activity)	Individually tailored; most participants attended daily or near-daily sessions as part of a structured, multidisciplinary programme over the course of their stay	NR	The mean±sd duration of the health resort treatment was 24.59±6.38 days Polish legal rules enabled 2–6 weeks of treatment	ΔmMRC VAS for respiratory symptoms	Significant improvement in mMRC in female (0.99±0.77, p<0.05) and male (0.66±0.65, p<0.05) participants following the intervention Dyspnoea at rest significantly improved following the intervention (mean±sd) – pre: 1.11±1.51; post: 0.31±0.62, p<0.0001 Exercise-induced dyspnoea significantly improved following the intervention – pre: 3.20±2.73; post: 1.19± 1.27, p<0.0001 Cough intensity scores significantly improved following the intervention (mean±sd) – pre: 0.96±1.41; post: 0.18±0.46, p<0.0001 Tightness of the chest scores significantly improved following the intervention (mean±sd) – pre: 0.75±1.47; post: 0.20±0.60, p<0.0001 Chest pain scores significantly improved following the intervention (mean± sd) – pre: 0.34±0.96; post: 0.08±0.30, p=0.002
Ostrowska [72]  2023  Switzerland	Multicomponent exercise rehabilitation	Outpatient	Physical training (aerobic, resistance, and breathing exercises) Education (30 min, 3 times weekly) and group psychotherapy (30 min, once weekly)	Three times weekly for 6 weeks: 90 min 30 min once weekly for 6 weeks 30 min	NR	6 weeks	CPET: V′O2peak Dyspnoea (mMRC and modified Borg scale (1–10)) Exercise capacity (6MWT and 30 s CST)	No statistically significant change in V′O2peak Dyspnoea improved by 1 point on both the modified Borg scale and the mMRC (p<0.0001) Exercise capacity significantly improved by 62.5 m in the 6MWT (p<0.0001) and by 3 in the 30 s CST (p<0.0001)
Ovejero [73]  2025  Spain	Balneotherapy programme	Outpatient (health resort/spa)	Thermal water inhalation (mineralised sodium chloride water at 35°C) Circular fine jet shower Aquatic exercise programme (joint mobility, strength, coordination and relaxation exercises) Whirlpool relaxation	Three sessions per week on alternate days Total: 12 sessions over 4 weeks Each session lasted 2 h	NR	2 months	mMRC dyspnoea SF-36 health survey	mMRC dyspnoea – no difference between groups following intervention or at 1 month follow-up Post-intervention β-coefficient 0.267 (95% CI −0.54–0.588), p=0.153 1 month follow-up β-coefficient 0.35 (95% CI 0.05–0.65), p=0.069 SF-46 general health – no difference between groups following intervention or at 1 month follow-up Post-intervention β-coefficient −2.519 (95% CI −8.065–3.028), p=0.448 1 month follow-up β-coefficient −2.993 (95% CI −9.599–3.614), p=0.555
Parker [74]  2023  UK	Multicomponent exercise rehabilitation	Outpatient (telephone)	WHO Borg CR-10 five-phase pacing protocol for PESE in post-COVID-19 syndrome: recommends activities across progressively intense phases aligned with perceived exertion	Five phases with 7 days per phase for 6 weeks duration	NR	6 weeks	Number of PESE episodes, Average number of symptoms per PESE QoL (EQ-5D VAS)	Number of PESE episodes per week reduced by 2.3 (16% reduction, 95% CI 9–24% CI) Average number of symptoms per PESE reduced by 1.0 (13% reduction, 95% CI 5–21% ) EQ-5D VAS increased by 9.2 (95% CI 3.2–15.2 points)
Philip [75]  2024  UK	Singing-based rehabilitation programme	Home-based (virtual)	Initial one-on-one consultation Weekly online group workshops led by vocal specialists Access to online learning resources (guided videos, playlists, audio tracks, written materials) for asynchronous use Ongoing email support	6 weeks, 1 h per week	NR	6 weeks	Dyspnea-12 CAT VAS – dyspnoea	Dyspnea-12 scores (mean±sd) significantly improved following the intervention – pre: 15.4±7.4; post: 11.1±7.2, p<0.001 CAT scores (mean±sd) significantly improved following the intervention – pre: 19.5±6.4; post: 16.9±6.9, p<0.001 VAS dyspnoea (mean±sd) (rest) significantly worsened following the intervention – Pre: 22.6±20.8; post: 23.9±22.3, p<0.001 VAS dyspnoea (walking) significantly improved following the intervention – pre: 36.8±23.0; post: 31.8±23.2, p<0.001 VAS dyspnoea (stairs) significantly improved following the intervention – pre: 58.8±24.7; post: 48.0±25.8, p<0.001 VAS dyspnoea (running) significantly improved following the intervention – pre: 83.7±22.0; post: 75.8±26.0, p<0.001
Resta [77]  2024  Italy	Spa rehabilitation	Outpatient (health resort/spa)	Inhalation therapy with mineral water Hydrokinesitherapy (water-based physiotherapy, featuring pursed lip breathing, diaphragmatic breathing and secretion mobilisation) High-flow noninvasive ventilation with mineral water, when indicated	Daily treatment for 12 days; 2–3 h per day	>3 months following acute infection	12 days	Subjective improvement in dyspnoea Borg dyspnoea	Upon completion of the intervention, 165 out of 187 patients (88.2%) reported an improvement in their sensation of dyspnoea Assessed in a subgroup of 118 patients: intervention associated with a median improvement in Borg dyspnoea by 2.0 (mean±sd pre-treatment 3.6±1.3, mean±sd post-treatment Borg was 1.4±0.7)
Resta [76]  2024  Italy	Spa rehabilitation	Outpatient (health resort/spa)	As above	As above	>3 months following acute infection	12 days	mMRC dyspnoea Borg dyspnoea	No significant improvement in dyspnoea following intervention (Δ−1.28±0.90) (note: unclear if this change in dyspnoea was measured using the mMRC scale or Borg scale)
Rzepka-  Cholasinska  [82] 2024  Switzerland	Multicomponent exercise rehabilitation	Outpatient (university hospital)	Warm up (active breathing) Exercise (resistance exercises with balls, bands, sticks; aerobic step exercise, respiratory muscle exercises with band) Cool down (stretching and breathing exercises)	6 weeks of 30 min sessions (three times weekly for aerobic; once weekly for resistance, progressing to twice weekly in week 3 and three times weekly in week 5)	NR	6 weeks	mMRC dyspnoea 30 s STS 6 WT distance Desaturation during 6MWT	mMRC dyspnoea (mean±sd) significantly reduced following intervention – pre: 2.2±0.5, post: 1.0±0.9, p<0.005 30 s STS (mean±sd) significantly improved following intervention – pre: 13.4±4.3, post: 17.1±4.7, p<0.005 6MWT distance (mean±sd) significantly improved following intervention – pre: 327.8±82.9, post: 384.0±74.6, p<0.005 Number of participants desaturating during the 6MWT did not significantly change following intervention – pre: 4 (4.4%), post: 0 (0%), p=0.13
Sick [78] 2025  Austria	Multicomponent exercise rehabilitation	Outpatient (local gym)	Endurance training group: 30 min continuous, sub ventilatory threshold cycle, elliptical or treadmill exercise for 2 weeks, then interval training, 1 min at ventilatory threshold with active recovery (ratio 1:2), max 60 min, progressive increases in volume as tolerated Concurrent: machine-based resistance exercises (leg press, leg curl, chest press, horizontal row at RPE 12 for 2 weeks, then RPE 14–18 for remaining weeks) followed by endurance training as described above but with lower volume to match energy expenditure	12 weeks of exercise training, three times weekly Endurance training: 30–60 min Concurrent training: 30–60 min (energy expenditure matched)	Mean±sd time post-acute infection (days) Endurance: 223±155 Concurrent: 238±137	12 weeks	Peak O2 consumption (mL·min−1·kg−1) Number of symptoms mMRC dyspnoea (% with score ≥1) HRQoL (SF-36 total score)	Significant within-group improvements for endurance and concurrent groups (p<0.05); no change in control group (p>0.05); both groups improved significantly relative to control (group×time interaction, p<0.05) Significant reduction in number of symptoms for both groups and controls (p<0.05); no group×time interaction Significant reduction in mMRC dyspnoea in concurrent group only (p<0.05); ANOVA not completed for between-group differences Significant increase in SF-36 total score for both groups and controls (p<0.05); no group×time interaction present
Smith [79]  2023  UK	Multicomponent exercise rehabilitation	Mixed virtual and community based	6 weeks virtual, 6 weeks in-person Group exercise sessions on-demand, pre-recorded guided exercises on a dedicated online platform Weekly telephone call with a rehabilitation specialist lasting up to 45 min	3×45-min exercise sessions for 12 weeks	Mean±sd period of 9.8±5.0 months post-diagnosis	12 weeks	Dyspnoea (Dyspnea-12) Functional capacity (DASI)	Significant improvements in Dyspnea-12 (p<0.001); MCID was exceeded for this measure Significant improvements in DASI values (p<0.001); MCID was exceeded for this measure
Szarvas [80]  2024  Hungary	Multicomponent exercise rehabilitation	Outpatient	Programme consisted of daily group and individualised exercise sessions 1) Group sessions included three 30-min breathing exercise classes per day, led by a physiotherapist, incorporating controlled breathing, chest mobility and muscle strengthening using body weight and dumbbells 2) Individualised sessions included two 30-min low-intensity endurance workouts (e.g. cycling, treadmill, rowing), with programme design based on factors such as COVID-19 severity, comorbidities, and physical condition; in severe cases, HIIT was implemented, involving alternating 60 s intervals of intense and low-intensity activity, with gradual progression and supervision	Daily for 2 weeks. 30 min breathing exercises and resistance exercises, 30 min endurance exercise training	NR	3 months	No pre- or post-intervention assessment; 2- and 3-month follow-up only; controls not assessed Peak work rate during CPET (Watts) V′O2peak (mL·min−1) V′O2peak (mL·kg·min−1) V′Epeak V′E/V′CO2peak	2-month follow-up peak work rate (median (IQR)): 106 W (82–139); no change at 3-month follow-up 2-month follow-up V′O2peak (median (IQR)): 1276 mL·min–1 (1070–1614); significant (unexplained) improvement at 3-month follow-up (1429.5 mL·min–1 (1191–1871)) 2-month follow-up V′O2peak (median (IQR): 14.6 mL·min–1 (11.95–18.2); significant (unexplained) improvement at 3-month follow-up (16 mL·min–1 (12.4–21)) 2-month follow-up V′Epeak (median (IQR): 53.2 (43.3–67.5); no change at 3-month follow-up. 2-month follow-up peak V′E/V′CO2 (median (IQR): 38.4 (34.6–49.2); no change at 3-month follow-up
Yasaci [81]  2025  Turkey	Telerehabilitation-based exercise programme	Home-based, virtual	Telerehabilitation programme delivered via video conferencing (WhatsApp or Zoom) by an experienced PT Each 45-min session included supervised breathing exercises (diaphragmatic and pursed-lip), relaxation, range-of-motion movements, lower-limb strengthening (e.g. mini-squats) and aerobic activities such as domestic walking Exercise intensity was tailored using ACSM guidelines and monitored via the Borg RPE scale (target: 11–13) A 1-h online orientation preceded the intervention The control group received a brochure outlining the same unsupervised home exercise programme and instructions Both groups were advised to stop exercise if symptoms worsened Exercises were selected based on clinical relevance, evidence of efficacy and feasibility in home settings	6-week telerehabilitation programme delivered twice weekly, 45 min per session	NR	6 weeks	1) mMRC dyspnoea	1) Significant reduction in mMRC dyspnoea following the intervention in intervention group (Δ −1.16 (95% CI −0.93–1.38), p=0.001) and the control group (Δ −0.36 (95% CI −0.17–0.55), p=0.001) A significant between-group difference was observed in favour of the intervention group (Δ 0.8 (95% CI 0.5–1.1), p=0.001)
Secondary research (n=20)
Aiyegbusi [83]  2021  UK	Multicomponent exercise rehabilitation	NR	Recommendations for physical therapy management	NR	NR	NR	Breathing, exercise capacity, muscle strength, quality of life and functional outcomes	It was recommended that appropriate rehabilitation to prevent this syndrome should start in ICUs as soon as sedation and clinical stability permit Physical rehabilitation should be started in ICU as soon as the patient is stable Rehabilitation may help improve patients’ breathing, exercise capacity, muscle strength, QoL and functional outcome Early mobilisation would help to improve functional, cognitive and respiratory conditions in these patients and may shorten hospital stay Nonhospitalised patients with long COVID may also benefit from physical rehabilitation
Arienti [84]  2023  Italy	Exercise-based rehabilitation programmes	Hospital (n=1 study), home-based (n=12 studies), general outpatient (n=6 studies), specialised post-acute (n=1 study), NR (n=1 study)	Aerobic/endurance exercise (n=12 studies) Resistance training (n=10 studies) Respiratory muscle training (n=4 studies) Breathing exercises (n=6 studies) Multicomponent rehabilitation (n=5 studies) Telerehabilitation (n=4 studies) HIIT (n=1 study) Yoga, pilates, relaxation (n=2 studies) Virtual reality exercise (n=1 study)	NR	NR	NR	Dyspnoea, exercise capacity, QoL	Overall, while rehabilitation interventions, particularly multicomponent and supervised approaches, appear to improve dyspnoea, exercise capacity and QoL in adults with PCC, although the certainty of evidence remains low to very low due to methodological limitations and small sample sizes
Cha [6] 2024  South  Korea	Various rehabilitation types	Inpatient: rehabilitation unit (n=1) Outpatient: health spa (n=1), multicentre (n=1), virtual (n=4) NR (n=3)	Number of studies with each component: traditional occupational rehabilitation (n=1), multidisciplinary involvement (n=3), personalised planning (n=2), physical training (n=4), respiratory training (n=3), virtual platforms (n=3), corrective actions (n=1), education (n=3), standardised rehabilitation guidelines	NR	NR	NR	Any clinically relevant outcome	Evidence on PCC management was weak in the literature Suggested management for PCC included a focus on respiratory, fatigue and psychological distress Several studies suggested utilising remote access and multidisciplinary access based on information technology
Dillen [44]  2023  Belgium	Multicomponent exercise rehabilitation	Outpatient	Exercise training, breathing exercises, nutritional supplementation, olfactory training, multidisciplinary treatment, narrative exposure therapy, aromatherapy, hydrogen inhalation, massage	NR	NR	NR	Dyspnoea, chest pain, physical capacity (i.e. physical fitness and muscle performance), pulmonary function, QoL, return to normal daily life activities, functional capacity (i.e. ability to perform “activities of daily living”)	Physical training programmes and breathing exercises may reduce dyspnoea and chest pain and may improve physical capacity and quality of life, but the supporting evidence is very weak The evidence underpinning the effect of nutritional supplements on dyspnoea, QoL and functional capacity is considered to be very poor Multidisciplinary treatment may have beneficial effects on dyspnoea, physical capacity, pulmonary function, QoL, return to normal daily life activities and functional capacity, but the evidence is very uncertain
Martinez-Pozas  [45] 2024  Spain	Multicomponent exercise rehabilitation	Face-to-face (n=5); telerehabilitation (n=5)	Components varied depending on modality Telemedicine: majority involved supervised sessions guided by physical therapist guide individuals in performing their exercises	Average: 24 sessions, 60–80 min per session, 3–10 weeks duration	NR	NR	Physical function, QoL	In multivariate network analysis, moderate quality of evidence concluded that face-to-face was superior to usual care and telerehabilitation, improving all physical function and quality of life measures In addition, low-quality evidence suggested that telerehabilitation was superior to usual care for all outcomes
Martinez-Pozas  [46] 2024  Spain	Multicomponent exercise rehabilitation	Face-to-face (n=2), telerehabilitation (n=5)	Programmes incorporated a range of components, most commonly breathing exercises or respiratory muscle training (n=7), followed by aerobic training (n=3), strength training (n=3) and educational or psychological support (n=3) Several interventions were multicomponent, while others focused on single elements such as breathing or strength training	Frequency NR, duration ranged 5–10 weeks	NR	NR	Dyspnoea (e.g. mMRC dyspnoea, Transition Dyspnea Index, Dyspnea-12)	Pooling seven studies (n=365) showed a large, significant effect of PR on dyspnoea, however high between study heterogeneity was identified After sensitivity analysis and removal of outliers, the effect remained significant and moderate (SMD 0.55, 95% CI 0.41–0.68) with no heterogeneity (I2=0%) Meta-regression indicated no differences between face-to-face and telerehabilitation delivery formats, and no publication bias was detected
McDowell [47]  2025  Ireland	Multicomponent exercise rehabilitation	In-person (n=3), telerehabilitation/home-based (n=4), hospital-based (n=2)	All interventions were low-to-moderate intensity, individualised and had various components (n=6 aerobic exercises, n=4 strength training, n=3 breathing exercises)	Frequency NR, duration ranged 2–10 weeks	NR	Ranged: 2–28 weeks (mean=8.5 weeks)	Dyspnoea, lung function	Across studies, results for dyspnoea were mixed Three studies reported no between-group differences, while others found significant improvements favouring intervention groups, with effect sizes ranging from moderate to large Evidence for changes in pulmonary function were limited and inconclusive
Neto [48] 2025  Brazil	Multicomponent exercise rehabilitation	Home-based/ telerehabilitation	NR	NR	NR	NR	Dyspnoea (mMRC dyspnoea)	Two studies reported substantial improvements in dyspnoea among PCC patients, with one study showing mMRC score improvement in 90.4% of patients and another reporting a 47.8% reduction in mMRC scores following telerehabilitation
Ora [38] 2023  Italy	Multicomponent exercise rehabilitation	Outpatient	NR	NR	NR	NR	HRQoL, 6MWT distance, feasibility, cardiorespiratory fitness, lung function, physical functioning, mobility, fatigue, perceived exertion	At the time of publication, there were no published studies on rehabilitation, but 17 studies were identified as registered clinical trials, including the COVID-Rehab study, which is investigating the effectiveness of an 8-week cardiopulmonary rehab programme on cardiorespiratory fitness, functional capacity, QoL, inflammation and oxidative stress
Pouliopoulou  [49] 2023  Canada	Respiratory training and exercise rehabilitation	Outpatient	Exercise rehabilitation (n=4), respiratory muscle training (n=4), combined exercise and respiratory muscle training (n=6).	NR	NR	NR	Primary: 6MWT distance Secondary: fatigue, functional leg strength, and endurance (30 s STS test), dyspnoea; respiratory function (FEV1, FVC) and QoL	Interventions were associated with a greater improvement in functional exercise capacity, dyspnoea and QoL compared with usual care No difference between rehabilitation interventions and usual care in either FEV1 or FVC
Romanet [50]  2025  France	Various rehabilitation types	Outpatient, telerehabilitation, in-patient	PR (n=6 studies), multicomponent home-based rehabilitation (n=6 studies), respiratory muscle training (n=6 studies), low-level laser therapy (n=1 study)	On average: 22 sessions, ranging 30–120 min, 6-week duration	NR	NR	Dyspnoea	Random-effects meta-analysis showed a significant positive treatment effect on dyspnoea post-intervention in favour of multi-component exercise rehabilitation interventions (SMD −0.63, 95 CI −1.03–−0.24, p<0.001, I2=88%) despite significant heterogeneity Sub-group analysis confirmed the impact of interventions on dyspnoea for high-intensity exercise interventions only For low-level laser therapy, no meta-analysis could be completed (n=1) Results show a significant effect in favour of intervention compared to a sham technique
Schurr [51]  2025  Germany	Multicomponent psychotherapeutic interventions	NR	Rehabilitation programmes were multi-component, incorporating psychotherapeutic interventions (e.g. cognitive–behavioural therapy, EMDR, psychoeducation) with combinations of physical training/exercise therapy, physiotherapy, breathing and relaxation techniques, and nutritional or occupational support	Ranging from 4 to 12 sessions, duration NR	NR	NR	Dyspnoea, 6MWT distance	Three studies reported significant improvements in dyspnoea and one intervention reported a statistically significant improvement in 6MWT distance following interventions Components of interventions contributing to intervention effects (i.e. exercise versus psychotherapeutics) not clear
Soril [52] 2022  Canada	Pulmonary rehabilitation	Outpatient	Most PR were multidisciplinary, tailored to the participants’ conditions or rehabilitation goals and frequently consisted of exercise training, education and/or counselling	Frequency ranged from once per week to every day Programme duration ranged from 4 to 8 weeks	Reported in four studies Ranged from 10 to 125 days following onset of acute symptoms or COVID-19 diagnosis	Ranged from 3 weeks to 3 months	6MWT distance, HRQoL and pulmonary function (e.g. FVC, FEV1, FEV1/FVC, DLCO), dyspnoea	The experimental and the controlled before and after studies commonly reported statistically significant improvements in 6MWT distance, HRQoL and pulmonary function Select studies also reported improvements in dyspnoea One RCT with 36 participants demonstrated at 6-weeks post-PR significant improvements in 6MWT distance, HRQoL and pulmonary function compared to their baseline values and to control participants (n=36; no rehabilitation intervention)
Tan [53] 2025  China	Multicomponent exercise rehabilitation	NR	Exercise training programme (n=8 studies), respiratory muscle training (n=8 studies), telerehabilitation (n=7 studies)	NR	NR	NR	Lung function (e.g. FEV1, FVC, FEV1/FVC) Exercise performance (e.g. V′O2peak), dyspnoea (e.g. mMRC dyspnoea scale) Respiratory muscle function (e.g. MIP, MEP) Functional exercise capacity (e.g. 6MWT distance, 30 s STS test)	Exercise training and respiratory muscle training were both associated with significant improvements in functional exercise capacity, dyspnoea, V′O2peak and respiratory muscle function compared to controls Across all rehabilitation types, pulmonary function measures showed no significant differences Telerehabilitation showed significant improvements in functional exercise capacity only
Torres [54]  2023  South  Africa	Multicomponent exercise rehabilitation	NR	Various rehabilitation interventions (HIIT, flexibility, proprioception, endurance exercise, respiratory exercise, muscular endurance exercise, gymnastics, yoga, balance and Liuzijue	NR	NR	NR	6MWT distance, lung function (FEV1 % pred, FVC % pred)	Meta-analysis found significant improvements favouring exercise rehabilitation in 6MWT distance, FEV1 %pred and FVC % pred Between-study heterogeneity (measured as I2) was noted for both measures of lung function
Xavier [55]  2024  Brazil	Respiratory muscle training	NR	Various types of respiratory muscle training exercise cycles to gain and/or maximise resistance and/or strength of the respiratory muscles, including the use of equipment with progressive linear loads, most commonly use device was the POWER-breathe®	Frequency ranged from 3 to 14 sessions per week, duration ranged from 2 to 12 weeks	NR	Up to 12 weeks post-intervention	Respiratory muscle strength (e.g. maximum inspiratory pressure and maximum expiratory pressure) Lung function (e.g. FEV1, FVC, FEV1/FVC) Dyspnoea (e.g. Transition Dyspnea Index, mMRC dyspnoea scale) HRQoL (e.g. EQ-5D-3L, SF-12) Functional exercise capacity (e.g. 6MWT distance, 60-s STS test)	Findings indicate that respiratory muscle training results in significant improvements in respiratory muscle strength, functional exercise capacity and HRQoL However, conclusions remain uncertain due to the low evidence quality of identified studies
Yang [56] 2024  China	Telerehabilitation	Home-based	Breathing exercises, resistance and/or aerobic training, strengthening exercises	NR	NR	NR	Dyspnoea (e.g. Borg, Dyspnea-12) Functional exercise capacity (e.g. 30 s STS test, 6MWT distance) Physical function (e.g. SF-36) Lung function (e.g. FEV1, FVC, FEV1/FVC) HRQoL (e.g. EQ-5D)	Pooled analysis demonstrated that, compared with no rehabilitation or usual care, telerehabilitation can improve 30 s STS test (6 RCTs, n=310, MD 1.58 reps, 95% CI 0.50–2.66; p=0.004); 6MWT distance (6 RCTs, n=324, MD 76.9 m, 95% CI 49.5–104.3; p<0.00001) and physical function from the 36-SF (5 RCTs, n=380, MD 6.12 units, 95% CI 2.85–9.38; p=0.0002) Pooled results did not indicate significant improvements in dyspnoea, lung function parameters or HRQoL
Zeraatkar [42]  2024  Canada	Multicomponent exercise rehabilitation	NR	Inspiratory muscle training (n=1), multicomponent exercise rehabilitation (n=1)	Frequency NR, duration ranged from 8 to 10 weeks	NR	NR	Dyspnoea (e.g. Transitional Dyspnea Index, mMRC dyspnoea, Dyspnea-12) Physical function (e.g. SF-36)	In both relevant investigations, improvements in dyspnoea were identified; however, evidence was low quality (concerns related to serious imprecision) One investigation (multicomponent exercise rehabilitation) also identified a potential improvement in physical function (low quality of evidence)
Zheng [43]  2022  UK	Telerehabilitation (n=1)	Virtual, home-based	Breathing control and thoracic expansion, aerobic and lower limb muscle strength exercise	6 weeks	NR	28 weeks	Dyspnoea (mMRC scale)	One RCT found improvement in mMRC dyspnoea compared to controls was found immediately after the 6-week intervention period (p=0.001), but not at the 28-week follow-up
	Multicomponent exercise rehabilitation (n=7)	Outpatient	NR	NR	NR	NR	Persistent breathlessness	Seven observational studies suggested that rehabilitation exercises were associated with reduced persistent breathlessness in hospitalised or mild cases of COVID-19 in a qualitative synthesis
Zheng [57]  2024  China	Multicomponent exercise rehabilitation	Outpatient	Aerobic training (n=17), resistance training (n=17), respiratory muscle training (n=5), flexibility (n=3), balance training (n=3), interval training (n=3), yoga (n=1), pilates (n=1)	2–5 times per week for 2–12 weeks	Range: 30 days after discharge from hospital to 4 months post-diagnosis	Range: 2 weeks to 3 months	Lung function (FEV1, n=11; FEV1/FVC, n=9), 6MWT distance (n=9), STS test (n=9), QoL (n=15), dyspnoea (n=11)	Meta-analysis found significant improvements favouring exercise rehabilitation in 6MWT distance, FEV1/FVC, QoL and dyspnoea

6MWT: 6-min walk test; ACSM: American College of Sports Medicine; ACT: acceptance and commitment therapy; A_X : area of reactance; CAT: COPD Assessment Test; CATS: Cognitive Activation Theory of Stress; CPET: cardiopulmonary exercise testing; D_LCO: diffusing capacity of the lung for carbon monoxide; EMDR: eye movement desensitisation and reprocessing; EQ-5D-5L: EuroQol five-dimension five-level questionnaire; EQVAS: EQ visual analogue scale; FEV₁: forced expiratory volume in 1 s; F_res: resonant frequency; FVC: forced vital capacity; HIIT: high-intensity interval training; HRQoL: health-related quality of life; ICU: intensive care unit; IME: inspiratory muscle endurance; ISWT: incremental shuttle walk test; IQR: interquartile range; MCID: minimal clinically important difference; MEP: maximal inspiratory pressure; MIP: maximal inspiratory pressure; mMRC: modified Medical Research Council; MVPA: moderate-to-vigorous physical activity; NR: not reported; NRS: numeric rating scale; OT: occupational therapist; PEF: peak expiratory flow; PESE: post-exertional symptom exacerbation; PHQ-9: Patient Health Questionnaire-9; PR: pulmonary rehabilitation; PT: physiotherapist; QoL: quality of life; QMVC: quadriceps maximal voluntary contraction; RER: respiratory exchange ratio; RMR: respiratory muscle training; RPE: rating of perceived exertion; RCT: randomised controlled trial; SARS-CoV-2: severe acute respiratory syndrome coronavirus 2; SCT: stair climbing test; SOBQ: shortness of breath questionnaire; S_pO2: peripheral oxygen saturation; SF-12: 12-Item Short Form Health Survey; SF-36: 36-Item Short Form Health Survey; SMD: standardised mean difference; STS: sit to stand; TUG: timed up-and-go; WHO: World Health Organization; VAS: visual analogue score; V′_CO2: carbon dioxide production; V′_CO2peak: peak carbon dioxide production; V′_E: minute ventilation; V′_Epeak: peak minute ventilation; V′_O2: oxygen uptake; V′_O2peak: peak oxygen uptake; VT1: first ventilatory threshold. ^#: Text extracted verbatim from included studies in some instances.

What are the expert recommendations/guidelines for the workup and/or management of pulmonary complications in PCC?

12 studies (guideline document, n=10; position statement, n=2) addressed research question 1, with representation from seven different countries (Canada, n=2; Korea, n=1; Israel, n=1; Italy, n=1, South Korea, n=1; Sweden, n=1; Switzerland, n=1; UK, n=2; USA, n=2). Six studies utilised formal Delphi methodology to generate recommendations for PCC assessment, workup and management, with the remaining studies performing a literature review followed by expert narrative synthesis (n=6). Guidance document quality ranged from medium (n=2) to high (n=10) with stakeholder involvement, rigour of development and applicability domains commonly awarded lower scores (see appendix 7). Specific recommendations for the workup, assessment and management of pulmonary-related PCC are summarised in table 1 and synthesised below.

TABLE 1.

Summary of findings for question 1

First author [ref.],  year, country	Guidance type	Method or evidence source	Post-COVID-19 pulmonary complications	Phase of care	Guidance or recommendations#	AGREE tool summary score
Beauchamp [19] 2021 Canada	Position statement	Expert opinion	General	Assessment, management	1)  Patients to refer to PR: new or ongoing respiratory symptoms (dyspnoea and/or cough and/or exercise intolerance) and functional limitations (difficulty in performing daily activities) after resolution of acute COVID-19 AND new or ongoing requirement for supplemental oxygen after resolution of acute COVID-19 OR at least one of 1) persistent radiographic pulmonary abnormality, 2) PFT results demonstrating new/persistent reduction in lung volumes, airflow limitation, and/or reduction in diffusing capacity after resolution of acute COVID-19. 2)  Principles of PR can be adapted to meet the needs of patients who had COVID-19 and complain of cough, dyspnoea, and/or reduced exercise capacity and functional limitations. 3)  For patients beginning a modified PR programme, an assessment of the following domains (examples given in brackets) are recommended: Dyspnoea (e.g. UCSD SOBQ). Exercise capacity (e.g. 6MWT, 1-min sit-to-stand test). Physical function (e.g. SPPB). Health-related QoL (e.g. EQ-5D-5L). Mental health (e.g. impact of event scale). Fatigue (e.g. VAS scale). Return to work items as applicable. 4)  The following safety precautions should be considered in the context of rehabilitation for COVID-19: Possibility of myocarditis. In such cases, rehabilitation should be postponed until approved by a cardiologist. Deep vein thrombosis and pulmonary embolism for patients on anticoagulation, rehabilitation can be considered in the absence of bleeding or other complications. Exercise-induced desaturation and pre-existing health issues should be considered.	Medium
Cheng [25] 2025 USA	Guideline	Delphi study	Respiratory symptoms	Assessment, management	1)  Differential diagnoses to consider: nonpulmonary contributions (e.g. neurologic, cardiovascular, neuromuscular); other pulmonary and airway conditions (e.g. asthma). 2)  The assessment considerations for breathing discomfort and respiratory sequelae in patients with long COVID include the following: 3)  Assessment considerations (history): Pulmonary history of acute COVID-19 course: Desaturation? Supplemental O2? ICU? Mechanical ventilation? Document character and severity of breathing discomfort: At rest or with exertion? Relieved with rest? Chest tightness or air hunger? Assess for nonpulmonary contributors that could also limit activity (e.g. fatigue, PEM/PESE, autonomic, neurologic, cardiovascular, neuromuscular). 4)  Assessment considerations (physical exam): Vital signs: HR, respiratory rate, BP, SpO2 (on room air, at rest, and while walking at a pace/duration tolerated by the patient). Cardiopulmonary exam: assess for crackles, use of accessory breathing muscles, peripheral oedema, acrocyanosis. Assess for jugular venous distention, split P2, peripheral oedema, peripheral vascular perfusion and skin integrity. 5)  Assessment considerations (labs and procedures): If not already performed, consider: CXR. If breathing discomfort not improving ≥8 weeks after acute COVID-19 infection, or if new/worsened breathing discomfort later in long COVID course, consider: PFTs (spirometry and DLCO), ECG, echocardiogram; If pulmonary exam, CXR or PFT abnormalities, consider: noncontrast chest CT. 6)  Management considerations (nonpharmacologic): PR (use caution if coexisting PEM or PESE, and modify programme accordingly). If symptoms are improving or supervised rehabilitation is not available: i) breathing techniques and self-monitored paced physical activity; ii) self-directed educational resources or online programmes. If chronic productive cough, difficulty clearing airway secretions or bronchiectasis: airway clearance techniques and consider prescribing an airway clearing device. If requires home oxygen (to maximise mobility, QoL, and participation in rehabilitation): portable oxygen device, when possible. Consider a home peak flow meter to track progress and quantify the impact of respiratory infection or another setback. 7)  Management considerations (pharmacologic): If evidence of impaired pulmonary function, consider oral corticosteroids, inhaled bronchodilators, and inhaled corticosteroids (not for isolated breathing discomfort). 8)  Suggested referrals, as needed: To pulmonary medicine, if: i) breathing discomfort is progressing or limiting activity; ii) productive cough and/or difficulty clearing airway, persistent and unexplained pulmonary exam abnormalities, new or worsened O2 desaturation with activity, or new or worsened PFT or imaging abnormalities. To PR, if: i) recent diagnosis of a qualifying condition (i.e. history of long COVID with abnormal PFTs or exercise test, interstitial disease on imaging, functional limitations and associated dyspnoea from oxygen desaturation with exertion or at rest); ii) if PR is unavailable, consider PT with supplemental oxygen for conditioning and OT for ADL training. For patients with mild limitations such that insurance may not cover PR, PT or OT, consider a gentle return-to-activity programme in a gym with a certified athletic trainer. To otolaryngology, if abnormal upper airway breath sounds or voice changes (e.g. stridor, hoarseness, unexplained episodic breathing discomfort, especially if history of endotracheal intubation and/or tracheostomy).	High
Ewing [26] 2025 Sweden	Guideline	Delphi study	Respiratory symptoms	Assessment, management	1)  Clinical and diagnostic assessment: Careful assessment for cardiorespiratory post-COVID-19 complications across body systems (including dizziness on standing, palpitations, chest pain, SOB, cough, wheezing and tachycardia with modest exertion) is important. CPET can be used to evaluate cardiorespiratory function and functional capacity in patients for whom the testing is safe and appropriate. Testing for thrombophilias or indicators for increased clotting risk (i.e. factor V Leiden, anti-phospholipid antibodies, homocysteine, prothrombin, haemochromatosis testing, etc.) can be helpful. Recommended diagnostic tests available to clinicians for evaluation of patients with long COVID respiratory symptoms include: pulse oximetry, CXR, CT of the chest, PFTs, sleep study. 2)  Treatment recommendations: Anticoagulant and antiplatelet drugs can be used to treat a subpopulation of patients with long COVID, as long as appropriate diagnostic tools for thrombotic endothelitis are available and treatment is overseen by experienced clinician. Anticoagulant drugs can be useful in treating some long COVID patients. PR can be useful in treating some long COVID patients. PR (an established exercise training and education programme for people with structural lung disease) is not indicated for the majority of people with long COVID-related breathlessness. Respiratory pathology should be excluded prior to recommending graded exercise testing. 3)  Recommendations for assessment following treatment: Re-imaging for specific long COVID complications such as pulmonary embolism, myocarditis and heart failure can be useful to monitor the response to treatment in patients with long COVID. If determined safe and appropriate following detailed screening for PEM, repeat CPET can be important to monitor changes in V′O2max and anaerobic threshold and to measure the effectiveness of treatments (including rehabilitation programmes) in long COVID.	High
Falcone [13] 2025 Canada	Guideline	Delphi study		Assessment, management	1)  Assessment recommendations: Electrocardiogram: small benefit for identifying cardiac issues in PCC with respiratory or cardiac symptoms; negligible cost; widely available; recommendation likely favours use despite no direct studies. Holter monitor: small potential benefit for PCC with respiratory or cardiac symptoms; higher cost and possible access issues; probably favoured, but availability and wait times may limit use. Troponin (high-sensitivity Troponin I): small benefit for detecting cardiac injury in PCC with respiratory or cardiac symptoms; negligible cost; widely available; probably favoured despite no direct evidence. CXR: small benefit in PCC with respiratory or cardiac symptoms; high cost; generally accessible but mobility and staffing issues may limit use; recommendation varies. Chest CT: benefit varies by patient risk level; high cost and limited access; may be useful in select high-risk cases; equity concerns for remote or low-income patients. 2)  Management recommendations Exercise-based interventions (no PEM/PESE): moderate improvements in fatigue and function; large cost for supervised therapy; probably favoured for patients without PESE. Exercise-based interventions (with PEM/PESE): evidence limited; potential risk of worsening symptoms; no clear benefit–harm balance; recommendation does not favour either approach.	High
Funke-Chambour [20] 2021 Switzerland	Guideline	Delphi study	Respiratory symptoms, interstitial lung abnormalities, cough, fibrosis	Assessment, management	1)   All patients hospitalised for COVID-19 should have a pulmonary follow-up (strong recommendation for, consensus reached). 2)   All COVID-19 affected patients, including those with mild disease, should have a pulmonary follow-up, if symptomatic (strong recommendation for, consensus reached). 3)   All patients with persistent respiratory symptoms after COVID-19 should have a pulmonary follow-up including plethysmography, diffusing capacity measurement and blood gas analysis (strong recommendation for, consensus reached). 4)   For patients with severe COVID-19 (WHO definition), a pulmonary follow-up is recommended including plethysmography and diffusing capacity, blood gases analysis independently of symptom persistence (moderate recommendation for, consensus reached when strong and weak recommendations are merged). 5)   Exercise testing (e.g. 6MWD or equivalent) is recommended after COVID-19 (weak recommendation for, consensus reached). 6)   Chest CT scan is recommended to be routinely performed in patients with persisting respiratory symptoms after COVID-19 (moderate recommendation for, consensus reached when strong and weak recommendations are merged). 7)   CPET is recommended to be routinely performed after COVID-19 with persistent symptoms (moderate recommendation for, consensus reached when strong and weak recommendations are merged). 8)   Patients with persistent symptoms after COVID-19 are recommended to have access to specialised multidisciplinary post-COVID-19 clinics or networks (moderate recommendation for, consensus reached when strong and weak recommendations are merged). 9)   Patients after COVID-19 who present with new obstructive lung disease are recommended to be offered empiric topic inhaled or systemic steroid treatment (moderate recommendation for, consensus reached when strong and weak recommendations are merged). 10)  Patients after COVID-19 who present with persistent cough are recommended to have empiric inhaled topic steroids (moderate recommendation for, consensus reached when strong and weak recommendations are merged). 11)  Patients after COVID-19 who present with interstitial abnormalities after exclusion of an active infection are recommended to be evaluated to receive an empiric systemic steroid trial (moderate recommendation for, consensus reached when strong and weak recommendations are merged). 12)  It is unclear whether patients after COVID-19 who present with signs of pulmonary fibrosis should receive specific antifibrotic drugs (suggestion for no recommendation, no consensus). 13)  Patients after COVID-19 who present with persistent respiratory symptoms are recommended to undergo a rehabilitation programme (strong recommendation for, consensus reached).	High
Kim [21] 2022 South Korea	Guideline	Literature review and expert opinion	Dyspnoea, cough, sputum, exercise limitation	Assessment, management	1)  Long COVID should be diagnosed after excluding underlying diseases (e.g. cardiopulmonary sequelae), COVID-19 complications and other diseases that can also show the same symptoms and when the symptoms persist even after 12 weeks from COVID-19 diagnosis. 2)  Regardless of symptoms, PFTs including DLCO can be considered 3 months after the diagnosis of COVID-19 in patients with severe or critical COVID-19, persistent dyspnoea after acute COVID-19 or underlying lung disease. 3)  When respiratory symptoms persist for 3 months, CXR can be considered to exclude other diseases and detect early lung fibrosis. Chest CT can be considered if there are abnormalities on CXRs or if symptoms persist even without CXR abnormalities. 4)  If cough persists for >3 months, CXR and chest CT can be considered to check for pulmonary parenchymal fibrosis or bronchial inflammation. 5)  If long COVID persists in patients with severe COVID-19 treated in ICU or patients >65 years of age, appropriate and specific respiratory rehabilitation can be considered in consultation with a rehabilitation specialist. 6)  Evidence is insufficient to recommend or against specific medical treatments (corticosteroids, antihistamines, ipratropium bromide, aminophylline or codeine) in patients with respiratory symptoms (dyspnoea and cough) in long COVID.	High
Man [27] 2023 UK	Position statement	Literature review and expert opinion	Respiratory symptoms	Assessment, management	1)  Several factors need to be considered when providing PR to individuals with PCC: Unidentified (and therefore untreated) pulmonary thromboembolic disease and myocarditis have been reported in the post-COVID-19 syndrome, which are relative contraindications to PR. Caution with unexplained chest pain. Consider patients with functional limitation and ongoing symptoms for post-COVID-19 rehabilitation. Individuals with post-intensive care syndrome have multisystemic symptoms and deficits, which may require individualisation of exercise and education components. Fatigue and PESE should be closely monitored through symptom, exertion, activity scores and diaries.	High
Mylvaganam [22] 2021 USA	Guideline	Literature review and expert opinion	Fibrotic changes	Assessment, management	1)  Obtain PFTs at 3 months post-discharge in all survivors of critical COVID-19 and in those with mild/moderate disease with abnormal radiographic imaging 2)  There should be standardisation for CT appearances in PASC such as i) predominantly ground glass; ii) predominantly fibrotic; and iii) mixed ground glass and fibrotic. 3)  Further research should assess how acute clinical risk factors of COVID-19 (i.e. duration of illness, invasive or noninvasive ventilation, use of corticosteroids) affects the likelihood of PASC-pulmonary fibrosis. 4)  Urge to develop proven therapeutic options for patients who develop PASC-pulmonary fibrosis as none currently exist.	Medium
Nurek [23] 2021 UK	Guideline	Delphi study	Respiratory symptoms	Assessment, management	1)   In someone with long COVID, symptoms of possible non-COVID-19-related issues should be investigated and referred as per local guidelines. Long COVID alone is not a sufficient diagnosis unless other causes have been excluded. 2)   Carry out a face-to-face assessment including a thorough history and examination, consider other non-COVID-19-related diagnoses and measure full blood count, renal function, C-reactive protein, liver function test, thyroid function, HbA1c, vitamin D, magnesium, B12, folate, ferritin and bone studies 3)   Be aware that simple spirometry may be normal, but patients may have diffusion defects indicative of scarring, chronic pulmonary embolisms or microthrombi. Consider referral to respiratory for full lung function testing. 4)   Measure oxygen saturation at rest and after an age-appropriate brief exercise test in people with breathlessness and refer for investigation if hypoxaemic or if any desaturation on exercise. 5)   For patients with fatigue and worsening symptoms hours to days following an activity, emphasise the importance of an initial phase of convalescence followed by careful pacing and rest. 6)   Support patients in shifting their mental timeline of recovery to reflect the likely prolonged course, with a possibly long phased return to work. 7)   Further support patients with signposting to patient resources. Applicable resources may include: management of post-exertional symptom exacerbation, activity pacing, acupuncture and diagnosis-specific management as relevant. 8)   Provide patients with signposting to social prescribing, sickness certification and financial advice. Discuss with the patient whether sickness certification will state long COVID as diagnosis. 9)   Clinicians should ensure that the occupational status of patients with long COVID is recorded (in/out of work, part-/full-time, student). 10)  Follow patients up regularly to monitor progress from a full biopsychosocial and occupational perspective. 11)  Encourage reporting of new symptoms (expected) and expectation of waxing–waning course. 12)  Consider contributing patient data to research on long COVID, using the WHO Case Report Form or similar. 13)  For breathing pattern disorder, consider specialist physiotherapy and/or using alternative therapies such as pranayama breathing and meditation. 14)  In those with respiratory symptoms, consider CXR at an early stage. Be aware that a normal appearance does not exclude respiratory pathology.	High
Seo [105] 2024 South Korea	Guideline	Literature review and expert opinion	Respiratory symptoms (e.g. cough, dyspnoea, chest pain), pulmonary fibrosis, pulmonary embolism	Assessment, management	1)  Evaluation methods for long COVID patients complaining of: For respiratory distress, heart and lung-related tests should be considered to evaluate the presence of cardiopulmonary diseases. For chest pain, clinicians should consider prioritising evaluation for the cardiovascular, respiratory, musculoskeletal and gastrointestinal systems to exclude relevant conditions. For cough, evaluation should be considered based on criteria for chronic cough. Simple CXR and PFTs are recommended as initial tests. 2)  Methods for managing long COVID patients: For dyspnoea, adjustments in the dosage or frequency of previously used medications (e.g. inhalers) may be considered, or specific treatment for newly diagnosed conditions can be initiated. For cough symptoms, empirically, antihistamines and intranasal corticosteroid use may be considered. The use of anticoagulants or antiplatelet drugs for the purpose of preventing blood clots is not recommended. However, if a blood clot is diagnosed, treatment with anticoagulants or antiplatelet drugs is recommended according to the relevant guidelines. If pulmonary fibrosis is suspected in patients with long COVID, chest CT is recommended to assess the degree of pulmonary fibrosis. The use of antifibrotic drugs is not recommended without confirming the degree of pulmonary fibrosis. Respiratory rehabilitation therapy is recommended for patients with long COVID, considering underlying lung conditions, the need for ICU treatment, the presence of comorbidities (neurological and muscular disorders) and other relevant factors.	High
Visca [24] 2023, Italy	Guideline	Delphi study	Exercise limitation, dyspnoea	Assessment, management	1)  Patients with sequelae not explained by an alternative diagnosis should be evaluated for possible post-COVID-19 lung disease. 2)  Patients with lung function impairment, reduced exercise tolerance, reduced QoL or other relevant signs or ongoing symptoms 4 weeks after the onset of the first symptoms, should be evaluated for treatment and PR. 3)  The PR programme should be based on feasibility, effectiveness and cost-effectiveness criteria, organised according to local health services and tailored to the patient's needs. 4)  Each patient undergoing and completing PR should be evaluated to determine its effectiveness and have access to a counselling/health education session.	High
Yelin [12] 2022 Israel	Guideline	Literature review and expert opinion	Dyspnoea, pulmonary fibrosis, functional limitation, pneumonia, persistent cough	Assessment, management	1)  In symptomatic patients, other serious/life-threatening conditions should be ruled out prior to considering long COVID. As a first step, collecting specific clinical history is recommended to rule out previous underlying conditions, as well as iatrogenic causes or complications related to the acute episode. Hence, any patient with persisting or new symptoms that last more than 12 weeks after acute COVID-19 should be referred to medical care. For patients with symptoms 4–12 weeks after acute infection, assessment should be considered on a case-by-case basis, according to the severity and course of symptoms. 2)  The following may be considered for symptomatic patients according to symptoms: For patients with decreased oxygen saturation, blood gases are recommended by some guidelines, although the benefit of this test is limited. D-dimer should not be used in patients without respiratory symptoms. 3)  Evidence is insufficient to provide a recommendation for or against PFT. Considering that the test is simple and noninvasive and that future studies may suggest beneficial treatment for patients with abnormal PFT, the panel recommends considering routine PFT, including diffusing capacity, in all patients with severe and critical COVID-19 at 3 months from diagnosis, regardless of symptoms, as well as considering completing PFT with diffusion for patients reporting persistent dyspnoea 3 months after acute disease and those with known prior lung disease. 4)  Evidence is insufficient to provide a recommendation for or against CXR. CXR may be considered in patients with long COVID and persistent respiratory symptoms at 3 months to rule out other diagnoses and for a possible early diagnosis of pulmonary fibrosis. 5)  Evidence is insufficient to provide a recommendation for or against chest imaging. Chest CT should be considered at 3–6 months in patients with dyspnoea or abnormal PFTs, regardless of symptoms, to rule out other causes and identify fibrotic changes. 6)  Evidence is insufficient to provide recommendations for or against any functional tests. Consider performing them at the beginning of an interventional/rehabilitation programme to assess progress. 7)  Evidence is insufficient to provide a recommendation for or against PR specifically for COVID-19. No data regarding persistent long COVID were identified. Until further evidence accumulates, it is reasonable that clinicians follow available consensus statements regarding multidisciplinary rehabilitation in the post-acute stage. 8)  Evidence is insufficient to provide a recommendation for or against any intervention (for the management of persistent pulmonary symptoms). 9)  Evidence is insufficient to provide a recommendation for or against any intervention (for the management of persistent cough).	High

^#: To accurately convey guidance/recommendations the majority of text was extracted verbatim for most included articles. 6MWD: 6-min walk distance; 6MWT: 6-min walk test; ADL: activities of daily living; AGREE: Appraisal of Guidelines for Research and Evaluation; BP: blood pressure; CPET: cardiopulmonary exercise testing; CT: computed tomography; CXR: chest radiography; D_LCO: diffusing capacity of the lung for carbon monoxide; EQ-5D-5L: EuroQol five-dimension, five-level questionnaire; HbA1c: haemoglobin A1c; HR: heart rate; ICU: intensive care unit; OT: occupational therapy; PASC: post-acute sequelae of COVID-19; PCC: post-COVID-19 condition; PEM: post-exertional malaise; PESE: post-exertional symptom exacerbation; PFT: pulmonary function test; PR: pulmonary rehabilitation; PT: physical therapy; QoL: quality of life; SOB: shortness of breath; S_pO2: peripheral oxygen saturation; SPPB: short physical performance battery; UCSD SOBQ: University of California San Diego shortness of breath questionnaire; VAS: visual analogue scale; V′_O2max: maximal oxygen uptake; WHO: World Health Organization.

Recommendations for workup and assessment

Across included articles there was consistent emphasis on a broad, multi-system diagnostic approach. Seven (58%) guidance documents emphasised the importance of excluding other causes of breathlessness (i.e. underlying pulmonary disease) and to consider nonpulmonary contributors (neurologic, cardiovascular, autonomic, neuromuscular and activity-related phenomena such as post-exertional malaise (PEM)) before attributing symptoms to pulmonary-related PCC [12, 21, 23, 25–27, 105]. While this recommendation applies generally to all PCC patients, it also applies to pulmonary-related PCC patients. Across six (50%) guidance documents, the utility of pulmonary function tests (PFTs) (including diffusing capacity measurement) was highlighted [12, 21–23, 25, 26]. It was noted that the relatively simple and noninvasive nature of this diagnostic tool can guide the detection of reduced lung volumes, airflow limitations or impaired pulmonary diffusing capacity in pulmonary-related PCC to further guide management [12, 21–23, 25, 26]. In addition, six (50%) recommended imaging techniques (i.e. chest radiography and computed tomography (CT) scan) be considered for patients with persistent respiratory symptoms and to rule out other diagnoses and identify pulmonary fibrosis [12, 13, 21, 23, 25, 26]. However, access and equity were noted as potential challenges for high-cost diagnostics such as chest CT and Holter monitoring [13]. Ewing et al. [26] explicitly recommend cardiopulmonary exercise testing (CPET) for appropriately screened patients to aid in characterising exercise limitation and monitoring responses to interventions.

Recommendations for symptom management

Three (25%) guidance documents highlighted the importance of a multidisciplinary approach for managing pulmonary-related PCC [20, 23, 24]. Multidisciplinary management is recommended to include consultation with specialists, rehabilitation teams and access to PCC clinics or networks to optimise patient care and outcomes. Four (33%) guidance documents specifically noted limited evidence supporting the use of pharmacological interventions for treating pulmonary-related complications in PCC [12, 20–22]. For example, Mylvaganam et al. [22] found no consensus could be reached on the use of antifibrotic medication in patients presenting with pulmonary fibrosis after acute infection, and both Kim et al. [21] and Yelin et al. [12] found there was insufficient evidence to provide a recommendation for or against any pharmacological treatment for pulmonary-related complications in PCC. In contrast, three (25%) guidance documents recommended the use of inhaled or systemic steroids for patients with new obstructive lung disease, persistent cough, impaired lung function and/or interstitial abnormalities after SARS-CoV-2 infection [20, 25, 105]. Two guidelines (17%), both published from 2024 onward, recommended anticoagulant therapy in cases where diagnosed blood clots are present, treatment is provided by a physician and follows relevant guidelines; however, anticoagulant therapy was not recommended as a preventative measure for undiagnosed respiratory PCC patients [26, 105].

Eight (67%) guidance documents recommended PR for pulmonary PCC patients with new or ongoing dyspnoea, exercise intolerance, functional limitation, abnormal PFTs, persistent radiographic abnormalities, requiring supplemental oxygen or experiencing oxygen desaturation with exertion [13, 19, 20, 24–27, 105], and one document (published in 2022) found that the evidence was insufficient to provide a recommendation for or against PR at the time of publication [12]. Of the eight recommending PR, all guidance documents recommended individualisation of programmes (n=8, 100%) and six (75%) emphasised the importance of an initial assessment for patients entering a PR programme, including an assessment of dyspnoea, exercise capacity, physical function, HRQoL, mental health, fatigue and considerations for return to work (if applicable) [13, 19, 20, 24, 25, 27]. Three (25%) guidance documents recommended safety monitoring for myocarditis, deep vein thrombosis, pulmonary embolism and exercise-induced desaturation during rehabilitation [19, 26, 27].

What are the existing observational, case–control/series or RCT data on medical interventions for any of the above complications?

Nine primary studies (RCT, n=4; retrospective cohort study, n=2, prospective nonrandomised controlled study, n=1; observational cross-sectional, n=1; retrospective case series, n=1) and seven secondary studies (systematic review, n=3; systematic review and meta-analysis, n=2; scoping review, n=1; living systematic review, n=1) addressed research question 2. Included studies were published internationally (Canada, n=1; China, n=4; Germany, n=2; Italy, n=1; Korea, n=1; Malaysia, n=1; Serbia, n=1; Sweden, n=1; UK, n=1; and US, n=3). In primary studies, the number of participants ranged from 28 to 871 with the proportion of female identifying participants ranging from 26%–80%. In secondary studies, the number of included studies ranged from 10 to 35. Across secondary studies, participant characteristics were poorly reported (i.e. age, sex and comorbidities not reported). For primary studies, various concerns impacted the quality of included studies (see appendix 5). None of the included studies employed random participant selection, introducing potential selection bias [28–36]. Other common concerns in multi-arm studies were the lack of equivalence between comparison groups on sociodemographic characteristics (n=3, 50%), and a lack of equivalence at baseline for outcome measures (n=3, 50%) potentially introducing confounding factors and reducing internal validity when evaluating group differences. In contrast, strengths of these studies included random assignment of participants in the majority of multi-arm studies (n=4, 67%) and follow-up rates of 80% or more were identified in most studies (n=8, 89%). For secondary research, systematic review quality ranged from low risk of bias (n=3, 43%) to high risk of bias (n=4, 57%) (appendix 6). Additional study characteristics are summarised in table 2.

Primary research findings

Primary study findings are summarised in table 3. Three broad types of medical interventions for PCC treatment and management were identified, including therapeutic, pharmacologic and pharmacist-led education and counselling. Therapeutic modalities, including intermittent hypoxia/hyperoxia training (IHHT), intermittent hypoxia exposure (IHE), ozone autohemotherapy, hyperbaric oxygen and whole-body hyperthermia, were investigated in the literature. Doehner et al. [28] found that adding IHHT to standard outpatient rehabilitation yielded significant between-group improvements in 6-min walk test (6MWT) distance and Borg dyspnoea (p<0.05) and within-group improvements in lung function variables in the intervention group only (e.g. percent predicted forced expired volume in 1 s (FEV₁); p<0.05). Similarly, Zha et al. [36] showed that 7 days of IHE produced improvements in forced vital capacity (FVC), FEV₁ (absolute and % predicted), modified Medical Research Council (mMRC) and Borg dyspnoea scores, and 6MWT distance relative to normoxia exposure (median Δ=47 m versus Δ=23.5 m; p=0.001). In contrast to the above studies, He et al. [30] found that ozone autohemotherapy did not produce greater improvements in 6MWT distance compared to usual care for individuals with PCC (intervention Δ=58.5 m versus control Δ=36 m; p>0.05). Although a significant between-group difference in resting tidal volume favoured the ozone group (p=0.037), no other lung function measures differed significantly between groups [30]. Kjellberg et al. [31] found the impact of hyperbaric oxygen treatment was not different from sham treatment in terms of RAND-36 physical function score, 6MWT distance, 30 s sit-to-stand (STS) and EuroQol five-dimension quality of life score (EQ-5D). Vagedes et al. [35] demonstrated that inpatient whole-body hyperthermia delivered relief of sensory and perceptual dyspnoea at discharge (p<0.05); however, these gains were largely resolved by 4-week follow-up and cough persisted at all time-points.

TABLE 3.

Summary of findings for included studies for question 2

First author [ref.], year, country	Characteristics of medical intervention				Control group	Outcome measures	Summary of findings#
	Type	Setting	Description	Intervention length
Primary research (n=9)
Doehner [23] 2024 Germany	Therapeutic	Outpatient	IHHT involving individualised cycles of low-oxygen (10–12% O2) and high-oxygen (30–35% O2) breathing In addition, standard rehabilitation included physical training, breathing exercises, relaxation, education, occupational therapy, psychological counselling and management of comorbidities	Three 45-min sessions per week (∼5 weeks)	Standard rehabilitation only	1) 6MWT distance 2) Lung function 3) Borg dyspnoea	1) Both the intervention (Δ=91 m, p<0.001) and control (Δ=32 m, p<0.001) groups improved following rehabilitation; a significant between-group difference was detected, favouring the intervention group (p<0.001) 2) FEV1 % pred: the intervention group significantly improved (Δ=2.3%, p=0.009), while the control group did not (Δ=1.6%, p>0.05) PEF % pred: the intervention group improved significantly (Δ=3.0%, p=0.005), while the control group showed no change (Δ=0.6%, p>0.05) VC % pred: the intervention group improved (Δ=2.3%, p=0.009), while the control group showed no change (Δ=−0.2%, p>0.05) 3) Borg dyspnoea: both groups improved significantly: intervention group (Δ=−1.7, p<0.001); control group (Δ=−0.9 points, p<0.001); a significant difference between groups was observed, favouring the intervention group (p<0.01)
Geng [29] 2024 USA	Pharmacologic	Outpatient	Nirmatrelvir (300 g), in combination with low-dose ritonavir (100 g)	Twice daily for 15 days	Placebo in combination with low-dose ritonavir (100 g)	1) Symptoms 2) PROMIS dyspnoea 3) 60 s STS test	1) Reduction in moderate to severe shortness of breath in the intervention group at 10 weeks (−20.6%) as well as the control group (−24.5%) 2) High proportion of weeks with mild or no shortness of breath reported in intervention (median (IQR)): 0.769 (0.25–1.0) and control: 0.62 (0.09–0.89) with no difference between groups (p>0.05) 3) PROMIS dyspnoea reduced by (mean±sd) −1.96±7.90 in the intervention group and −2.38±6.13 in controls with no difference between groups (p>0.05) 4) No between-group differences
He [30] 2024  China	Therapeutic	Outpatient	Major ozone autohaemotherapy: involved collection of venous blood, mixing of ozone and re-infusion into participant's bloodstream	Daily treatment for 7 days	Conventional treatment	1) 6MWT distance 2) Lung function	1) Both groups improved 6MWT distance significantly: intervention group (Δ=58.5 m, p<0.001); control group (Δ=36 m, p<0.001); a significant difference between groups was detected for 6MWT distance %pred (p>0.05) 2) Both groups improved FEV1, FEV1 %pred, FVC, FVC %pred, FEV1/FVC, FEV1/FVC %pred and VT; a significant between-group difference in resting VT was detected in favour of the intervention group (p=0.037); no other between-group differences were identified
Kjellberg [31] 2025 Sweden	Therapeutic	Outpatient	Hyperbaric oxygen therapy (100% oxygen, 2.4 bar, 90 min per treatment)	10 sessions over 6 weeks	Sham treatment (room air treatment)	1) RAND-36 physical function 2) 6MWT distance 3) 30 s STS EQ-5D-5L index	1) No difference in RAND-36 physical function between groups at post-intervention follow-up (least square difference±se): 4.87±4.44, p=0.277 2) No difference in 6MWT distance between groups at post-intervention follow-up (least square difference±se): −7.48±15.22, p=0.62 3) No difference in 30 s STS between groups at post-intervention follow-up (least square difference±se): −0.37±0.59, p=0.53 4) No difference in EQ-5D-5L index between groups at post-intervention follow-up (least square difference±se): −0.04±0.04, p=0.36
Ntiamoah [32] 2024 USA	Pharmacologic	Outpatient (post-COVID-19 pulmonary clinic)	Systemic corticosteroids	Time between assessments ranged from 2 to 10 months	Nonsteroid-treated group with post-COVID-19 who visited the clinic	1) Spirometry 2) Lung imaging	1) Compared to baseline, significant improvements in FEV1, FVC and FVC % pred occurred in the steroid group only FEV1 (L) – steroid group: baseline (mean±sd): 2.33±0.67; follow-up: 2.66±0.90, p=0.008 Nonsteroid group – baseline (mean±sd): 2.64±0.70, follow-up: 2.72±0.73, p=0.3 FEV1 %pred – steroid group: baseline (mean±sd): 86±30, follow-up: 86±20, p=0.7 Nonsteroid group – baseline (mean±sd): 90±16, follow-up: 90±9, p=0.2 FVC (L) – steroid group: baseline (mean±sd): 2.93±0.82, follow-up: 3.28±1.11, p=0.007 Nonsteroid group: baseline (mean±sd): 3.33±0.91, follow-up: 3.45±1.11, p=0.3 FVC % pred – steroid group: baseline (mean±sd): 82±22, follow-up: 85±23, p=0.0008 Nonsteroid group: baseline (mean±sd): 90±18, follow-up: 89±13, p=0.2 FEV1/FVC – steroid group: baseline (mean±sd): 80±5, follow-up: 82±8, p=0.6 Nonsteroid group: baseline (mean±sd): 80±6, follow-up: 80±6, p=0.5 2) Compared to baseline imaging, there was a significant improvement in ground-glass opacities on follow-up in both the steroid and nonsteroid groups (Fisher's exact test <0.05 for both)
Sipetic [33] 2023 Serbia	Pharmacist delivered patient care counselling	Outpatient (community pharmacy)	SMART pharmacist programme: standardised evidence-based post-COVID-19 education and guidance to help patients manage and monitor symptoms (self-medication advice, follow-up visit, referrals as needed)	3–4 weeks	NA	Self-reported symptom severity (Wilcoxon signed-rank test) at first and second visits	More than 50% of patients with respiratory symptoms, especially coughing, were successfully treated with traditional over-the-counter interventions based on pharmacist advice A large spectrum of post-COVID-19 symptoms could be efficiently identified, treated and controlled in the self-medication process, based on pharmacist recommendations and counselling; however, pharmacists could also appropriately refer patients to other healthcare providers, whenever needed Follow-up rate was high (over 80% for patients with respiratory symptoms) Both patient and pharmacists need to prepare for the follow-up visit, which leads to better adherence, better information exchange, open dialogue and the identification of drug-related problems
Torres [34] 2024 USA	Pharmacologic	Outpatient	Treatment of post-COVID-19 ILD patients with prednisone (20–40 mg) daily and mycophenolate mofetil (500–1500 mg) twice daily Dose adjustments were based on patients’ response to treatment or development of side-effects Routine infection prophylaxis included trimethoprim–sulfamethoxazole	6 months	NA	1) FVC, FVC % pred 2) Oxygen requirement at rest (via nasal cannula) 3) 6MWT distance	1) FVC % pred was significantly improved at 6-month follow-up (baseline (mean±sd): 52±0.17%, 6 months: 60±0.20%, p=0.019); no significant change in FVC (L) following intervention (baseline (mean±sd): 1.94±0.53, 6 months: 2.24±0.62, p=0.086) 2) Oxygen requirement at rest was significantly reduced at 6-month follow-up (baseline (mean±sd): 1.96±1.79), 6 months: 0.89±1.29, p=0.005) 3) No significant change in 6MWT distance following intervention (baseline (mean±sd): 243±82.5, 6 months: 266±97.2), p=0.50)
Vagedes [35] 2025 Germany	Therapeutic	Inpatient	Fever-range WBH using specialised infrared devices. Patients laid in a heating chamber and were monitored by trained nurses The procedure involved heating the body to 38.8–39.8°C followed by a period to cool down	1–3 sessions each lasting up to 3 h with a 2-h cool-down period	NA	1) MDP sensory and immediate perception domains 2) Respiratory symptoms 3) Note: measured at hospital admission (t0), at discharge (t1) and 4 weeks after discharge (t2)	1) MDP sensory quality domain was significantly improved at t1 (−5.8, 95% CI −9.3–−2.4, p<0.05) and not different at t2 (−3.1, 95% CI −7.5–1.3, p≥0.05) MDP immediate perception score was significantly improved at t1 (−7.5, 95% CI −11.5–−3.6, p<0.05) and not different at t2 (−3.9, 95% CI −9.0–1.3, p≥0.05) 2) Persistent cough was not different at t1 (−0.2, 95% CI −0.4–0.1, p≥0.05) or at t2 (0.1, 95% CI −0.1–0.4, p≥0.05) Chest pain was significantly improved at t1 (−0.5, 95% CI −0.8–−0.1, p<0.05) and not different at t2 (−0.2, 95% CI −0.6–0.2, p≥0.05)
Zha [36] 2024 China	Therapeutic	Inpatient	IHE: participants underwent five cycles of alternating 5-min hypoxia (inhaled oxygen concentration 10–12%) and 5-min normoxia, targeting an SpO2 range of 75–85%; if SpO2 dropped below 75% or if discomfort occurred, the session was halted and oxygen was administered until SpO2 returned to 95%; IHE was delivered for a minimum of 7 days Routine therapies: all participants received symptom-based standard care, including inhaled bronchodilators and nebulised therapies with corticosteroids and anticholinergic bronchodilators	Minimum of 7 days of treatment, median (IQR) treatment of 10.0 (9.0–15.0) days	NE: participants inhaled compressed air continuously for 50 min while receiving the same nitrogen flow as in IHE; the gas mixture was adjusted to blind participants to group allocation; NE was also administered for at least 7 days Routine therapies: all participants received symptom-based standard care, including inhaled bronchodilators and nebulised therapies with corticosteroids and anticholinergic bronchodilators	1) Lung function 2) Borg dyspnoea 3) mMRC dyspnoea 4) 6MWT distance	1) Significant improvement in FVC following intervention in the IHE group (0.2, 95% CI 0.1–0.4 L) compared to the NE group (0.1, 95% CI 0.0–0.3 L), p=0.001 Significant improvement in FVC % pred following intervention in the IHE group (6.1, 95% CI 4.2–10.6%) compared to the NE group (3.2, 95% CI –0.9–8.8%), p=0.001 Significant improvement in FEV1 following intervention in the IHE group (0.1, 95% CI 0.1–0.3 L) compared to the NE group (0.1, 95% CI 0.0–0.2 L), p=0.001 Significant improvement in FEV1 % pred following intervention in the IHE group (5.3, 95% CI 4.1–9.9%) compared to the NE group (2.1, 95% CI –0.8–6.9%), p=0.001 2) Significant decrease in Borg dyspnoea score following intervention in the IHE group (1.0, 95% CI 0.0–1.0 points) compared to the NE group (0.0, 95% CI 0.0–1.0 points), p=0.031 3) Significant decrease in mMRC dyspnoea score following intervention in the IHE group (0.0, 95% CI 0.0–1.0 points) compared to the NE group (0.0, 95% CI 0.0–0.0 points), p=0.001 4) Significant improvement in 6MWT distance following intervention in the IHE group (47.0, 95% CI 30.0–61.0 m) compared to the NE group, p=0.001
Secondary research (n=7)
Li [37] 2023  China	Symptom management and therapeutic regimens	NR	Stopping smoking, avoiding pollutants, and regular exercise, inspiratory muscle training, music-based breathing and human UC-MSC administration	NR	NR	NR	Inspiratory muscle training and music-based breathing training elicited clinical improvements in chest symptoms and difficult breathing UC-MSC administration revealed an excellent outcome symptoms and lung lesion improvement; nevertheless, current treatment strategies remain inadequate and the majority are based on previous experience in similar diseases and pilot studies with crude designs Although hundreds of clinical trials have been registered, few of them have been widely used in clinical practice Multidisciplinary approaches should be applied to manage long COVID
Ora [38] 2023 Italy	Pharmacologic	NR	Sodium pyruvate nasal spray, cannabis sativa 50 mm·mL−1 Combination product: hyperbaric oxygen therapy, montelukast, S-1226 8% (gas mixture containing carbon dioxide and small volumes of nebulised perflubron), coenzyme Q10, hyperbaric oxygen, sirolimus, Ampion	NR	NR	Symptom change, symptom severity score, tolerability, number of side-effects, exercise tolerance, 6MWT, dyspnoea perception, CAT scale, 1-min STS test, O2 desaturation, treatment-emergent adverse effects normalisation in 6MWT distance, number of self-reported symptoms (EQ-5D-5L), RAND 36 change, prevalence of pulmonary fibrosis as evidenced by CT scan, 10% threshold for pulmonary fibrosis evidenced by CT, safety	Only one completed trial with available results was found (sodium pyruvate nasal spray) which showed improved coughing and trouble breathing with no adverse events To date, there are no established drugs for the treatment of long-COVID; most are focused on drugs already known or tested in other pathologies that act on the reduction or modulation of systemic inflammation This is probably due to the difficulty of identifying a single pathogenetic mechanism
Seo [39] 2025 South Korea	Scoping review	NR	Antifibrotic agents (nintedanib and pirfenidone)	Range: 12 weeks to 12 months	NR (two studies) Prednisolone (one study)	PFTs, 6MWT distance, oxygen saturation, radiological scores, CTSS	Nintedanib showed superior improvement in 6MWT distance and oxygen saturation but was associated with frequent side-effects (e.g. diarrhoea in 80% of patients) 12 months of pirfenidone and nintedanib did not result in significant differences between groups in lung function parameters (p>0.05) Significant improvement in CTSS with nintedanib compared to pirfenidone plus corticosteroids and corticosteroids alone (p<0.05); no significant advantage of pirfenidone over steroids alone
Tsang [40] 2025 China	Four studies: CHM One study: CHM plus standard rehabilitation exercise	NR	Bufei huoxue capsule (0.35 g): 4 capsules three times daily for 3 months (n=1) Qingjin yiqi granules plus usual care (10 g): twice daily for 2 weeks (n=1) Ludangshen oral liquid (10 mL): twice daily for 2 weeks (n=1) Shengmai yin (10 mL): three times daily for 2 weeks (n=1) Jinshuibao capsule (0.42 g) four capsules three times daily for 2 weeks (n=1)	2 weeks (n=4) 3 months (n=1)	Usual care (placebo)	Borg dyspnoeas scale, 6MWT distance	Compared to the control, CHM reduced dyspnoea on the Borg dyspnoea scale score (MD −0.2, 95% CI −0.65–0.25) with moderate certainty Meta-analysis of two RCTs (517 patients) showed that CHM clinically improves exercise intolerance by increasing 6MWT distance (MD −15.92, 95% CI −10.20–42.05) with substantial heterogeneity (I2=68%) and low certainty of evidence
Yong [41] 2023 Malaysia	Pharmacologic (n=1)	NR	Treamid	NR	Placebo	Functional exercise capacity, lung function, dyspnoea, quality of life, lung fibrosis, recovery, recurring symptoms	Of the drugs examined in completed RCTs, only Treamid was relevant to pulmonary-related PCC Treamid did not lead to improvements in respiratory symptoms for people with respiratory PCC The authors concluded there were no effective pharmacological treatments for respiratory PCC at the time of publication
Zeraatkar [42] 2024 Canada	Pharmacologic (n=1)	NR	Leronlimab	8 weeks	Placebo	Dyspnoea (measurement instrument NR)	No significant impact on dyspnoea (MD −0.23, 95% CI −0.75–0.29), low certainty of evidence
Zheng [43] 2022 UK	Supplementation (n=1)	NR	Pycnogenol–Centellicum	NR	NR	Dyspnoea (breathlessness)	One small-scale observational study showed that the use of Pycnogenol–centellicum supplementation was associated with improved breathlessness after COVID-19

6MWT: 6-min walk test; CAT: COPD Assessment Test; CHM: Chinese herbal medicine; CT: computed tomography; CTSS: computed tomography severity score; FEV₁: forced expiratory volume in 1 s; EQ-5D-5L: EuroQol five-dimension five-level questionnaire; FVC: forced vital capacity; IHE: intermittent hypoxia exposure; IHHT: Intermittent hypoxic–hyperoxic training; ILD: interstitial lung disease; IQR: interquartile range; MD: mean difference; MDP: multidimensional dyspnoea profile; mMRC: modified Medical Research Council; PCC: post-COVID-19 condition; PFT: pulmonary function test; NA: not applicable; NE: normoxia exposure; NR: not reported; PEF: peak expiratory flow; PROMIS: Patient-Reported Outcomes Measurement Information System; RCT: randomised controlled trial; SMART: specific, measurable, achievable, relevant and time-bound; S_pO2: peripheral oxygen saturation; STS: sit to stand; UC-MSC: umbilical cord-derived mesenchymal stem cell; VC: vital capacity; V_T: tidal volume; WBH: whole-body hyperthermia. ^#: Text extracted verbatim for most included studies.

Pharmacological interventions yielded limited evidence of effectiveness for treatment and management of PCC. Geng et al. [29] showed that nirmatrelvir+ritonavir reduced the number of reported weeks participants experienced moderate-to-severe dyspnoea, but no difference was observed compared to the control group (placebo+ritonavir). In addition, no between-group differences in symptom reduction, dyspnoea or 60-s STS performance were identified (p>0.05 for all) [29]. Ntiamoah et al. [32] found that systemic corticosteroids significantly improved FEV1, FVC and FVC % pred at follow-up (p<0.01), whereas the control group showed no improvement. Radiologic findings did not improve in either group and between-group comparisons were not reported [32]. Torres et al. [34] observed that 6 months of prednisone+mycophenolate significantly increased FVC % pred (baseline: 52%, follow-up: 60%, p=0.019) and reduced resting oxygen needs (baseline: 1.96, follow-up: 0.89 L·min⁻¹, p=0.005) but did not have a significant impact on 6MWT distance (baseline: 243 m, follow-up: 266 m, p=0.50).

Sipetic et al. [33] evaluated a pharmacist-delivered education and counselling programme, featuring standardised, evidence-based education and medication guidance in a community pharmacy setting. The authors reported that more than half of patients with respiratory PCC symptoms (e.g. cough) achieved symptom relief via over-the-counter interventions guided by pharmacists, with high follow-up adherence (>80%) and timely referrals for complex cases.

Secondary research findings

Secondary study findings were summarised in table 3. Secondary studies identified various additional medical interventions for respiratory-related complications in PCC [37, 38, 41, 43]. The review by Ora et al. [38] examined completed and in-progress registered clinical trials to understand what interventions were in the pipeline for PCC management. Only one completed trial relevant to respiratory PCC symptom management was identified, which studied the impact of a nasal sodium pyruvate spray in people with PCC and found a reduction in coughing with no adverse events [38]. A variety of other pharmacological interventions were identified among trials that were still ongoing (10 registered trials were recruiting participants and three were not yet recruiting) [38]. Similarly, out of four published RCTs and over 20 ongoing registered RCTs, Yong et al. [41] identified only one completed RCT and two ongoing RCTs that were relevant to pulmonary-related PCC. In the completed RCT, the authors were investigating Treamid for symptom management and found it did not lead to improvements in respiratory symptoms for people with respiratory PCC [41]. At the time of writing, both ongoing trials have been terminated without results posted (NCT04695704, NCT04652518) [41]. In contrast, the review by Li et al. [37] identified one RCT (n=100) that found human umbilical cord-derived mesenchymal stem cell administration improved symptoms and improved lung lesion volume in people with respiratory PCC. Additionally, Zheng et al. [43] reported in their review that one small observational study (n=18) found Pycnogenol-Centellicum supplementation was associated with improved breathlessness. Further, Zeraatkar et al. [42] found there was no significant effect of leronlimab on dyspnoea (mean difference (MD) −0.23, 95% CI −0.75–0.29; low certainty of evidence), Tsang et al. [40] identified small effects of Chinese herbal medicines on Borg dyspnoea (MD −0.2, 95% CI −0.65–0.25; moderate certainty) and 6MWT distance (MD −15.92, 95% CI −10.20–42.05; I²=68%, and low certainty of evidence), and Seo et al. [39] found that nintedanib may improve exercise capacity (6MWT distance) and CT severity scores compared to controls (p<0.05), but adverse events were common (i.e. 80% of participants reported diarrhoea).

What are the existing observational, case–control/series or RCT data on rehabilitation interventions for any of the above complications?

28 primary research studies (RCT, n=9; prospective cohort study, n=5; prospective observational study, n=5; retrospective cohort study, n=4; retrospective observational study, n= 2, quasi-experimental study, n=2; prospective nonrandomised controlled study, n=1) and 20 secondary studies (systematic review and meta-analysis, n=9; systematic review, n=6; scoping review, n=2; rapid review, n=1, living systematic review, n=1; systematic review and network meta-analysis, n=1) addressed research question 3. The highest proportion of primary studies were published in the UK (n=5, 18%), followed by Norway (n=3, 11%) and Italy (n=3, 11%), with additional publications from India (n=2, 7%), Brazil (n=2, 7%), France (n=2, 7%), Spain (n=2, 7%) and Switzerland (n=2, 7%), and single contributions (n=1, 4% each) from Germany, Sweden, US, Poland, Austria, Hungary and Turkey.

Among the included primary studies, the number of participants ranged from 32 to 1438 and the proportion of female-identifying participants ranged from 34% to 86%. 21 (75%) of the primary studies reported participant comorbidities, with hypertension (n=12, 43%), diabetes (n=10, 36%), and asthma (n=9, 32%) commonly reported (see table 4). Three (15%) of the included secondary studies were published in Canada and the rest internationally (Belgium, n=1; Brazil, n=2; China, n=3; France, n=1; Germany, n=1; Ireland, n=1; Italy, n=1; South Africa, n=1; South Korea, n=1; Spain, n=2; UK, n=2). Among the secondary studies, the number of included studies ranged from two to 38. 12 (60%) secondary studies reported the number of included participants (range 314–1886); however, participant characteristics were not well reported (i.e. age, sex and comorbidities not reported).

TABLE 4.

Characteristics of included studies for question 3

First author [ref.], year, country	Study design	Participants, n	Participant characteristics			Post-COVID-19 condition (PCC)		Summary of risk of bias or study quality
			Age, mean±sd years	Female, n (%)	Comorbidities, n (%)	Definition of PCC	Pulmonary complications, n (%)
Primary research (n=29)
Acharya [58] 2025  India	Prospective cohort study	114	43.2±11.8	54 (47%)	Hypertension 32 (28.1%), diabetes 25 (21.9%), asthma 12 (10.5%)	Persistent symptoms ≥12 weeks post-infection	NR	See appendix 5
Altmann [60] 2023  Germany	Prospective observational study	42 (n=21 COVID-19; n=21 long COVID)	NR for overall n	21 (50%)	2 (10%) COPD/asthma	Persistent symptoms >4 months after acute infection	Pulmonary embolism 3 (14%); pulmonary artery hypertension 1 (5%); dyspnoea (NR)	See appendix 5
Armstrong [85] 2024  UK	RCT	32 (intervention: n=17, control: n=15)	Intervention: 44±12 Control: 46±12	Intervention: 9 (53%) Control: 8 (53%)	NR	Persistent symptoms ≥12 weeks post-infection	Intervention: breathlessness 15 (88%), chest tightness 4 (24%), cough 11 (65%) Control: breathlessness 13 (87%), chest tightness 5 (33%), cough 10 (67%)	See appendix 5
Bargaje [59] 2024  India	Retrospective cohort study	155	52.2±13.9	53 (34%)	Hypertension 46 (29.7%), diabetes mellitus 39 (25.2%), COVID-induced diabetes 23 (14.8%), bronchial asthma 8 (5.2%), ischaemic heart disease 2 (1.3%), hypothyroidism 2 (1.3%)	Persistent symptoms ≥3 months post-infection	Cough 7 (5%), breathlessness 130 (84%)	See appendix 5
Bileviciute-Ljungar [61] 2024 Sweden	RCT	116 (n=67 intervention, n=56 control)	Intervention: 43±9 Control: 47±9	Intervention: 52 (78%) Control: 37 (88%)	Hypertension: intervention 2 (3%), control 1 (5%) Hypothyroidism: intervention 3 (5%), control 1 (2%) Asthma: intervention 5 (8%), control 4 (10%) Allergies: intervention 3 (5%), control 1 (2%) Skin disease: intervention 1 (1%) Anxiety/depression: intervention 7 (10%), control 2 (5%) Bipolar disorder: intervention 1 (1%), control 1 (2%) PTSD: intervention 1 (1%) Arthritis: intervention 1 (1%), control 1 (2%) Chronic pain: intervention 3 (5%), control 1 (2%) Gastrointestinal disease: intervention 3 (5%) Kidney disease: control 1 (2%) Gynaecological disease: intervention 1 (1%), control 2 (5%) Sleep disorders: control 2 (5%)	Persistent symptoms ≥12 weeks post-infection	Breathing function impaired (n not specified)	See appendix 5
Campos [86] 2024  Brazil	Prospective nonrandomised controlled study	37 (n=22 intervention, n=15 control)	Intervention: 40.8±10.0 Control: 45.4±10.5	Intervention: 12 (55%) Control: 6 (40%)	Depression: intervention 7 (31.8%), control 4 (26.7%) Hypercholesterolaemia: intervention 5 (22.7%), control 4 (26.7%) Diabetes: intervention 3 (13.6%), control 2 (13.3%) Hypertension: intervention 2 (9.1%), control 3 (20.0%)	Not exactly specified, but all participants were at least 3 months post-acute infection (table 1)	At baseline, 87% of the sample reported dyspnoea	See appendix 5
Campos [87] 2024  Brazil	Retrospective observational study	65 (n=33 intervention, n=32 control)	Intervention: 41.3±10.6 Control: 45.2±10.9	Intervention: 33 (73%) Control: 32 (69%)	Hypertension: intervention 9 (27.3%); control 13 (40.6%) Diabetes: intervention 6 (18.2%); control 10 (31.2%) Cardiopathy: intervention 4 (12.1%); control 4 (12.5%) Asthma: intervention 6 (18.2%); control 2 (6.3%) COPD: intervention 2 (6.1%); control 3 (9.4%)	Persistent symptoms ≥3 months post-infection	NR	See appendix 5
Colas [62] 2023  France	Prospective observational study	114 (n=38 long COVID, n=38 coronary artery disease, n=38 fibromyalgia)	Long COVID: 46.9±12.7, Coronary artery disease: 61.4±9.45 Fibromyalgia: 47.4±9.93	65 (57%)	Overweight 57 (50%), tobacco consumption 14 (37%)	Persistent symptoms ≥3 months post-infection	Exertional dyspnoea 32 (84%), persistent cough 2 (<5%)	See appendix 5
Daynes [63] 2025  UK	RCT	181 (n=56 face-to-face rehab, n=63 virtual rehab, n=62 control)	Face-to-face rehab: 61±13 Virtual rehab: 55±11 Control: 62±11	Face-to-face rehab: 25 (45%) Virtual rehab: 34 (54%) Control: 23 (37%)	One comorbidity: face-to-face rehab 11 (20%), virtual rehab 9 (14%), control 15 (24%) ≥2 comorbidities: face-to-face rehab: 34 (61%), virtual rehab: 34 (54%), control: 36 (58%)	Persistent symptoms ≥12 weeks post-infection	Baseline mMRC dyspnoea (median (IQR)): face-to-face rehab: 3 (2–4), virtual rehab: 2 (2–3), control: 3 (2–3)	See appendix 5
del Corral [64] 2025  Spain	RCT	64 (n=32 intervention, n=32 control)	Intervention: 49.0±10.4 Control: 51.4±10.6	Intervention: 21 (66%) Control: 20 (62%)	NR	Persistent symptoms ≥3 months post-infection	Dyspnoea: intervention 32 (100%), control 32 (100%) Chest pain: intervention 6 (19%), control 6 (19%)	See appendix 5
Frisk [65] 2023  Norway	Quasi-experimental study	78	40.3±12	64 (82%)	Obesity 16 (21%)	Persistent symptoms ≥3 months post-infection	Exercise limitation or desaturation (NR), dyspnoea 49 (63%)	See appendix 5
Frisk [66] 2025  Norway	Quasi-experimental study	As above	As above	As above	As above	As above	As above	See appendix 5
Mammi [67] 2023  Italy	Retrospective observational study	50	53.0±11.4	29 (58%)	NR	Persistent symptoms ≥3 months post-infection (n=31 (62%) met this definition)	Exercise limitation or desaturation (NR)	See appendix 5
Moine [68] 2024  France	Prospective observational study	47	51±12	29 (62%)	Respiratory 18 (38%), cardiovascular 14 (30%), metabolic 13 (28%)	Persistent symptoms ≥6 months after the initial COVID-19 infection and with a confirmed association between persistent symptoms and COVID infection using clinical examination with a medical doctor	Dyspnoea 43 (91%), chest pain 20 (43%), cough 14 (30%), sore throat 5 (11%)	See appendix 5
Nasrullah [69] 2024  USA	Retrospective cohort study	55	58±14	33 (60%)	Congestive heart failure 2 (4%), obstructive sleep apnoea 13 (24%), asthma 26 (47%), pulmonary hypertension 4 (7%), diabetes mellitus 7 (13%), hypertension 51 (93%), ILD 1 (2%), chronic kidney disease 53 (96%), liver cirrhosis 0 (0%), dialysis dependent 0 (0%), HIV infection 0 (0%), malignancy – active 2 (4%), remission 5 (9%), vascular disease 1 (2%), depression 23 (42%), transplant 0 (0%)	Persistent symptoms ≥3 months post-infection	NR	See appendix 5
Nerli [70] 2024  Norway	RCT	310 (n=154 intervention, n=156 usual care)	Intervention: 43±12 Usual care: 42±12%	Intervention: 122 (78%) Usual care: 103 (66%)	None: intervention 93 (60%), control 94 (61%) Asthma, allergy and atopy: intervention 27 (17%), control 27 (17%) Psychological: intervention 5 (3.2%), Control 6 (3.8%) Migraine and headache syndromes: intervention 7 (4.5%), control 7 (4.5%) Cardiovascular: intervention 8 (5.1%), control 8 (5.1%) Endocrinologic: intervention 7 (4.5%), control 9 (5.7%) Gynaecological: intervention 5 (3.2%), control 2 (1.3%) Gastroenterological: intervention 12 (7.6%), control 5 (3.2%) Pain syndromes: intervention 4 (2.5%), control 6 (3.8%) Other: intervention 12 (7.6%), control 13 (8.3%)	Persistent symptoms ≥3 months post-infection	Shortness of breath 167 (54%), cough 132 (43%)	See appendix 5
Onik [71] 2024  Poland	Retrospective cohort study	122	65.0±8.6	71 (58.2%)	Hypertension 70 (57%), osteoarthritis 27 (22%), type 2 diabetes mellitus 23 (19%), hypothyroidism 16 (13%), gout 4 (3%), benign prostatic hyperplasia 1 (1%)	Persistent symptoms ≥3 months post-infection	Dyspnoea at rest (NR), exercise-induced dyspnoea (NR), cough (NR), chest tightness (NR), chest pain (NR), sputum (NR)	See appendix 5
Ostrowska [72] 2023  Switzerland	Prospective observational study	97	Median (IQR): 60 (50–68)	53 (54.6%)	Hypertension 45 (46.4%), hyperlipidaemia 24 (27.7%), coronary artery disease 18 (18.6%), heart failure 6 (6.2%), COPD 12 (12.4%), active smoker 11 (11.3%)	Persistent symptoms ≥3 months post-infection	Significant dyspnoea (23.7%), pulmonary embolism (2.0%), exercise intolerance (73.2%), persistent cough (24.7%)	See appendix 5
Ovejero [73] 2025  Spain	RCT	98 (n=51 intervention, n=47 control)	Intervention: 47.8±9.9 Control: 49.0±8.4	Intervention: 45 (88%) Control 39 (83%)	NR	Persistent symptoms ≥3 months post-infection	Respiratory symptoms at baseline intervention: 46 (90%) Control: 35 (75%)	See appendix 5
Parker [74] 2023  UK	Prospective cohort study	34	47±9	22 (71%)	Hypertension 5 (16%), diabetes 4 (13%), respiratory conditions 3 (13%), anxiety 5 (16%), depression 4 (13%), cardiovascular conditions 4 (13%)	Persistent symptoms ≥3 months post-infection	Exercise limitation of desaturation (NR), dyspnoea (NR)	See appendix 5
Philip [75] 2024  UK	Retrospective cohort study	1438	49.0±11.9	1150 (80%)	Asthma 360 (25%), COPD 9 (1%), heart disease 30 (2%), hypertension 180 (13%), diabetes mellitus 77 (6%)	Not specifically defined; mean±sd days of persistent symptoms at baseline 423±216	Persistent breathlessness 1438 (100%)	See appendix 5
Resta [77] 2024  Italy	Prospective cohort study	187	58.1±13.5	87 (46.5%)	Hypertension 71 (38.0%), Left heart failure 18 (4.8%), COPD: 16 (8.6%), type 2 diabetes 14 (7.5%), asthma 13 (7.0%), thyroid disease 10 (5.4%), chronic ischaemic heart disease 3 (1.6%), arrhythmias 2 (1.1%), pulmonary fibrosis 1 (0.5%), fibromyalgia 1 (0.5%), multiple sclerosis 1 (0.5%), ulcerative colitis 1 (0.5%), psoriasis 1 (0.5%), neoplasm 1 (0.5%)	Persistence of dyspnoea >3 months after the latest infection	Dyspnoea 134 (71.7%)	See appendix 5
Resta [76] 2024  Italy	Prospective cohort study	327	58.8±14.1	151 (46.3%)	182 (55.7%) participants had one or more comorbidities (chronic respiratory conditions accounting for over 20%)	Persistence of dyspnoea >3 months after the latest infection	Dyspnoea 247 (75.5%)	See appendix 5
Rzepka-Cholasinska [82] 2024   Switzerland	Prospective cohort study	90	61.7±5.4	49 (NR)	Hypertension 38 (42.2%), diabetes 14 (15.5%), osteoarthritis 23 (25.6%), history of MI 10 (11.1%), bronchial asthma 4 (4.4%), bronchiectasis 1 (1.1%), COPD 12 (13.3), thyroid disorders 15 (16.7%)	Persistent dyspnoea >12 weeks after acute infection (but <1 year)	Dyspnoea 90 (100)	See appendix 5
Sick [78] 2025  Austria	RCT	42 (n=14 endurance training; n=14 concurrent training; n=14 control)	Endurance: 41.8±11.8 Concurrent: 41.6±14.7 Control: 40.3±10.8	Endurance: 12 (85.7%) Concurrent: 10 (71.4%) Control: 11 (78.6%)	NR	Laboratory-confirmed SARS-CoV-2 infection at least 12 weeks prior to enrolment and reported at least one symptom specific to PCC according to the NICE definition	Breathlessness: endurance: 7 (50%); concurrent: 10 (71.4%); control: 6 (42.9%) Chest tightness: endurance: 3 (21.4%); concurrent: 6 (42.9%); control: 6 (42.9%) Cough: endurance: 3 (21.4%); concurrent: 2 (14.3%); control: 2 (14.3%)	See appendix 5
Smith [79] 2023  UK	Prospective observational study	601	47±10	465 (77.4%)	Comorbidities per participant (mean±sd) 2.9±1.7	≥12 weeks post-infection	NR	See appendix 5
Szarvas [80] 2024  Hungary	RCT	200 (intervention: n=100, control: n=100)	Intervention (median (IQR)): 56 (48–68) Control: 56 (47.8–66)	Intervention: 43 (43%) Control: 43 (43%)	NR	Persistent symptoms ≥12 weeks post-infection that cannot be otherwise explained	Intervention: chest pain 70 (70%), coughing 53 (53%) Control: chest pain 92 (92%), coughing 51 (51%)	See appendix 5
Yasaci [81] 2025  Turkey	RCT	64 (intervention: n=32, control: n=32)	Intervention: 56.5±13.4 Control: 55.5±9.5	Intervention: 14 (44%) Control: 17 (61%)	NR	Persistent symptoms ≥3 months post-infection	mMRC dyspnoea ≥2 at baseline (100%)	See appendix 5
Secondary research (n=20)
Aiyegbusi [83] 2021  UK	Systematic review	27 studies	NR	NR	NR	>4 weeks post-infection	Dyspnoea NR (32%), cough NR (18%)	High risk of bias
Arienti [84] 2023  Italy	Systematic review	21 studies (respiratory relevant), 1443 participants	NR	NR	NR	≥3 months after the onset	Dyspnoea, pneumonia, or lung function impairment (n=21 studies, 100%)	High risk of bias
Cha [6] 2024  South Korea	Scoping review	34 studies (three relevant to pulmonary, seven relevant to general)	NR	NR	NR	>4 weeks post-infection	NR	High risk of bias
Dillen [44] 2023  Belgium	Systematic review	38 studies	NR	NR	NR	>4 weeks post-acute infection	Dyspnoea (NR), general pulmonary function (NR)	Low risk of bias
Martinez-Pozas [45] 2024  Spain	Systematic review and network meta-analysis	10 studies, 765 participants	Range: 22–66	561 (73.3%)	NR	≥3 months after onset of symptoms	NR	High risk of bias
Martinez-Pozas [46] 2024  Spain	Systematic review and meta-analysis	Seven studies (respiratory relevant), 365 participants	Range: 45.2–52.2 (range is mean years across studies)	NR	NR	≥3 months after initial infection	Dyspnoea (n=365 participants, 100%)	High risk of bias
McDowell [47] 2025  Ireland	Systematic review of RCTs	Eight studies (respiratory relevant), 447 participants	49.9±NR	Mean 53.9% (n=NR)	NR	Multiple definitions (i.e. persistent symptoms ≥40 days after an initial infection; ≥2 months; ≥3 months)	Dyspnoea (NR)	Low risk of bias
Neto [48] 2025  Brazil	Scoping review	Two studies (respiratory relevant)	NR	NR	NR	>12 weeks	Dyspnoea (NR)	High risk of bias
Ora [38] 2023#  Italy	Systematic review	35 studies (n=17 medical interventions)	NR	NR	NR	Persistent symptoms >4 weeks after an initial infection	Dyspnoea (NR), fibrosis (NR), unspecified pulmonary symptoms (NR), exercise limitation (NR)	High risk of bias
Pouliopoulou [49] 2023  Canada	Systematic review and meta-analysis	14 studies, 1244 participants	50±NR	559.8 (45%)	NR	>3 months after initial infection	Dyspnoea (n=8 studies, n=573 participants), exercise limitation (NR), respiratory function (n=6 studies, 363 participants)	Low risk of bias
Romanet [50] 2025  France	Systematic review and meta-analysis	19 studies, 1292 participants	49±NR	NR	NR	≥3 months post-initial infection	Persistent rest or exertional breathlessness/dyspnoea (n=19 studies, 100%)	Low risk of bias
Schurr [51] 2025  Germany	Systematic review	Four studies (respiratory relevant), 349 participants	NR	NR	NR	Persistent symptoms ≥12 weeks post-infection	Dyspnoea (NR), general respiratory symptoms (NR)	High risk of bias
Soril [52] 2022  Canada	Rapid review	Nine studies, 764 participants	NR	330 (43.2%)	NR for overall n	>3 weeks after acute infection	Dyspnoea (NR), exercise limitation (NR), exercise induced dyspnoea (NR), respiratory symptoms (NR)	Low risk of bias
Tan [53] 2025  China	Systematic review and meta-analysis	48 studies (number respiratory relevant NR)	NR	NR	NR	≥3 months after initial infection	General pulmonary involvement, cough, dyspnoea (NR)	High risk of bias
Torres [54] 2023  South Africa	Systematic review and meta-analysis	32 studies, 1886 participants	Range: 18–75	NR	NR	>3 weeks from the initial detection of acute symptoms	Exercise limitation (NR), respiratory function (NR)	High risk of bias
Xavier [55] 2024  Brazil	Systematic review and meta-analysis	Seven studies, 527 participants	Range: 44±11.3–50.40±12.1	223 (58.8%) reported in 6 of 7 studies, total n=379	NR	NR	Symptoms lasting >12 weeks	High risk of bias
Yang [56] 2024  China	Systematic review and meta-analysis	Three studies (respiratory relevant)	NR	NR	NR	Definitions ranging from months to years post-acute infection	NR	Low risk of bias
Zeraatkar [42] 2024#  Canada	Living systematic review	Two studies (pulmonary relevant), 314 participants	NR	NR	NR	≥3 months after initial infection	Pulmonary symptoms (NR)	Low risk of bias
Zheng [43] 2022#  UK	Systematic review and meta-analysis	Nine studies (relevant to rehabilitation)	NR	NR	NR	Ongoing symptoms lasting >4 weeks after acute infection	Dyspnoea (NR), reduced exercise capacity (NR)	Low risk of bias
Zheng [57] 2024  China	Systematic review and meta-analysis	23 studies, 1579 participants	Range: 18–84	752 (47.6%)	NR	≥3 months post-infection	Dyspnoea (NR), persistent cough (NR)	High risk of bias

ILD: interstitial lung disease; IQR: interquartile range; MI: myocardial infarction; mMRC: modified Medical Research Council; NICE: National Institute for Health and Care Excellence; NR: not reported; PCC: post-COVID-19 condition; PTSD: post-traumatic stress disorder; RCT: randomised controlled trial; SARS-CoV-2: severe acute respiratory syndrome coronavirus 2. ^#: Study also reported in table 2; this study provides data to support research questions 2 and 3.

Various concerns impacted the quality of the included primary studies. Out of 28 studies, the most common sources of bias (no or NR responses; see appendix 5) were a lack of random selection of participants (n=28, 100%) and absence of a control/comparison group (n=15, 54%). Other frequent issues were unclear or inadequate follow-up (n=11, 39%) and lack of equivalence between comparison groups on sociodemographics (n=5, 18%) or at baseline on outcome measures (n=5, 18%). Study quality for secondary research ranged from low risk of bias (n=8, 40%) to high risk of bias (n=12, 60%) (appendix 6). Additional study and participant characteristics are summarised in table 4.

Primary research findings

All primary study findings are summarised in table 5. Frisk et al. [65, 66] reported on the same sample of participants at different time-points and are thus being summarised as one record. Multicomponent exercise rehabilitation was the most common intervention studied (n=17, 63%), followed by spa rehabilitation/balneotherapy programmes (n=4, 15%) and single-study interventions including micro-choice rehabilitation (n=1, 4%), telerehabilitation-based exercise (n=1, 4%), a specialised long COVID rehabilitation programme (n=1, 4%), personalised health-behaviour support (n=1, 4%), cognitive–behavioural rehabilitation (n=1, 4%) and singing-based rehabilitation (n=1, 4%). Overall, rehabilitation interventions were heterogenous in setting, components, frequency and duration. Most studies delivered outpatient (clinic-based) programmes (n=20, 74%), with several hybrid/virtual or home-based telerehabilitation programmes (n=6, 22%), and one based out of an inpatient setting (4%). Programmes commonly combined aerobic and resistance training, breathing exercises or inspiratory muscle training, education, self-management, and psychological or counselling support. Many programmes were multidisciplinary, featuring physiotherapists, exercise physiologists, nurses and psychologists. Programme length ranged from 3 days to 12 weeks and additional intervention characteristics (length of follow-up, frequency, duration, timing of initiation) are summarised in table 5.

Broadly, the main outcomes evaluated among the primary studies included lung function (e.g. FEV₁, FVC, peak expiratory flow (PEF), diffusing capacity for carbon monoxide (D_LCO)), respiratory muscle strength and endurance (e.g. maximum inspiratory and expiratory pressures), exercise performance (e.g. peak oxygen consumption (V′_O2peak), incremental shuttle walk test (ISWT) score), functional exercise capacity (6MWT distance, 30 s STS test), HRQoL (e.g. COPD Assessment Test (CAT), EQ-5D) and symptoms (e.g. dyspnea-12 scale, mMRC, visual-analogue scales).

Across six studies reporting lung-function outcomes, significant within-group improvements in lung function measures were identified following rehabilitation [58, 60, 64–66, 86, 87]. For instance, del Corral et al. [64] reported within-group improvements in D_LCO and PEF following rehabilitation and Acharya et al. [58] reported improved FVC % pred, FEV₁ % pred and D_LCO% pred following rehabilitation. However, statistically significant between-group differences favouring the intervention group were only identified for PEF [64, 86] and FEV₁/FVC [87]. Of two studies directly measuring respiratory-muscle strength or endurance [60, 64], one RCT found statistically significant improvements in maximum inspiratory pressure, maximum expiratory pressure and inspiratory-muscle endurance in the intervention group, with significant between-group differences versus sham treatment [64]. Further, a nonrandomised study cross-sectionally observed lower inspiratory-muscle strength in participants with PCC compared to controls [60].

Three RCTs reported mixed results for changes in V′_O2peak following rehabilitation. One showed significant improvements compared with control groups [78], one identified within-group gains but no significant between-group differences [89] and one did not complete baseline testing, preventing any conclusions from being drawn [80]. Among the four nonrandomised studies reporting on V′_O2peak, three observed improvements [62, 65, 66], while one found no statistically significant change following rehabilitation [72]. Across all studies, magnitude of improvement in V′_O2peak ranged from 0.4 mL·kg⁻¹·min⁻¹ to 4.0 mL·kg⁻¹·min⁻¹. In addition, two RCTs and one nonrandomised controlled study assessed changes in ISWT distance following rehabilitation [63, 85, 86]. One RCT reported significant improvements compared to usual care for both face-to-face and remote delivery [63], while another RCT observed a small, nonsignificant increase [85]. The nonrandomised study reported a significant between-group difference favouring the intervention group [86]. Across all studies, ISWT improvements ranged from 21 m to 100.8 m.

Five nonrandomised studies assessed change in 6MWT distance following rehabilitation, with four studies reporting significant improvements ranging from 56.2 m to 405 m [59, 69, 72, 82], and one retrospective study reporting a higher 6MWT distance in participants who completed rehabilitation (compared to those who did not) [87]. Four nonrandomised studies examined changes in 30 s STS test score following rehabilitation, with all four identifying improvements following rehabilitation (range of improvements: 3.3 to 4.6 repetitions) [65, 66, 69, 82].

Seven studies reported on EQ-5D (i.e. EQ-5D-5L, EQ-5D index or EQ-5D visual analogue scale). Nonrandomised studies showed significant within-group improvements [58, 59, 67, 74], but included RCTs did not find significant between-group differences in EQ-5D scores [63, 64, 85]. In addition, two nonrandomised, uncontrolled studies showed statistically significant within-group improvements in CAT following rehabilitation [69, 75] and one RCT showed a significant between-group difference in CAT in favour of the intervention group [85].

One RCT [63] and four nonrandomised studies [65, 66, 68, 75] examined changes in Dyspnea-12 following rehabilitation interventions. While the RCT (exercise-based rehabilitation) found no between-group differences following the intervention [63], all four nonrandomised studies identified a significant within-group increase following rehabilitation (three exercise-based programmes and one singing-based programme) [65, 66, 68, 75]. In addition, six studies (n=4, RCT; n=2, non-randomised) assessed changes in mMRC dyspnoea following rehabilitation. Within RCTs, mixed results were identified with one study showing significant between-group differences in favour of the rehabilitation group [81], two studies showing nonsignificant between-group differences following rehabilitation [70, 73] and between-group differences not reported in one [61]. Both nonrandomised studies identified significant improvements in mMRC dyspnoea following rehabilitation [68, 69].

Secondary research findings

Secondary research findings are summarised in table 5. Multicomponent exercise rehabilitation was the most common type of rehabilitation studied among the included reviews (n=13, 65%), followed by telerehabilitation (n=2, 10%), various rehabilitation types (n=2, 10%), PR (n=1, 5%), combined respiratory muscle training and exercise rehabilitation (n=1, 5%), multicomponent psychotherapeutic interventions (n=1, 5%), respiratory muscle training (n=1, 5%), and exercise-based rehabilitation (n=1, 5%). Any reported intervention characteristics (i.e. length of follow-up, outcome measures, rehabilitation setting, components, frequency, duration and timing of initiation) are summarised in table 5. However, intervention characteristics were not explicitly reported in the majority of the included secondary studies.

10 reviews (50%) summarising multicomponent programmes and/or structured respiratory muscle training programmes were associated with improvements in dyspnoea, functional exercise capacity, respiratory muscle strength and HRQoL [43, 45, 46, 49, 50, 52, 54–56, 84]. Conversely, two reviews found that pulmonary function outcomes (i.e. FEV₁, FVC, FEV₁/FVC) were inconclusive [47, 53]. Soril et al. [52], Aiyegbusi et al. [83] and Schurr et al. [51] all noted the importance of starting rehabilitation early to prevent deterioration in patient condition and to promote recovery. In alignment with findings from research question one, engaging in a multidisciplinary approach was also recommended [44].

Six (30%) reviews highlighted methodological limitations across primary studies, such as small sample sizes, imprecision and inconsistent reporting of intervention dose/timing, which led to predominantly low to very low quality of evidence and recommended caution in interpretation of findings [6, 42, 44, 47, 48, 84]. Similarly, several of the included systematic reviews and meta-analyses exhibited a high risk of bias. For instance, Zheng et al. [57] reported improvements in 6MWT distance, FEV₁/FVC, quality of life and dyspnoea (n=1579, high risk of bias) particularly among RCTs or controlled clinical trials. Similarly, Torres et al. [54] reported improvements in 6MWT distance, FEV₁ % pred and FVC % pred (1886 participants, high risk of bias). In both reviews, there was evidence of low to moderate heterogeneity in these pooled estimates (I²=0–65%) [54, 57]. In two (10%) reviews, sensitivity analyses were executed to help manage heterogeneity between studies and it was found that significance of findings were preserved after removing outliers [46, 50].

Discussion

Through the present rapid review of the international literature, 73 unique guidance documents, secondary studies and primary studies for workup or management of pulmonary-related complications in PCC were identified. To our knowledge, this synthesis represents the most comprehensive and up-to-date review on this topic to date. The evidence base was strong in terms of hierarchy of evidence, with 24 secondary studies (including n=10 systematic reviews and meta-analyses, n=7 systematic reviews, n=1 living systematic review and n=1 systematic review and network meta-analysis) and 13 RCTs included for final analysis. However, study quality and risk of bias among included articles were heterogenous and, as such, the reported findings should be interpreted with caution. A summary of practical recommendations based on the findings of this review are provided in supplementary figure S2.

12 guidance documents from seven different countries provided recommendations on the assessment and workup of pulmonary-related PCC. Most were of high quality, though domains related to stakeholder involvement, rigour of development and applicability were commonly rated lower. Despite some variability in methodological approaches, there was consistent emphasis on the need for a broad, multi-system diagnostic evaluation. For instance, recommendations highlighted ruling out alternative causes of breathlessness [12, 21, 23, 25–27, 105], incorporating PFTs (including diffusing capacity) [12, 21–23, 25,26] and considering imaging (chest ray or CT) for persistent symptoms or suspected fibrosis [12, 13, 21, 23, 25, 26]. One recent publication recommended CPET for selected patients to help characterise exercise limitation and guide exercise prescription [26], which also aligns recently published practical guidelines [106].

With regard to pharmacological symptom management, there was limited evidence across primary studies, secondary studies and guidance documents to support the use of pharmacological interventions for treating pulmonary-related PCC. However, three guidance documents (AGREE II quality rating: high) recommended the use of inhaled or systemic steroidal medications for PCC patients with new obstructive lung disease, persistent cough, impaired lung function and/or interstitial abnormalities [20, 25, 105], which was further supported by Ntiamoah et al. [32] who identified improvements in lung function and imaging parameters in a group provided with systemic corticosteroids. In the systematic review by Yong [41], one completed RCT found that Treamid was not an effective treatment for pulmonary-related PCC and two additional ongoing registered RCTs were identified. As of November 2024, both ongoing trials have been terminated without results posted (NCT04695704, NCT04652518). Other recently published primary studies [29, 34] and systematic reviews [39, 40, 42] have not identified any pharmacological interventions with robust evidence supporting their efficacy and safety in the treatment and management of pulmonary-related PCC. Collectively, aside from corticosteroid use for symptom management, the current evidence base does not provide convincing support for the use of pharmacological treatments for pulmonary-related complications in PCC.

With regard to non-pharmacological interventions for pulmonary-related PCC management, findings for therapeutic approaches were mixed. Investigations of the impact of intermittent hypoxia-hyperoxia and intermittent hypoxia treatments showed significant improvements in exercise capacity, lung function and dyspnoea [28, 36], while ozone autohemotherapy and hyperbaric oxygen therapy did not demonstrate clear benefits [30, 31], and whole-body hyperthermia produced only short-term symptom relief [35]. One observational study from Serbia found self-reported improvements in respiratory symptoms and intervention adherence through pharmacist-delivered respiratory PCC patient care counselling [33]. This may present as a potential model to couple with effective pharmacological therapies (i.e. as a multi-component programme) once they emerge.

PR emerged as a widely endorsed intervention, with eight (67%) of guidance documents recommending its use for patients with ongoing dyspnoea, functional limitations or abnormal assessment results (e.g. radiographic abnormalities, oxygen desaturation with exercise) [13, 19, 20, 24–27, 105]. All eight emphasised individualisation of PR programmes, with many recommending structured baseline assessments and some highlighting the need for safety monitoring of adverse events during exercise. In addition, three guidance documents supported multidisciplinary approaches to care (e.g. specialist consults, rehabilitation teams, diverse outpatient PCC teams or networks) [20, 23, 24]. Such recommendations may speak to either the array of symptoms and needs of those presenting with pulmonary-related PCC or the evolving understanding of the condition itself. From included primary and secondary studies, there was evidence to support use of rehabilitation interventions for managing pulmonary-related complications in PCC, secondary to improvements in respiratory muscle strength and endurance [64], V′_O2peak [62, 65, 66, 78], ISWT distance [63, 86], dyspnoea [65, 66, 68, 75], and HRQoL [58, 59, 67, 69, 74, 75]. Similarly, 10 (50%) secondary studies supported improvements in dyspnoea, functional exercise capacity, respiratory muscle strength and HRQoL following rehabilitation programmes [43, 45, 46, 49, 50, 52, 54–56, 84]. However, primary and secondary study design features for rehabilitation interventions introduced risk of bias (e.g. more than 50% of primary studies lacked a control group), warranting caution in interpretation of findings. There was common emphasis towards multi-component exercise programmes as these were featured in 17 (63%) of primary studies investigating rehabilitation interventions. Unfortunately, limited and heterogenous information on the individual rehabilitation programme components or characteristics was reported among secondary studies. There was also limited evidence evaluating the effectiveness of individual rehabilitation components and it is not clear whether a comprehensive rehabilitation programme was effective when compared to exercise therapy (or any individual rehabilitation component) alone. Of note, a particular challenge in utilising pre-existing rehabilitation programmes for pulmonary-related PCC patients has been identified previously and remains relevant: referring PCC patients to PR programmes could significantly strain current programmes, leading to longer waitlists, requiring considerable resources for programme expansion and potentially limiting access for individuals with other respiratory conditions who rely on this established therapy for treatment (e.g. COPD patients) [52].

Limitations and future considerations

The rapid review methodology was selected due to its efficiency and the clinical importance of answering the research questions. However, there are a few limitations to our approach worth noting. Consistent with rapid review methodology, titles, abstracts and full texts were not screened in duplicate. This may introduce a greater risk of selection bias compared with duplicate screening. Potentially relevant studies may have been excluded from the final analysis in error, particularly those where relevance was less clear from the abstract alone. Additionally, because the review excluded non-English language publications, studies conducted in non-English-speaking settings may have been missed, potentially limiting the generalisability of our findings. While the number of potential studies not captured is likely small, given the broad range of outcomes reported in included studies (tables 3 and 5) and the predominance of English-language publication in this field, it remains possible that omitted studies could have contributed additional evidence, particularly from underrepresented regions or populations. In addition, because both primary and secondary sources were included, some study findings appeared in more than one review, creating the potential for over-representation of findings. In evidence syntheses that include meta-analyses, overlapping populations can artificially increase the apparent precision of effect estimates and produce misleading conclusions [107]. In this synthesis, however, no meta-analysis was conducted. To minimise duplication, we excluded individual primary studies already captured in the included systematic reviews; however, we could not control for the overlap across the reviews themselves. To maintain transparency, overlapping studies are identified in appendix 4.

Due to the rapid nature of this methodology, meta-analyses and assessment for publication bias were not performed. Given that numerous systematic reviews performed meta-analyses, pooled analyses (e.g. evidence gap maps) could be considered in future work. In addition, the reference lists of guidance documents were not screened for relevant records. Further, studies were also conducted across international settings, thus the generalisability of study findings to other specific contexts or jurisdictions is unclear. To this end, extracting information on implementation considerations (e.g. barriers and/or facilitators) for locally adopting the identified management strategies and programmes may help to clarify the applicability of study findings in any given setting. In addition, while the study populations for each of the included primary research studies met our inclusion criteria for PCC definition (i.e. probable or confirmed SARS-CoV-2 infection, 3 months from initial onset with symptoms that last for at least 2 months and cannot be explained by an alternative diagnosis), it could not always be clearly distinguished whether pulmonary complications were a continuation from acute infection, a result of worsening of pre-existing pulmonary conditions, or if they were new symptoms which developed within 3 months after the initial infection. At minimum, the included secondary studies had criteria specifying that the individual studies reported symptoms lasting at least 8 weeks from onset. Despite this, the possibility remains that the treatment needs for those with or without pre-existing pulmonary conditions may be different and, as a result of inconsistent reporting on pre-existing conditions in the source data, we cannot make a delineation in this review. Finally, as it was outside the scope of this review, we were not able to evaluate the impact of acute infection severity or vaccination status on outcomes in people with PCC. Given the wide variability in patient presentation within PCC, it would be valuable for future research to summarise the influence of these factors on patient outcomes, assessment, and management.

Conclusion

The objective of this rapid review was to identify, appraise and summarise the literature concerning the management and treatment of pulmonary-related complications in those with PCC. The rising prevalence of pulmonary-related PCC worldwide demonstrates not only the substantial and continued burden of SARS-CoV-2 infection, but also signals the pressing need for contemporary, evidence-informed guidance on timely assessment and effective treatment for those impacted. This rapid review identified numerous primary studies, secondary studies and guidance documents with relevance to the assessment, workup, and treatment of pulmonary-related complications in PCC. Common recommendations for assessment were identified (e.g. the use of PFTs with diffusing capacity for assessment) and emerging evidence suggests that multicomponent exercise rehabilitation programmes led by multidisciplinary teams may be of value. In accordance with the findings of previous secondary research and guidance documents, there remains insufficient evidence to recommend pharmacological treatment of pulmonary-related complications in PCC at current time, outside of possible monitored trials of corticosteroid use for symptom management. PR remains a cornerstone intervention; however, the optimal type and duration of programmes to improve pulmonary outcomes in patients with PCC remain unclear. Moreover, study design limitations in the identified literature introduce risk of bias, warranting caution in the interpretation of these findings. Overall, the breadth of new literature identified in the present review, coupled with the changing cultural landscape, including infrastructure and resources, to care for those with PCC suggest the need for an updated, evidence-based and “living” guideline focused on pulmonary-related complications in PCC.

Points for clinical practice

• A rapid review of evidence on the assessment, workup and/or management of pulmonary-related PCC complications was completed. We identified 73 unique articles, including 12 guidance documents, 24 secondary studies and 37 primary research studies (13 RCTs). Key recommendations for assessment were identified, such as the use of PFTs with diffusing capacity measurement, radiological imaging and cardiopulmonary exercise testing. Emerging evidence supports the potential benefits of multicomponent exercise rehabilitation programmes delivered by multidisciplinary teams.

• Aside from a possible trial of corticosteroid use for symptom management, there is insufficient evidence to recommend pharmacological treatment for pulmonary-related complications of PCC.

• The study quality and risk of bias among included records was heterogenous and, as such, the reported findings should be interpreted with some caution.

Supplementary material

Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.

Supplementary figure: PRISMA flowchart

ERR-0010-2025.SUPPLEMENT

Supplementary figure: Practical recommendations

ERR-0010-2025.SUPPLEMENT2

Supplementary appendices

ERR-0010-2025.SUPPLEMENT3