Navigating the efficacy, safety and outcome of intranasal platelet-rich plasma for persistent olfactory loss: a systematic review and meta-analysis

Abstract

Background

Persistent olfactory dysfunction (OD) poses a significant clinical challenge with limited therapeutic options. Intranasal platelet-rich plasma (PRP), an autologous source of regenerative growth factors, presents a biologically plausible intervention, yet its efficacy across various OD etiologies remains synthetically underexplored.

Methods

This systematic review and meta-analysis adhered to PRISMA guidelines. A comprehensive search of PubMed/MEDLINE, Web of Science, and Cochrane Library (2010–2025) identified studies assessing intranasal PRP for persistent OD (> 6 months). Random-effects models pooled standardized mean differences (SMD) for objective (e.g., Sniffin’ Sticks TDI) and subjective outcomes.

Results

Seven studies (n = 473) were included. Meta-analysis of controlled studies (n = 314) demonstrated a moderate, significant improvement in objective olfactory scores favoring PRP over control (SMD 0.61, 95% CI 0.38–0.84). The pooled mean TDI change from baseline was 7.73 points (95% CI 6.59–8.87), exceeding the minimal clinically important difference of 5.5 points. Benefits were observed across post-COVID-19, post-viral, and post-traumatic etiologies. Adverse events were predominantly mild and transient (e.g., epistaxis).

Conclusion

Intranasal PRP appears to be a generally well-tolerated and promising therapeutic strategy, yielding clinically meaningful improvements in persistent OD of diverse origins. However, the evidence remains preliminary and should be interpreted cautiously, as it is based on only two small RCTs plus non-randomized or single-arm studies. Future studies must optimize its protocol and evaluate its potential synergy with standard therapies like olfactory training.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s00405-026-10467-7.

Introduction

Olfactory dysfunction (OD) affects a substantial proportion of the general population. Quantitative olfactory deficits are present in approximately 20% of adults, with complete anosmia occurring in nearly 5% [1]. Acquired OD has a multifactorial etiology, most commonly arising from chronic sinonasal disease, but also from post-infectious causes (e.g., following common respiratory viral illnesses), head trauma, and idiopathic olfactory loss [1]. Upper respiratory tract infections account for approximately 14% of OD cases, while head injury contributes to nearly 6% [1]. In the context of severe traumatic brain injury, anosmia has been reported in up to 25–30% of affected patients [2]. After exclusion of identifiable causes, nearly 8% of olfactory losses are classified as idiopathic; importantly, isolated idiopathic anosmia has been recognized as a potential prodromal marker of neurodegenerative disorders, such as Parkinson’s disease [1]. Regardless of etiology, persistent OD can profoundly impair quality of life due to diminished flavor perception, increased safety risks, and social isolation, and it may reflect underlying central nervous system alterations, including reduced olfactory bulb volume and disrupted cortical processing.

Persistent OD reflects injury to the olfactory system and impaired regenerative capacity. Viral infections and environmental toxins can damage the olfactory epithelium, including olfactory receptor neurons and supporting cells, thereby inducing local inflammation and, in some cases, invading central olfactory pathways [3]. Similarly, head trauma may result in shearing of olfactory axons at the cribriform plate or direct injury to the olfactory bulbs or associated cortical regions [2]. Such insults, compounded by chronic neuroinflammation, may overwhelm the normally robust regenerative potential of the olfactory epithelium, ultimately leading to long-term anosmia or hyposmia. In idiopathic cases, subtle central nervous system alterations or early neurodegenerative processes may underlie the observed dysfunction [1]. In sum, persistent OD arises from a complex interplay of epithelial injury, inflammatory processes, and altered neural processing that collectively impede recovery.

Platelet-rich plasma (PRP) represents an autologous regenerative therapy with a compelling biological rationale for olfactory repair. PRP is derived from the patient’s own blood and is processed to concentrate platelets along with their associated growth factors and cytokines [4]. It contains multiple neurotrophic and angiogenic mediators, including transforming growth factor–β, vascular endothelial growth factor, nerve growth factor, and insulin-like growth factor-1, which are known to support cell survival, proliferation, and axonal regeneration [4]. Experimental evidence from in vitro and animal studies suggests that the olfactory neuroepithelium can regenerate in the presence of such factors, and PRP has demonstrated encouraging results in improving anosmia in preclinical models [4]. Clinically, PRP has been widely applied in otolaryngology to enhance wound healing, including in sinonasal and tympanic procedures [4]. Owing to its autologous nature, PRP is generally well tolerated and avoids immunogenic reactions, rendering it an attractive therapeutic option for promoting olfactory neuron repair.

Intranasal administration of PRP enables direct targeting of the olfactory cleft and mucosa. Endoscopic injection of PRP into the superior nasal cavity allows precise delivery of growth factor–rich plasma to the damaged olfactory epithelium [5]. This approach ensures interaction with the olfactory neuroepithelium located within the olfactory cleft, rather than dispersion into the lower nasal passages [5]. Consequently, intranasal PRP provides localized therapeutic exposure with minimal systemic involvement. Recent pilot studies have demonstrated the feasibility and safety of endoscopic PRP injections into each olfactory cleft, with several patients reporting both subjective and objective improvements in olfactory function [4, 5].

Despite these promising findings, the current evidence base remains limited and is largely focused on post–COVID-19–related olfactory dysfunction. To our knowledge, existing systematic reviews have primarily evaluated PRP in the context of COVID-19–associated anosmia [4], while no comprehensive synthesis has examined the efficacy of intranasal PRP across all etiologies of persistent OD. Therefore, the aim of the present study was to systematically evaluate the efficacy and safety of intranasal PRP injections for chronic olfactory dysfunction, encompassing post-viral, post-traumatic, and idiopathic causes.

Methodology

Literature search and study selection

This systematic review and meta-analysis were conducted according to a predefined protocol registered with PROSPERO (CRD420251266058) and adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [6, 7]. A comprehensive literature search was performed across PubMed/MEDLINE, Web of Science, and the Cochrane Library from 2010 inception to 2025. The search strategy employed a combination of Medical Subject Headings (MeSH) and Emtree terms, including “Olfaction Disorders,” “Anosmia,” and “Platelet-Rich Plasma,” combined with free-text terms such as (“smell loss” OR hyposmia OR parosmia) AND (“Platelet-Rich Plasma” OR PRP). All identified records were uploaded to Rayyan, a web-based systematic review platform, for collaborative screening and deduplication [8].

Inclusion criteria required studies of adult patients (≥ 18 years) with persistent quantitative olfactory dysfunction (post-viral, post-traumatic, or idiopathic) lasting more than 6 months, who underwent intranasal injection of autologous PRP. Studies must have reported quantifiable pre- and post-treatment outcomes. Objective olfactory outcomes were defined as validated psychophysical olfactory tests, including Sniffin’ Sticks TDI, UPSIT, or CCCRC. Subjective patient-reported outcomes, including VAS, ODQ, or Likert-scale smell ratings, were extracted and analyzed separately as secondary outcomes and were not considered equivalent to objective psychophysical testing. Studies relying solely on subjective VAS-based olfactory outcomes were not included in the objective olfactory meta-analysis. Case series with fewer than 10 patients, review articles, non-human studies, and publications without extractable data were excluded.

Screening and data extraction

The screening process was conducted in three phases to ensure methodological rigor. Initially, 88 records were identified through database searching. After removing 35 duplicates, 53 unique records underwent title and abstract screening. This phase excluded 38 irrelevant studies, leaving 16 articles for full-text assessment. During the eligibility phase, these 16 full-text articles were evaluated against the predefined inclusion criteria. Nine studies were excluded, with six being irrelevant to the research question and three excluded due to inaccessible full texts despite attempts to locate them. Ultimately, seven studies met all inclusion criteria and were included in the systematic review [9–15]. Figure 1 presents the PRISMA flow diagram detailing the complete screening process.

Fig. 1.

PRISMA flow diagram summarizing the screening process

A standardized data extraction form was developed in Microsoft Excel to document study characteristics (first author, publication year, country, study design), patient demographics (age, gender, etiology, and duration of olfactory loss), intervention details (PRP preparation protocol, injection volume, number of sessions), and outcomes. Primary outcomes were recorded as mean and standard deviation for objective psychophysical olfactory test scores (TDI, UPSIT) at pre-treatment and post-treatment time points. Secondary outcomes included subjective patient-reported scores (VAS), the proportion of patients achieving clinically significant improvement, and the incidence of adverse events. Two reviewers cross-verified all extracted data, with discrepancies resolved through consensus discussions.

Quality assessment and risk of bias

The methodological quality and risk of bias of included studies were assessed independently by two reviewers using validated assessment tools appropriate to study design. For randomized controlled trials, the Cochrane Risk of Bias 2 (RoB 2) tool was used, evaluating the following domains: (1) randomization process, (2) deviations from intended interventions, (3) missing outcome data, (4) measurement of the outcome, and (5) selection of the reported result [16]. For non-randomized studies, including cohort studies and controlled trials without randomization, the Newcastle-Ottawa Scale (NOS) was applied, assessing selection of study groups, comparability of groups, and ascertainment of exposure and outcomes [17]. Studies were categorized as having low, moderate (some concerns), or high risk of bias. Disagreements between reviewers were resolved through discussion.

Among RCTs (Table S5), Yan et al. [9] had some concerns due to unclear allocation concealment and differential dropout (22% vs. 0%), while Evman and Cetin [10] had a high risk of bias due to an open-label design without a sham control or trial registration [9, 10]. For non-randomized studies (Table S6), Steffens et al. [11] demonstrated good quality (9/9 stars, low risk), while Lechien et al. [13] and Lechien [15] showed fair quality (7/9 stars, moderate risk) with limitations in control group selection [11, 14, 15]. Both case series (Table S7) scored 9/10 on the Joanna Briggs Institute (JBI) checklist, indicating low risk of bias with clear inclusion criteria, validated outcome measures, and consecutive enrollment [12, 13].

Statistical analysis

All statistical analyses were performed using R version 4.3.x with the meta package (version 7.0–0) for meta-analysis, the metafor package for additional analyses, and readxl for data import. Forest plots were generated using the meta package and exported using the ggpubr and gridExtra packages. For controlled studies comparing PRP intervention to control groups, the standardized mean difference (SMD) using Hedges’ g correction was calculated for continuous outcomes. The SMD was computed as the difference in mean change from baseline between the PRP and control groups, divided by the pooled baseline standard deviation. This approach was chosen because studies reported change scores rather than post-treatment scores consistently, baseline standard deviations were available for all studies, and standardization allows comparison across studies using different outcome scales. For dichotomous outcomes such as clinical improvement, risk ratios (RR) with 95% confidence intervals were calculated using the Mantel-Haenszel method. For single-arm analyses pooling all PRP-treated patients, the raw mean change from baseline was calculated using the metamean function. For proportions such as clinical improvement rate, the logit transformation was applied using metaprop with back-transformation for interpretation.

Statistical heterogeneity was assessed using Cochran’s Q statistic with corresponding p-value, the I² statistic interpreted as low (0–40%), moderate (41–60%), substantial (61–75%), or considerable (> 75%), and τ² (tau-squared) as the between-study variance estimate. A random-effects model was employed for all analyses using the restricted maximum likelihood (REML) estimator for τ², given the expected and heterogeneity across studies.

Pre-specified subgroup analyses were conducted to explore potential sources of heterogeneity, including study design (RCT vs. Non-RCT vs. Single-arm), etiology (Post-COVID-19 vs. Post-viral non-COVID vs. Post-traumatic), follow-up duration (≤ 2 months vs. > 2 months), and number of PRP sessions (single vs. multiple). Subgroup analyses were performed using the update function with subgroup specification, and results were combined using metabind for visual presentation. Tests for subgroup differences were reported with corresponding p-values.

Sensitivity analyses were conducted using leave-one-out analysis, which involves sequentially omitting each study to assess the influence of individual studies on the pooled estimate, and influence diagnostics to identify potentially influential studies. Publication bias was not assessed since the number of studies was < 10. A two-sided p-value less than 0.05 was considered statistically significant for all analyses, and 95% confidence intervals were reported for all effect estimates. The feasibility of planned quantitative syntheses was assessed according to outcome availability and comparability across studies, as summarized in Tables S1 and S2.

Result

Characteristics of included studies

Seven studies met the inclusion criteria (Table 1), comprising a total of 473 participants across multiple countries. Two studies employed randomized controlled designs (Yan et al., [9]; Evman & Cetin, [10]), while five used non-randomized controlled or single-arm prospective designs (Steffens et al., [11]; Lechien et al., [12–14]; Lechien, [15]) [9–15]. The majority of studies (n = 5) focused on post-COVID-19 olfactory dysfunction (Yan et al., [9]; Evman & Cetin, [10]; Steffens et al., [11]; Lechien et al., [12, 13], while one examined post-viral non-COVID olfactory loss (Lechien et al., [14]) and one addressed post-traumatic olfactory dysfunction (Lechien, [15]) [9–15]. Sample sizes ranged from 25 to 159 participants, with mean ages between 31.8 and 58.9 years and a predominantly female population (50–77%). Duration of olfactory dysfunction varied from 8.9 months to over 8 years. Most studies administered a single PRP injection session, except Yan et al. [9], who used three sessions at two-week intervals [9]. Follow-up periods ranged from 1 to 6 months.

Table 1.

Characteristics of included studies

Study	Country/Setting	Study design	N	Age (years)	Gender	Etiology	OD duration	PRP sessions	Outcome measures	Follow-up
Yan et al., [9]	USA (Multi-center)	RCT, single-blinded, placebo-controlled	30 (17 PRP, 12 placebo analyzed)	44.1 ± 14.0 years	50% Female	Post-COVID-19	8.9 ± 2.2 months	3 sessions (2 weeks apart)	Sniffin’ Sticks TDI, UPSIT, VAS	3 months
Evman & Cetin, [10]	Turkey (Single-center)	RCT	25 (12 PRP, 13 control)	PRP: 31.8 ± 6.9; Control: 33.5 ± 11.1 years	52% Female	Post-COVID-19	> 12 months	1 session	CCCRC (STC, SIC)	1 month
Steffens et al., [11]	Belgium (Single-center)	Prospective non-randomized controlled	56 (30 PRP, 26 control)	PRP: 39 ± 12; Control: 44 ± 11 years	64.3% Female	Post-COVID-19	10.8 ± 2.5 months	1 session	Sniffin’ Sticks TDI, Likert scale	1 month
Lechien et al., [12]	Belgium (Multi-center)	Prospective single-arm	87 (37 completed follow-up)	41.6 ± 14.6 years	71.3% Female	Post-COVID-19	15.7 ± 7.5 months	1 session	Sniffin’ Sticks TDI, ODQ	2 months
Lechien et al., [13]	Belgium, Italy, France (Multi-center)	Controlled (non-randomized)	159 (81 PRP, 78 control)	45.2 ± 12.7 years	71% Female	Post-COVID-19	PRP: 15.7 ± 7.5; Control: 11.0 ± 3.0 months	1 session	Sniffin’ Sticks TDI, ODQ	10 weeks
Lechien et al., [14]	Belgium (Single-center)	Prospective case-series	43 (35 completed follow-up)	58.9 ± 16.8 years	56% Female	Post-viral (non-COVID)	104.7 ± 67.2 months	1 session	Sniffin’ Sticks TDI, ODQ	6 months
Lechien, [15]	Belgium (Single-center)	Preliminary controlled (non-randomized)	73 (33 PRP, 40 control)	46.5 ± 11.1 years	57.6% Female	Post-traumatic	55.6 ± 45.1 months	1 session	Sniffin’ Sticks TDI, ODQ	3 months

CCCRC Connecticut Chemosensory Clinical Research Centre test, COVID-19 coronavirus disease 2019, N number of participants, OD olfactory dysfunction, ODQ Olfactory Disorder Questionnaire, PRP platelet-rich plasma, RCT randomized controlled trial, SIC Smell Identification Component, STC Smell Threshold Component, TDI Threshold-Discrimination-Identification score, UPSIT University of Pennsylvania Smell Identification Test, VAS Visual Analog Scale

All seven studies demonstrated significant improvements in objective olfactory function following PRP treatment. Among controlled studies, Yan et al. [9] reported a significantly greater TDI improvement in the PRP group compared to placebo (+6.25 vs. +2.58 points; p =0.047), with 57.1% of PRP-treated patients achieving clinically meaningful improvement(≥5.5 TDI increase) compared to only 8.3% in the placebo group [9]. Similarly, Evman and Cetin [10] found significant between-group differences in both smell threshold (p = 0.037)and smell identification (p < 0.001) components [10]. Steffens et al. [11] reported a mean TDI improvement of +6.7 points in the PRP group versus no significant change in controls (p< 0.001) [11]. The largest controlled study by Lechien et al. [13] demonstrated a TDI improvement of +8.7 points in the PRP group compared to +3.9 points in controls (p =0.011), with 80.3% of PRP-treated patients reporting subjective improvement [13]. Single arm studies also showed favorable results, with Lechien et al. [12] and Lechien et al. [14] reporting TDI improvements of +5.7 and +9.8 points, respectively [12, 14]. Notably, Lechien [15] demonstrated efficacy in post-traumatic olfactory dysfunction, with PRP patients showing greater improvement than controls (+8.1 vs. +4.0 points; p = 0.038) [15]. Regarding safety, adverse events were generally mild and transient, with epistaxis being the most reported complication (35.6–46.5% in studies reporting this outcome). VAS outcomes were considered subjective patient-reported measures and were not pooled with objective psychophysical olfactory outcomes. (Table 2).

Table 2.

Olfactory outcomes

Study	Time point	Objective test	Baseline score	Post-treatment score	Change (p-value)	Between-group p	Subjective outcomes	Clinical improvement rate	Adverse events
Yan et al., [9]	3 months	Sniffin’ Sticks TDI	PRP: 24.3 ± 6.4; Placebo: 26.0 ± 4.4	PRP: 30.6*; Placebo: 28.6*	PRP: +6.25 (p < 0.0001); Placebo: +2.58 (p = 0.061)	0.047	VAS (0–10): PRP + 2.13 (p < 0.0001); Placebo + 1.25 (p = 0.014); Between-group p = 0.167	TDI ≥ 5.5 increase: PRP 8/14 (57.1%); Placebo 1/12 (8.3%)	Short-term nasal congestion and pressure (up to 24 h) in both arms; 1 self-resolved photophobia (placebo); No long-standing adverse effects
Evman & Cetin, [10]	1 month	CCCRC (STC, SIC)	PRP: STC 5.63 ± 0.68, SIC 11.42 ± 1.17; Control: STC 5.69 ± 0.66, SIC 11.20 ± 1.12	PRP: STC 6.46 ± 0.45, SIC 15.17 ± 0.39; Control: STC 5.77 ± 0.70, SIC 11.85 ± 1.57	PRP: STC + 0.83 (p < 0.001), SIC + 3.75 (p < 0.001); Control: STC + 0.08 (p = 0.165), SIC + 0.65 (p = 0.089)	STC: 0.037; SIC: <0.001	Not reported	Not reported	No adverse effects or procedure-induced morbidity reported
Steffens et al., [11]	1 month	Sniffin’ Sticks TDI	PRP: 21.3 ± 7.4; Control: 24.5 ± 7.4	PRP: 28.0 ± 5.0; Control: 25.0 ± 7.7	PRP: +6.7 (p < 0.001); Control: NS	< 0.001	Likert scale (0–3): PRP 1.8 ± 1.0; Control 0.3 ± 0.6; p < 0.001	Not explicitly reported	No adverse effects reported
Lechien et al., [12]	2 months	Sniffin’ Sticks TDI	20.3 ± 10.5	26.0 ± 11.2	+ 5.7 (p = 0.009)	single-arm	ODQ (0–87): Pre 51.0 ± 18.0 → Post 40.7 ± 10.9; Change − 10.3 (p = 0.001)	29/37 (78.4%) reported subjective improvement (20 substantial improvement of anosmia/hyposmia, 9 improvement of parosmia)	54/87 (62.1%) experienced ≥ 1 event: Transient epistaxis 35.6%, Parosmia during anesthesia 11.5%, Vasovagal episode 4.6%, Panic attack 2.3%, Postnasal drip 5.7%, Nausea 2.3%; 66.6% judged injection moderately/severely painful
Lechien et al., [13]	10 weeks	Sniffin’ Sticks TDI	PRP: 19.8 ± 9.5; Control: 21.5 ± 8.4	PRP: 28.5 ± 9.1; Control: 25.4 ± 7.7	PRP: +8.7 (p = 0.001); Control: +3.9 (p = 0.001)	0.011	ODQ (0–87): PRP − 12.7 (p = 0.001); Control + 0.9 (NS); Between-group p = 0.020	PRP: 65/81 (80.3%) reported subjective smell improvement	Not reported for this cohort
Lechien et al., [14]	6 months	Sniffin’ Sticks TDI	10.3 ± 10.2	20.1 ± 12.1	+ 9.8 (p = 0.001)	single-arm	ODQ (0–87): Pre 29.8 ± 13.0 → Post 23.4 ± 11.3; Change − 6.4 (p = 0.013)	71.4% self-reported significant improvement (ODQ); 87.5% achieved significant TDI increase (MCID)	22/43 (51.2%) experienced ≥ 1 event: Transient nasal bleeding 46.5% (within 10 min); Transient postnasal drip 4.7% (within 3 days)
Lechien, [15]	3 months	Sniffin’ Sticks TDI	PRP: 8.4 ± 8.1; Control: 9.3 ± 5.0	PRP: 16.5 ± 9.6; Control: 13.3 ± 8.5	PRP: +8.1 (p = 0.003); Control: +4.0 (p = 0.001)	0.038	ODQ (0–87): PRP − 3.5 (NS); Control + 0.2 (NS)	Binary subjective improvement: PRP 22/33 (66.7%); Control 15/40 (30.5%)	Not reported

Values are presented as mean ± standard deviation unless otherwise specified. * Post-treatment values calculated from baseline + change. “—” indicates no control group (single-arm study).

CCCRC Connecticut Chemosensory Clinical Research Center test, MCID minimal clinically important difference, NS not significant, ODQ Olfactory Disorder Questionnaire (lower scores indicate better function), PRP platelet-rich plasma, SIC Smell Identification Component, STC Smell Threshold Component, TDI Threshold-Discrimination-Identification score (Sniffin’ Sticks), VAS Visual Analog Scale

Objective olfactory function (TDI scores)

The pooled standardized mean difference (SMD) comparing PRP to control groups across four studies, including 314 patients (158 PRP, 156 control), yielded an SMD of 0.61 (95% CI: 0.38–0.84, P < 0.001), indicating a moderate effect favoring PRP (Fig. 2A) [9, 11, 13, 15]. The prediction interval (0.24–0.98) suggests that future studies are likely to show positive effects. Heterogeneity was negligible (I² = 0.0%; P = 0.84) [9, 11–15].

Fig. 2.

Forest plots of TDI score changes following PRP treatment for olfactory dysfunction: (A) PRP versus control (SMD); (B) Mean change from baseline

The pooled mean TDI change from baseline across six studies comprising 230 patients treated with PRP was 7.73 points (95% CI: 6.59–8.87), exceeding the minimum clinically important difference of 5.5 points (Fig. 2B) [9, 11–15]. The prediction interval (6.24–9.22) indicates consistent benefit across settings. Heterogeneity remained low (I² = 1.3%; P = 0.41). The robustness of the primary TDI double-arm meta-analysis was further evaluated using leave-one-out sensitivity analysis, with results summarized in Table S3.

Subjective olfunction function (ODQ scores)

The pooled standardized mean difference (SMD) comparing PRP to control groups across two studies, including 232 patients (114 PRP, 118 control), yielded an SMD of − 0.62 (95% CI: −1.13 to − 0.11, P = 0.02) (Fig. 3A) [12, 15]. Substantial heterogeneity was observed (τ² = 0.0949; I² = 69.4%; P = 0.07). The pooled mean ODQ change from baseline across four studies comprising 186 patients treated with PRP was − 8.24 points (95% CI: −12.36 to − 4.12), indicating symptom improvement (Fig. 3B) [12–15]. Heterogeneity was substantial (τ² = 11.86; I² = 68.8%; P = 0.02).

Fig. 3.

Forest plots of ODQ score changes following PRP treatment for olfactory dysfunction: (A) PRP versus control (SMD); (B) Mean change from baseline

Clinical improvement and responder analysis

The pooled risk ratio comparing PRP to control groups across two studies, including 99 patients, was not statistically significant (47 PRP, 52 control), yielding an RR of 2.49 (95% CI: 0.79–7.82, P = 0.118, (Fig. 4A) [9, 15]. Heterogeneity was moderate (τ² = 0.3976; I² = 43.6%; P = 0.18). The pooled proportion of patients achieving clinically significant improvement (≥ 5.5-point TDI increase) across four studies comprising 165 patients was 74.92% (95% CI: 65.78–82.28, Fig. 4B) [9, 12–13, 15]. Heterogeneity was moderate (τ² = 0.0107; I² = 38.1%; P = 0.18).

Fig. 4.

Forest plots of clinical improvement following PRP treatment for olfactory dysfunction: (A) PRP versus control (risk ratio); (B) Pooled proportion

Subgroup and sensitivity analyses

Subgroup analyses were performed according to the pre-specified subgroup analysis plan shown in Table S4. Overall heterogeneity was negligible (τ²=0; I²=0%; P = 1.00), and no subgroup variable significantly modified the treatment effect (Fig. 5).

Fig. 5.

Subgroup analysis of TDI score changes (SMD) comparing PRP versus control by study design, etiology, follow-up duration, and PRP sessions

Discussion

Current clinical data suggest that intranasal PRP can produce modest but clinically meaningful improvements in persistent olfactory dysfunction. In a randomized trial of COVID-19–related anosmia, Yan et al. [9] found that PRP injections led to a 3.67-point greater increase in the composite TDI (threshold–discrimination–identification) score than placebo at 6 weeks [9]. This corresponded to a higher “responder” rate (57% vs. 8%) with PRP. A recent meta-analysis by Albazee et al. [18] pooled four RCTs (n = 198) and similarly reported significant gains in both objective and subjective olfactory scores with PRP versus control [18]. Long-term data, though limited, also favor PRP. Fieux et al. [19] followed patients up to one year and showed that those receiving PRP had significantly greater improvements in UPSIT scores and higher rates of minimal clinically important difference than untreated controls [19]. Importantly, encouraging results have been seen across etiologies. In chronic posttraumatic anosmia (mean duration ~ 56 months), a preliminary cohort study found that one olfactory-cleft PRP injection yielded subjective improvement in 67% of patients and increased mean TDI from 8.4 to 16.5 after 3 months – a gain significantly larger than that seen with olfactory training alone [15]. Further supporting the potential of regenerative approaches for traumatic etiologies, a recent study by Jiang & Chiang [20] evaluated intranasal injection of PRP combined with hyaluronic acid in 28 patients with post-traumatic OD. They reported subjective improvement in 71.4% of patients at 1 month and 85.7% at 3 months, with objective improvement in phenyl ethyl alcohol (PEA) odor detection thresholds observed in 78.6% and 53.6% of patients at the same time points, respectively. The treatment was well-tolerated, with transient epistaxis being the most common adverse event [20]. Even qualitative smell disorders responded: Abo El Naga et al. [21] conducted a pilot RCT in 60 patients with persistent post-COVID parosmia refractory to conventional therapies, and three PRP injections significantly reduced parosmia severity (VAS) versus continuation of standard care [21]. Taken together, these trials and case series consistently show that PRP accelerates olfactory recovery in cases that might otherwise plateau. The magnitude of improvement (generally on the order of + 3–7 points on TDI or equivalent test scales) exceeds the change typically achieved by training or spontaneous recovery alone.

Importantly, PRP appears generally well tolerated in the available studies. No serious adverse events have been reported in any trial. In Yan et al.’s RCT, no complications occurred in either the PRP or placebo groups [9]. Across studies, only minor, transient nasal symptoms were observed – e.g., brief congestion or pressure during/after injection (often also seen with sham) [18]. Duffy et al. [22] reported that topical PRP was well tolerated with no treatment-related harms over 12 months [22]. Review authors emphasize that autologous PRP eliminates risks of immune reaction or infection, and indeed, no systemic side effects have been noted [4].

For context, olfactory training remains the first-line therapy for non-sinonasal anosmia, as endorsed by guidelines [4]. Training is low-risk and can yield gradual recovery in many post-infectious or post-traumatic cases. However, recovery rates with training alone are often limited (e.g., only ~ 10–30% reach normal smell after months), especially when loss is chronic [15]. Pharmacological options have shown only modest or inconsistent benefit. Corticosteroids (oral or intranasal) are indicated mainly for inflammatory causes (e.g., chronic sinusitis), and have not reliably improved isolated post-viral smell loss [4]. Nutraceuticals have likewise failed to move the needle: a controlled trial of high-dose omega-3 did not improve COVID-related anosmia over placebo [23]. By contrast, a small pilot found that adding a short course of oral vitamin A to olfactory training significantly boosted early recovery compared to training alone [24]. Even so, vitamin A protocols are still experimental. Compared with these approaches, PRP’s effect appears at least comparable to that of olfactory training and superior to placebo or supplements. In head-trauma patients, PRP produced greater psychophysical gains than training [15]. And unlike systemic steroids, PRP is localized and free of systemic risks; unlike vitamins or omega-3, it delivers potent growth factors directly to the olfactory mucosa. Importantly, no study has yet rigorously compared PRP head-to-head with training or drugs. It is conceivable that PRP might be used as an adjunct to, rather than a replacement for, conventional therapy.

The proposed mechanism of PRP in olfaction is biologically plausible. Platelet concentrates contain very high levels of growth factors and cytokines that promote tissue repair and modulate inflammation [4, 21]. Key factors include TGF-β, EGF, VEGF, PDGF, NGF, and IGF, all released from platelet α-granules. These mediators stimulate angiogenesis, cell proliferation and differentiation, and inhibit apoptosis, creating a microenvironment conducive to neuroregeneration [4]. In vitro and animal studies support this: in a mouse anosmia model, topical PRP markedly preserved olfactory epithelial thickness and histology, leading to faster recovery of odor-driven behavior compared to saline [25]. Animal experiments also show that growth factor treatments (and stem-cell grafts) can regenerate damaged olfactory epithelium, suggesting the olfactory mucosa has regenerative capacity if given the right signals [4]. Platelets also carry neurotrophins (e.g., BDNF) and anti-inflammatory cytokines, which may protect surviving olfactory neurons and reduce chronic mucosal inflammation [26]. Biomedicine review authors posit that PRP could mitigate the local injury (viral or immune-mediated) and actively stimulate basal progenitor cells to differentiate into new receptor neurons [4, 26]. Thus, PRP may act by both quelling inflammation in the olfactory cleft and directly fostering neuronal regrowth. Nevertheless, the exact pathways remain to be fully defined. It is not known how long PRP’s effects persist or whether repeated doses yield incremental benefit. Also unclear is whether PRP principally augments mucosal healing, enhances axonal reconnection at the bulb, or both. Further in vitro studies on human olfactory epithelial cells (for example, measuring PRP’s effect on neurite outgrowth and cytokine profiles) would clarify these mechanisms.

Limitations

The evidence base remains limited by small sample sizes, heterogeneous PRP protocols (dose, preparation, number of injections), short follow-up in most studies, and variable outcome measures (different psychophysical tests and subjective scales). Few trials were fully randomized or blinded, and patient populations (post-COVID, post-viral non-COVID, post-traumatic) were mixed without consistent subgroup reporting, which reduces certainty and external validity. A major additional limitation is that only two randomized controlled trials were available, while the remaining studies were non-randomized controlled or single-arm designs. Moreover, several of these non-randomized studies originated from overlapping centers or research groups, which may limit the independence of the evidence and reduce generalizability. Another limitation is that the literature search was restricted to PubMed/MEDLINE, Web of Science, and the Cochrane Library, while other major databases such as Embase and Scopus were not searched; this may have reduced the comprehensiveness of the search and could have resulted in missing relevant studies indexed exclusively in those databases. Therefore, the pooled findings should be interpreted cautiously and considered preliminary rather than definitive until larger, independent, standardized, long-term randomized trials are available.

Conclusion

Intranasal platelet-rich plasma appears to be a generally well-tolerated and promising regenerative therapy for persistent olfactory dysfunction, providing clinically meaningful improvements in objective and subjective olfactory outcomes across different etiologies. However, these findings should be interpreted cautiously because the evidence is still preliminary, includes only two small randomized controlled trials, and is supported by non-randomized or single-arm studies. Future research should focus on clarifying optimal dosing strategies, treatment timing, and patient selection, as well as determining whether PRP offers an additive benefit when combined with established interventions such as olfactory training.

Supplementary Information

Below is the link to the electronic supplementary material.

Funding

This work was supported and funded by the Deanship of Scientific Research at Imam Mohammad Ibn Saud Islamic University (IMSIU) (grant number IMSIU- DDRSP2601).

Data availability

Data are available from the corresponding author upon reasonable request.

Declarations

Ethics approval

Not required for this systematic review.

Consent for Publication

Not applicable.

Conflicts of Interest

The authors have no competing interests to declare that are relevant to the content of this article.

Protocol Registration

The review protocol was registered in PROSPERO under the registration number: CRD420251266058.