Effect of water- and land-based exercise on lung function in children with post-COVID-19 condition: secondary results from a randomised controlled trial

Abstract

Objectives

This study aimed to assess the effect of water- and land-based exercise on lung function in children with post-COVID-19 condition.

Methods

This was a randomised controlled trial with multigroup pre-–post-test design. Children with post-COVID-19 condition aged 10–12 years were randomly assigned to water-based exercise (AQUA), land-based exercise (LAND) or a control group with no exercise (CONTROL). The outcomes were the changes in lung function (Lungtest Handy) from baseline to immediately after the 8-week intervention. Parameters measured included forced expiratory volume in 1 s (FEV₁), forced vital capacity (FVC) and vital capacity (VC).

Results

After the intervention, there was a significant difference (p<0.01) between the AQUA and LAND groups compared to the CONTROL group for FEV₁ values (F(2,64)=6.80; p=0.91; p<0.01, η2=0.18) and significant differences between the CONTROL and AQUA groups (p<0.01) and the CONTROL and LAND groups (p<0.05) for FEV₁ (F(2,64)=6.96; p=0.91; p<0.01, η²=0.18). Repetitions–groups interactions for FEV₁/%FVC (F(2,64)=0.71, p=0.162, p>0.05, η²=0.030) showed that the changes that occurred varied from group to group and the reason for this was an upward trend in the LAND and AQUA groups and a downward trend in the CONTROL group.

Conclusion

The study found that a supervised twice weekly 8-week exercise training programme in water and on land improved lung function in children with post-COVID-19 condition.

Shareable abstract

This study found that a supervised, twice-weekly, 8-week exercise training programme in water and on land improved lung function in children with post-COVID-19 condition https://bit.ly/3ZZp2oi

Introduction

Nearly 800 million confirmed cases of COVID-19 across the world had been reported to the World Health Organization (WHO), including over 7 million deaths. The WHO Delphi consensus on the definition of post-COVID-19 condition states that it occurs in individuals with a history of probable or confirmed severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, usually 3 months from disease onset, with symptoms that last for at least 2 months and cannot be explained by an alternative diagnosis [1]. According to the most recent global analyses, the cumulative prevalence of post-COVID-19 condition in children ranges between 9% and 63% and is up to six-fold higher than that of similar post-viral infection conditions [2]. In children and adolescents with post-COVID-19 condition, the most frequently reported respiratory symptoms are persistent cough, fatigue and exertional dyspnoea [3].

There is no consensus in the literature about the impact of SARS-CoV-2 infection on pulmonary function. The severity of respiratory failure during COVID-19 correlates with the degree of pulmonary function impairment and respiratory quality of life in patients in the 12 months after acute infection [4, 5]. The inflammatory process due to COVID-19 may continue regardless of its severity and, consequently, peripheral airways may be affected [5]. Some studies have demonstrated that patients with post-COVID-19 infection have impaired lung function [6], cardiorespiratory symptoms, fatigue and decreased functional capacity from 3 to 12 months after infection [7]. In contrast, some studies have suggested that SARS-CoV-2 is no more likely than other infections to cause long-term pulmonary sequelae in children and adolescents [8, 9]. Pulmonary function was found to be normal in 75% of children and adolescents with persistent respiratory symptoms [8]. The discrepancy between persisting respiratory symptoms and normal pulmonary function suggests that a different underlying pathology, such as dysfunctional breathing, may be present [9].

Female sex and older age are mentioned as risk factors related to post-COVID-19 condition in children and adolescents [10]. The differential growth patterns of airways and lung parenchyma in girls and boys during puberty have significant implications for pulmonary function. Female airways and lung parenchyma grow proportionally, whereas in males the growth of the airway lags behind the growth of the lung [11]. This suggests that larger male lungs would have longer conducting airways, being disadvantageous during expiration.

To date, rehabilitation appears the most effective treatment for post-COVID-19 condition [12, 13]. Exercise-based rehabilitation, including aerobic exercise, resistance training, respiratory muscle training [14] and behavioural interventions [15], plays an important role in managing fatigue, exercise intolerance, dyspnoea, mental health and sleep-related problems, and musculoskeletal pain in children. The previously described water- and land-based exercise for children with respiratory problems [16, 17] may highlight the potential benefits for developing programmes in a similar environment for those experiencing long COVID.

This study aimed to assess the effect of water- and land-based exercise on lung function in children with post-COVID-19 condition.

Methods

Study design

This was a randomised controlled trial following the recommendations of the Consolidated Standards of Reporting Trials (CONSORT) statement of 2010 [18]. It was conducted at the Faculty of Rehabilitation, Jozef Pilsudski University of Physical Education in Warsaw (AWF Warsaw), Poland, by following all principles described in the Declaration of Helsinki. The study was approved by the Ethics Committee of the AWF Warsaw (01–55/2021), as well as prospectively registered on ClinicalTrials.gov (NCT05216549). The protocol has previously been published elsewhere [19]. All legal guardians of the children signed informed consent to participate in the study.

Participants

Children with post-COVID-19 condition were included. The following inclusion criteria were used: 1) symptoms typical of post-COVID-19 condition, including fatigue and shortness of breath/respiratory issues following an initial COVID-19 infection; 2) aged 10–12 years old. The exclusion criteria included 1) absolute contraindications to exercise, 2) unstable cardiac conditions and 3) currently engaged in regular supervised exercise training more than twice per week. Children were recruited from primary care clinics and schools in Warsaw (Poland). The parents of the potential participants answered an online pre-screening questionnaire. If eligible and following the provision of written consent, children were assessed by a general practitioner.

Randomisation

Participants were randomly assigned to either a water-based exercise (AQUA), land-based exercise (LAND) or a control group with no exercise (CONTROL). A researcher external to the study conducted randomisation before the initiation of the study, using an Excel random number generator. Allocation was concealed in sealed opaque envelopes. Assignment to groups was prepared in five random versions. The allocation providing the lowest F value for MANOVA was used when comparing the basic characteristics of the subjects. Randomisation was stratified by age, sex and exercise capacity. It was impossible to blind the physiotherapist or participants due to the nature of the exercise interventions.

Exercise intervention

The water-based and land-based exercise programmes were performed for 8 weeks, with a frequency of 2 days per week (16 sessions in total) and the duration of each session was 45 min. The exercises in the water and on land were matched as closely as possible in terms of intensity, duration and muscle groups trained. The training consisted of an 8 min warm-up period (consisting of upper and lower limbs aerobics, breathing exercises, and stretches), 32-min aerobic training (circuit stations with endurance exercises for upper and lower limbs and breathing patterns) and 5-min cool down (upper limb and thoracic cage stretch and breathing control). Participants were encouraged to exercise at an intensity of 6–8 (“getting quite hard” to “hard”) on the Pictorial Children's Effort Rating Table [20]. The water-based exercise sessions were held in a community swimming pool (depth 1.2–1.5 m; length 12 m; width 3 m; water temperature approximately 30 °C; humidity 60%). The land-based exercise sessions were conducted in a university gymnasium with a controlled temperature. All sessions in the water and on land were supervised by experienced physiotherapists. Participants in the control group (no exercise) were told not to change their current exercise routine and were offered the option to join the water and/or land exercise programme after completion of the final study evaluation.

Both the water-based and land-based interventions have been described in detail in the published trial protocol [19].

Outcomes

The primary outcomes of this study were exercise capacity and fatigue and have been previously reported [21]. The secondary outcomes of this study were the changes in lung function from baseline to immediately after the 8-week intervention, the results of which are presented in this report. Pulmonary function tests were performed using the portable Lungtest Handy (MES LLC, Cracow, Poland). Parameters measured included forced expiratory volume in 1 s (FEV₁), forced vital capacity (FVC) and vital capacity (VC). All spirometric values were analysed using a Z-score according to the reference values of the Global Lung Function Initiative (GLI) powered by the European Respiratory Society (ERS) [22]. All measurements were performed according to the American Thoracic Society (ATS) and ERS consensus criteria [23].

Sample size

The necessary minimum total number of participants (n=69) was determined using G*Power software 3.1.9.7, assuming the detection of a moderate effect size of REPETITION_ GROUP interaction (d=0.50 or η²=0.04) for lung function with a significance level of 0.05 and statistical power of 0.85.

Statistical analysis

Statistical analyses were performed with the Statistica 14.0.0.15 program (TIBCO Software Inc., Palo Alto, CA, USA, 2020). The distribution and normality of the data were analysed with the Shapiro–Wilk test. The data were expressed as means, standard deviation (mean±sd), percent (%), group size (n), minimum (min) and maximum (max) together with a 95% confidence interval. The main method used was an analysis of variance (ANOVA) for repeated measures used to test the interaction between GROUP (3) (fixed factor) and REPETITION (2) to answer the question of whether the interventions caused different responses. Detailed comparisons were made using Tukey's post hoc test. ANOVA results were described by Fisher's F with degrees of freedom in parentheses, p-value and partial η² as effect size. It was assumed that η² values of approximately 0.01, 0.06 and 0.14 could be considered small, moderate and large, respectively. The MANOVA for repeated measures was used to assess changes in the severity of obstructive and restrictive disorders in the tested groups. Missing data was pairwise removed. The level of statistical significance was set at α=0.05.

Results

Recruitment began in January 2022 and the post-intervention data were collected in June 2022. 179 children were assessed for eligibility and 86 who met the inclusion criteria participated in the study and were randomised to one of the three groups (figure 1). The mean attendance rate was 84% in the AQUA group and 77% in the LAND group. There was no significant difference in attendance between the two groups and no adverse events were recorded.

FIGURE 1.

Consolidated Standards of Reporting Trials (CONSORT) flowchart of study participants. AQUA: water-based exercise group; CONTROL: control group with no exercise; LAND: land-based exercise group.

74 participants completed the intervention. Table 1 shows the general characteristics of the study groups.

TABLE 1.

Participant demographics

	AQUA (n=25)	LAND (n=23)	CONTROL (n=26)
Girls, n	10	9	11
Age (years)	10.5±0.5	10.6±0.7	10.5±1.1
Body mass (kg)	41.8±12.1	41.7±9.6	39.6±8.9
Height (cm)	149±6	150±8	146±7
BMI (kg·m−2)	18.5±4.2	18.2±2.8	18.5±4.3
Boys, n	15	14	15
Age (years)	11.0±0.9	11.3±0.6	10.6±1.0
Body mass (kg)	47.5±15.2	44.5±12.9	37.3±10.2
Height (cm)	153.4±9.7	152.3±6.6	145.3±9.6
BMI (kg·m−2)	19.9±4.9	19.0±4.3	17.4±3.2

Data are presented as mean±sd unless otherwise stated. AQUA: water-based exercise group; BMI: body mass index; CONTROL: control group with no exercise; LAND: land-based exercise group.

The results of the two examination sessions for lung function (pre- and post-test) are shown in tables 2 and 3, for girls and boys respectively. Due to a difficulty some children had in performing the lung function tests, only those tests made in accordance with the ERS were used for analysis. The exact number of valid data for each parameter is listed in tables 2 and 3.

TABLE 2.

Spirometry test results in boys

	AQUA				LAND				CONTROL
	n	Pre	Post	Mean difference (95% confidence interval)	n	Pre	Post	Mean difference (95% confidence interval)	n	Pre	Post	Mean difference (95% confidence interval)
FEV1 (L)	14	2.24±0.55	2.35±0.64	0.09 (−0.18–0.36)	13	2.34±0.52	2.36±0.59	0.12 (−0.13–0.36)	14	2.11±0.39	1.88±0.46	−0.24 (−0.49–0.02)
FEV1 (Z)	14	−1.11±1.49	−0.76±1.41	0.20 (−0.75–1.16)	13	−0.60±1.66	−0.55±1.85	0.39 (−0.43–1.20)	14	−0.53±0.69	−1.35±1.59	−0.85 (−1.76–0.06)
FEV1 % pred	14	87.27±17.25	91.35±16.04	2.40 (−8.64–13.45)	13	93.03±19.19	93.63±21.25	4.53 (−5.00–14.06)	14	94.01±7.88	84.45±18.34	−9.92 (−20.50–0.65)
FVC (L)	14	2.50±0.68	2.66±0.72	0.15 (−0.17–0.47)	13	2.56±0.51	2.51±0.63	0.02 (−0.25–0.29)	14	2.21±0.41	2.15±0.54	−0.05 (−0.24–0.14)
FVC (Z)	14	−1.59± 1.35	−1.08±1.30	0.39 (−0.60–1.37)	13	−1.21±1.33	−1.41±1.40	0.01 (−0.80–0.81)	14	−1.46±0.60	−1.62±1.25	−0.17 (−0.76–0.42)
FVC % pred	14	81.85±15.30	87.63±14.87	4.35 (−6.80–15.51)	13	86.25±15.07	83.95±15.88	0.03 (−9.09–9.15)	14	83.16±6.88	81.45±14.34	−1.92 (−8.67–4.83)
FEV1/FVC %	9	80.35±17.53	83.67±9.36	3.31 (−7.84–14.46)	11	84.51±15.30	82.87±8.17	0.03 (−9.09–9.15)	12	88.80±7.93	82.53±12.69	−5.56 (−15.46–4.34)

Data presented as mean±sd unless otherwise stated. AQUA: water-based exercise group; BMI: body mass index; CONTROL: control group with no exercise; FEV₁: forced expiratory volume 1 s; FEV₁ % pred: percentage of predicted value for FEV₁; FVC: forced vital capacity; FVC % pred: percentage of predicted FVC; LAND: land-based exercise group; Z: Z score.

TABLE 3.

Spirometry test results in girls

	AQUA				LAND				CONTROL
	n	Pre	Post	Mean difference (95% confidence interval)	n	Pre	Post	Mean difference (95% confidence interval)	n	Pre	Post	Mean difference (95% confidence interval)
FEV1 (L)	9	2.10± 0.32	2.40±0.35	0.33 (0.15–0.51)	8	2.19±0.42	2.31±0.35	0.14 (−0.06–0.34)	10	1.99±0.27	1.89±0.33	−0.14 (−0.33–0.06)
FEV1 (Z)	9	−0.72±0.91	0.46±0.81	1.22 (0.61–1.83)	8	−0.58± 1.45	−0.08± 0.94	0.48 (−0.29–1.26)	10	−0.69±1.30	−1.07±1.42	−0.53 (−1.31–0.24)
FEV1 % pred	9	91.45± 10.76	105.26± 9.35	14.30 (7.16–21.45)	8	93.18±16.72	99.05±10.85	5.68 (−3.33–14.68)	10	91.72±15.42	87.15± 16.75	−6.38 (−15.54–2.79)
FVC (L)	9	2.42±0.29	2.57±0.44	0.20 (0.05–0.35)	8	2.39±0.49	2.59±0.41	0.22 (0.00–0.44)	10	2.23±0.48	2.00±0.45	−0.26 (−0.45–−0.06)
FVC (Z)	9	−0.64± 0.75	−0.13±0.85	0.60 (0.15–1.05)	8	−0.97±1.53	−0.28±0.95	0.67 (−0.11–1.46)	10	−0.88±1.63	−1.71±1.34	−0.94 (−1.61–−0.27)
FVC % pred	9	92.50±8.80	98.60±10.08	7.08 (1.81–12.34)	8	88.86±17.39	96.80±11.06	7.72 (−1.34–16.79)	10	90.01±18.96	80.39±15.33	−10.93 (−18.58–−3.28)
FEV1/FVC %	8	85.56±7.23	78.48±5.96	−11.62 (−51.64–28.39)	8	84.61±8.30	87.24±5.98	1.49 (−6.69–9.67)	9	82.39±14.51	79.56±12.25	−6.83 (−16.33–2.67)

Data presented as mean±sd unless otherwise stated. AQUA: water-based exercise group; BMI: body mass index; CONTROL: control group with no exercise; FEV₁: forced expiratory volume 1 s; FEV₁ % pred: percentage of predicted value for FEV₁; FVC: forced vital capacity; FVC % pred: percentage of predicted FVC; LAND: land-based exercise group; Z: Z score.

No significant differences were found at baseline neither between groups (AQUA, LAND and CONTROL) nor between genders in terms of spirometric results. An increase in all spirometric parameters expressed as % predicted values was observed in the active girls (AQUA and LAND) group. The largest increase was found in FVC % pred in the girls in the LAND group (8.9%) and in the AQUA group (6.6%). In the AQUA boys group, an increase was found in all measured parameters with the largest increase (7.1%) in FVC % pred. In the boys LAND group, an increase was only found in FEV₁% pred (0.6%). In the control group, all spirometric parameters decreased, with the largest in the boys' group FEV₁% pred 11.3% and in the girls' group FVC % pred 12%; however, the post hoc test did not detect any significant differences within groups.

To assess the effect of intervention, GROUP×REPETITION interactions were considered. The interaction was significant for absolute FEV₁ values (F (2,64)=6.80; p<0.01, η2=0.18) and manifested itself in such a way that while the groups (AQUA, LAND and CONTROL) did not differ significantly before the intervention, after the intervention there was a significant difference (p<0.01) between the two exercise groups compared to the control group, in which post-intervention values were lower. A similar interaction was observed in FEV₁, expressed as a percentage of the norm for healthy children (F(2,64)=6.96; p<0.01, η2=0.18).

Analysing absolute values of FVC, GROUP×REPETITION (pre-–post-test) interaction was observed (F(2,64)=4.63; p<0.05, η2=0.13). There were significant differences in post-test results between the CONTROL and AQUA groups (p<0.01) and the CONTROL and LAND groups (p<0.05). A similar interaction was noted in FVC expressed as a percentage of the norm for healthy children (F(2,64)=4.99; p<0.01, η2=0.14); however, there were no significant differences within groups or between groups. GROUP×REPETITION interactions for FEV₁/%FVC (F(2,64)=0.71, p>0.05, η2=0.030) showed that the changes that occurred varied from group to group and the reason for this was an upward trend in the LAND and AQUA groups and a downward trend in the CONTROL group.

Lung function abnormalities were classified according to GLI standards (figure 2). Following the intervention, a shift in the severity of obstruction towards normal values was observed; however, two new cases from the control group with restrictive disorders were found.

FIGURE 2.

Distribution of lung function abnormalities among participants pre- and post-intervention. FEV₁: forced expiratory volume in 1 s; FVC: forced vital capacity.

Discussion

This was the first randomised controlled trial examining the effect of exercise training on lung function in children with mostly respiratory symptoms typical for post-COVID-19 condition, including fatigue, shortness of breath and other respiratory issues following an initial COVID-19 infection. Our results demonstrated lung function results recorded predominantly below normal (i.e. obstructive) or at the lower level of the normal for healthy children. This may indicate that the inflammatory process due to COVID-19 may persist in some children beyond the acute period, even when clinical findings on admission were not severe, and that the peripheral airways may have been correspondingly affected.

In adults, post-COVID-19 condition typically manifests with abnormalities in pulmonary function tests and/or chest imaging in approximately 30% of cases [24]. Conversely, research on pulmonary function tests in paediatrics is limited, with some evidence indicating that the lasting respiratory effects of COVID-19 in children are generally mild, with a positive prognosis. Most studies have demonstrated spirometry and plethysmography results within normal ranges (between 93 and 105%) for the majority of children with post-COVID-19 condition [25–27]. Sansone et al. [28] suggest that the mild lung inflammation during acute COVID-19 was unable to induce fibrotic change, persistently affect the lungs or decrease lung function parameters.

There are few studies evaluating pulmonary function in children after COVID-19 infection with similar results to our study. A systematic review with meta-analysis showed that children and adolescents may have persistent abnormalities in lung imaging and function between 3 and 12 months after COVID-19 infection [7]. Öztürk et al. [5] evaluated 50 children for ongoing respiratory symptoms and pulmonary function tests 3 months after infection. The authors observed persistent respiratory symptoms in 28% of children, a significant decrease in FEV₁/FVC and an increase in lung clearance index in this group of patients. In another study, children were assessed at 6 months post-COVID-19 infection and although they did not demonstrate respiratory symptoms, 34% had mild abnormalities on computed tomography and 10.5% had pulmonary function test abnormalities [29]. Vezir et al. [30] evaluated spirometry in 57 paediatric patients with COVID-19 and three patients with comorbid respiratory disease had impaired pulmonary function tests (obstructive-type deficit in two patients and restrictive-type deficit in one patient). An obstructive pattern was identified in 5% of the children and adolescents studied by Campos et al. [7]. In our study, 16 participants had an obstructive deficit and two had a restrictive-type deficit. The cause of such impaired ventilatory lung function could have been viral pneumonia followed by long-term complications such as prolonged cough, chronic bronchitis, asthma and bronchodilation, which led to respiratory dysfunction and caused obstructive and restrictive lung disease [31].

Considering the effects of exercise training on lung function, the training interaction in our study was significant for the absolute FEV₁ values, which reflect the volume of air exhaled in the first second of forced expiration after maximal inspiration. With regard to this parameter, there was a significant difference between the two exercise groups after the intervention compared to the CONTROL group, in which the values were lower after the intervention. A similar interaction was found for the FVC values: a significant difference was observed between the post-test results of the control and exercise groups. The combined analysis of both parameters (FEV₁/%FVC) indicates the presence of opposing trends, each showing an increase in values after exercise in the LAND and AQUA groups and a decrease in values in the CONTROL group. All these observations allow us to conclude that the training interventions (both AQUA and LAND) were effective in improving some dynamic spirometry parameters in children with post-COVID-19 conditions. However, due to the limited duration of the exercises and the still relatively small sample size, no actual changes in lung function could be observed in the children who participated in our intervention, so it is difficult to say with confidence that the intervention we propose will fully improve the lung function of children struggling with the consequences of COVID-19. Further studies with a larger number of participants over a longer period of time are undoubtedly necessary.

Nevertheless, it should be indicated that our results are consistent with studies previously conducted in adults worldwide. Evidence of exercise training counteracting pulmonary function deficits in paediatric populations with post-COVID-19 condition is limited. A review of 14 randomised clinical trials with 1244 adult participants found that patients with post-COVID-19 condition who participated in respiratory training and exercise-based rehabilitation programmes experienced improvement in lung function, shortness of breath and quality of life compared to those who received standard treatment [32]. A few other studies have demonstrated positive effects of physical training on improving lung function in adults. Our results demonstrate that a protocol-based intervention in water and on land in children after COVID-19 infection had an impact on lung function, in terms of dynamic spirometry parameters such as FEV₁ and FVC, expressed both in absolute values and in % of predicted values. These results are consistent with those obtained by Hockele et al. [33]. An exercise programme for post-COVID-19 patients, based on 16-session rehabilitation protocol (inspiratory muscle training, aerobic exercise and peripheral muscle strength performed two times per week, for 60 min each) over 16 sessions (the number of sessions, average 12.7±2.7) resulted in improvement in pulmonary function and respiratory muscle strength. According to the authors, the improved respiratory performance was a consequence of strengthening the inspiratory force and not lung capacities and volumes. Similar results were found by Araújo et al. [34] who demonstrated that a cardiopulmonary rehabilitation programme consisting of continuous moderate-intensity aerobic and resistance training over 6 weeks improved lung function, respiratory muscle strength in patients post-COVID-19. Exercise protocols available in the literature to improve lung function have been conducted on land. Our exercise training in water has additional advantages over land-based exercise training. Water provides resistance in all directions, requiring more effort and engagement of multiple muscle groups simultaneously resulting in increased resistance compared to exercising on land.

Our study showed increases in all spirometric parameters in girls in the exercising intervention groups, regardless of the type of training (AQUA and LAND group). In boys, a similar picture emerged only in the group that trained in the water (AQUA group), while in the boys who trained in the gym (LAND group) a slight improvement was observed only in one spirometric parameter (FEV₁ % pred). One of the reasons for the observed differences could be the different anatomical and functional development of the individual elements of the respiratory system in children aged 10–12 years. Studies suggest that maximal expiratory flow and total lung capacity increases more in girls than in boys between 8 and 12 years of age [35]. There are also other reports of higher volume-corrected maximal expiratory flows in young girls [36] and studies indicating higher volume-corrected flows in female adolescents [37] compared to boys and males, respectively. In addition, girls have a greater increase in residual volume and lower maximal expiratory and inspiratory pressures. Due to the smaller lung volume of girls, shorter airways would be expected; however, research suggests that girls generate larger flows, meaning that their airways may be wider than those of boys. There is also evidence of unequal airway and airspace growth between 8 and 12 years of age [35]. Measurement of total lung capacity using chest radiographs has shown that lung volumes grow similarly in males and females, but that girls peak earlier [38]. In girls, growth is greater between the ages of 8 and 12 years compared to boys in both the airways and air spaces, which is due to their greater somatic growth. Therefore, it is likely that girls reach their peak lung growth around the age of 12 years, while lung growth continues in boys into adolescent [35]. Accelerated lung growth is also associated with the diverse functionality of this organ; this is reflected in lung function measurements, which show an earlier growth peak in spirometric measurements in girls [39], which seems to confirm our observations.

Despite the careful planning of our study, we are aware that our research has limitations. First is the lack of standardisation in the diagnosis of post-COVID-19 conditions. Although recommendations exist from various research institutions and medical centres, they are yet to be standardised [40, 41]. Secondly is the performance of spirometry, which is an accurate and repeatable test when performed correctly. However, performing spirometry correctly is not a technically simple task, especially in children. Although there are studies suggesting that even preschool children are able to perform technically acceptable and repeatable spirometry under normal conditions so that spirometry may be a useful screening test for abnormal lung function in this age group [42], there are reports of only 54% of children being able to perform acceptable and repeatable spirometry according to ATS criteria [43]. Finally, lung function was a secondary outcome of this randomised controlled trial and our study was not powered for changes in lung function; therefore, future trials with larger sample sizes and longer interventions are recommended.

In conclusion, this randomised controlled trial found that a supervised twice-weekly 8-week exercise training programme in water and on land improved lung function, in terms of some dynamic spirometry parameters, in children with post-COVID-19 condition characterised by fatigue and shortness of breath/respiratory issues.