Feasibility and Tolerance of High-Intensity Interval Training in Adults With Cystic Fibrosis

Highlights

• •High-intensity interval training (HIIT) is feasible, safe, and well tolerated in most people with cystic fibrosis (CF).
• •HIIT is less time-consuming but as effective as conventional physical rehabilitation.
• •HIIT may be an alternative for new CF profiles emerging with modulator therapy.

Abstract

Objectives

To assess the feasibility and the tolerance of a high-intensity interval training (HIIT) program compared with conventional rehabilitation during a 3-week hospital stay in adults with cystic fibrosis (CF) and to analyze the effectiveness and impact on health indicators of HIIT.

Design

A prospective, 2-center, open-label, randomized, controlled study.

Setting

Cystic Fibrosis Resource and Expert Centers, Roscoff and Giens (France).

Participants

A total of 62 adults (aged: 36.1±11y; M/F ratio: 1.3) with CF were included.

Interventions

Intervention group performed 3 HIIT sessions, alternating 30 seconds of work and 30 seconds of rest repeated 6 times ≥80% of maximum heart rate, and 2 low-intensity sessions per week. Control group (conventional rehabilitation) performed 5, 20-to-30-minute sessions per week at the first aerobic ventilatory threshold.

Main Outcome Measures

(1) Feasibility was evaluated by compliance; (2) tolerance by the degree of dyspnea and adverse effects; (3) effect on physical capacity using the 6-minute walk, body composition, the forced expiratory volume in 1 second (FEV1), and the multidimensional dyspnea profile; (4) impact on health indicators by the Fibrosis Quality of Life Questionnaire and the Hospital Anxiety and Depression; and (5) participant satisfaction by the STARFISH (perception of physical activity) and the Physical Activity Enjoyment Scale.

Results

Compliance with exercise sessions was marginally higher but statistically significant in the HIIT (93.3%) than the control (86.7%) group, P=.032. However, tolerance was lower in those with severe respiratory impairment and those without cystic fibrosis transmembrane conductance regulator (CFTR) modulator therapy. Effectiveness on physical capacity (6-minute walk test, muscle strength, and FEV1) and quality of life was similar between the training methods.

Conclusions

HIIT is feasible, safe, and well tolerated and could be used as a timesaving alternative to conventional treatment, particularly in people under CFTR modulator therapy.

For people with cystic fibrosis (CF), it is recommended to perform 20-45-minute physical activity sessions at least 3 times a week, combining endurance and resistance training to improve their physical capacity, lung function, and quality of life.¹ However, it can be difficult for people with CF to achieve the target heart rate and dyspnea threshold because of excessive breathlessness and muscle fatigue. Health care and work/school constraints limit adherence to physical activity programs, which are considered “time-consuming.” High-intensity interval training (HIIT) consists of alternating short periods of intense effort with equally short recovery periods. The number of cycles and session durations are adapted to the individual’s abilities and objectives. The breaks allow partial recovery from breathlessness and muscle fatigue, optimizing work times. HIIT is well tolerated by healthy people² and people with chronic obstructive pulmonary disease,³ is less time-consuming,^4,5 and yields similar results to continuous training but with less dyspnea and muscle fatigue^6,7 and greater enjoyment.⁸ In small CF cohort studies, 6- to 12-week HIIT programs in outpatients improved VO₂max^9,10 and exercise capacity10, 11, 12, 13 similarly or slightly superior to conventional training with less dyspnea11, 12, 13 and higher satisfaction.^10,14 Here, we wished to evaluate (1) the feasibility, (2) the tolerance, and (3) the benefits of HIIT during a 3-week inpatient rehabilitation stay, with multidisciplinary management including nutritional program assessed by dieticians, daily chest physiotherapy, and exercise sessions supervised by sport teachers and physiotherapists.

We also wanted to explore the impact of several factors, such as respiratory impairment severity, the presence of diabetes, nutritional status and treatment with cystic fibrosis transmembrane conductance regulator (CFTR) modulators, on the feasibility and tolerability of the HIIT program, with the aim of potentially establishing management recommendations.

Methods

Design

We conducted a 2-center (Cystic Fibrosis Resource and Expert Centers) prospective, randomized controlled trial between September 2021 and September 2023. Participants were randomly assigned (1:1) to HIIT or conventional (control) rehabilitation using an electronic system. The study was registered on ClinicalTrials.gov (NCT04888767) and approved by a national ethics committee (n°21.0168.001390 CPP ouest VI).

Sample

Eligible adults (≥18y) with confirmed CF (sweat test and/or genetic test), and able to participate in the training were identified from preadmission files. CFTR modulator treatment must not have been initiated within 4 weeks before inclusion.

Exclusion criteria were pregnancy, lung transplant or on the transplant waiting list, unstable clinical condition (eg, exacerbation within 72h), dialysis, or pulmonary hypertension.

Potential participants received information at least 10 days before admission and were contacted by a research nurse. Written informed consent was obtained during the admission consultation.

Intervention

All participants were hospitalized for 18 days of rehabilitation. Exercise was performed on a cycle ergometer (Bike Reha EN-Cardio ENRAF-NONIUS 230v/50hz)^a, under professional supervision. The intervention group participated in 3 HIIT sessions (9 sessions; minimum 8) and 2 low-intensity sessions (6 sessions; minimum 5) per week, 1 session per day from Monday to Friday. The HIIT sessions began with a 2-minute warm-up at minimal or no power, followed by 30 seconds of work alternating with 30 seconds of rest (without pedaling), which was repeated 6 times. Initial intensity was set at 80% of predicted maximal heart rate (defined by the formula age -220), progressively increased toward 100% depending on tolerance. The sessions ended with a 2-minute recovery at minimum or no power. The low-intensity sessions were performed at <10% increase from resting heart rate.

The control group participated in 5 sessions/week, 1 session per day from Monday to Friday (15 sessions; minimum 13). Sessions began with a 2-minute warm-up at minimum or no power followed by continuous work for 20-30 minutes (the duration increased over the sessions and was individualized to patients’ abilities) at the 1st aerobic ventilatory threshold defined during the 6-minute walk test (6MWT).¹⁵ The session ended with a 2-minute recovery at minimum or no power.

Assessments

Assessments were conducted at admission (V1) and discharge (V2), unless otherwise specified.

Feasibility was evaluated by compliance with the sessions, ie, the proportion of sessions completed during the stay. The number and nature of adverse events (AEs) were recorded throughout the study.

Tolerance was assessed by comparing the degree of dyspnea at rest using the modified Borg scale¹⁶ and lower limb muscle fatigue using a 0-10 scale, as well as blood creatine kinase (CPK) dosage (spectrophotometric analysis, medical biological analysis).

Physical capacity was evaluated using the 6MWT. The test was conducted according to the American Thoracic Society and the European Respiratory Society recommendations.17, 18, 19 Dyspnea assessed using the modified Borg scale and heart rate and oxygen saturation were evaluated before 6MWT and after 3 minutes of rest.

Body composition (fat mass, lean mass, total body water, and basal metabolic rate) was determined by impedancemetry (Bodystat Quadscan 4000)^b, forced expiratory volume in 1 second (FEV1) by spirometry according to the American Thoracic Society recommendations, and dyspnea by the multidimensional dyspnea profile.^20,21 For the quadriceps strength measurement, participants were seated and the knee flexed at 90°. Quadriceps maximal voluntary force was measured using a microFET2^c dynamometer during a maximal isometric knee extension held for 3-5 seconds.²² Quadriceps endurance was assessed during a single sustained submaximal isometric contraction at 50%-60% of maximal voluntary contraction, with endurance defined as time to task failure (s).

The following self-report questionnaires were administered: Cystic Fibrosis Quality of Life Questionnaire,^23,24 Hospital Anxiety and Depression (HAD),^25,26 and STARFISH (perception of physical activity).²⁷ The Physical Activity Enjoyment Scale (PACES) was administered at discharge.²⁸

Participants with diabetes underwent blood glucose measurements using a “FreeStyle” Libre Glycemic Holter^d. Measurements were taken at the beginning, at session end, and after a 15-minute recovery period. Episodes of hypoglycemia during sessions were recorded.

Statistical analysis

Because of the lack of literature, sample size was estimated heuristically based on regression modeling principles and following Green's²⁹ recommendation (target n=100).

All statistical analyses were performed using the Statistical Software R version 4.3.0^e. All analyses were performed on the intention-to-treat population, except tolerance, which was performed on the safety population (at least 1 session completed).

Qualitative variables are expressed as numbers and frequencies, and quantitative variables as mean ± SD or median (interquartile range) depending on distribution normality (Shapiro-Wilk test). Changes from admission to discharge were calculated for all the quantitative variables and, along with compliance, compared between groups using the Student or Mann-Whitney-Wilcoxon test depending on the normality of the distributions. The same tests compared changes and compliance between the binary modalities of participant characteristics for a fixed group. Respiratory impairment severity had 3 levels and was compared using the Kruskal-Wallis test. The paired Student's t test or Wilcoxon signed ranks test were then applied to evaluate within-group changes.

Qualitative safety criteria (patients’ rate, AE rate) were compared between groups or within the same group between patient characteristics, using the chi-square or Fisher's exact test.

Sensitivity analyses were performed using linear regression models to explain changes, with adjustments for the group effect on the baseline parameter, compliance, etc. The regression application conditions were verified on the residuals (homoscedasticity and normality). A log transformation or bootstrap method was applied where necessary.

As blood glucose levels were recorded at 3 time points in participants with diabetes, changes were compared between groups using a mixed linear model with group × time interaction, and adjusting on initial level and compliance. A bootstrap method was used to calculate the P values.

A significance level of 5% was set for all analyses.

Results

Sample

Of the 67 eligible individuals, 62 were randomly divided and included in the intention-to-treat population: 31 in the control group and 31 in the intervention group. In the control group, 1 participant withdrew consent before the end of the study and did not complete all the visits; 2 did not complete any training sessions but completed the assessment visits and were excluded from the safety population (fig 1).

Fig 1.

Flowchart of inclusions.

The demographic and clinical characteristics of the control and intervention groups were similar (table 1).

Table 1.

Clinical characteristics of the control and HIIT group participants

	Sample Characteristics
	Total Sample (N=62)	Control Group (n=31)	HIIT Group (n=31)	P Value
Sex, n (%)
Female	27 (43.5)	13 (41.9)	14 (45.2)	1
Male	35 (56.5)	18 (58.1)	17 (54.8)
Age (y)	36.1±11.1	35.9±11	36.3±11.3	.87
Weight (kg)	62.2 [55.5; 72.1]	66.4 [56.2; 71.8]	60.7 [55.3; 73]	.79
Height (cm)	167.2±8.9	167.3±8.9	167.1±9.2	.91
BMI	22.5 [20; 25.1]	22.6 [20.3; 25.2]	22.3 [19.8; 24.6]	.83
Diabetes, n (%)	19 (30.6)	10 (32.3)	9 (29)	1
Respiratory impairment severity, n (%)
Mild (FEV1 ≥70)	16 (25.8)	9 (29)	7 (22.6)	.74
Moderate (FEV1 41-69)	31 (50)	14 (45.2)	17 (54.8)
Severe (FEV1 ≤40)	15 (24.2)	8 (25.8)	7 (22.6)
Malnourished (BMI ≤18.5), n (%)	6 (9.7)	4 (12.9)	2 (6.5)	.67
CFTR modulators, n (%)	52 (83.9)	25 (80.6)	27 (87.1)	.73

Abbreviation: BMI, body mass index.

Feasibility

Compliance with the sessions was significantly marginally higher but statistically significant in the HIIT (93.3%) than in the control group (86.7%), P=.032 in the safety population (fig 2).

Fig 2.

Box plot of participant compliance with physical exercise sessions: percentage of sessions completed during the entire stay for the control and the HIIT groups in the safety population.

Tolerance

In the safety population, the rate of participants with at least 1 AE did not differ between groups (P=.43). Eighteen AEs (14 participants) were recorded in the HIIT and 24 (17 participants) in the control group (table 2). None were serious. Only 9.5% of these required temporary training cessation; in 83.3% of cases, there was no change to the training. Moderately severe AEs were more frequent in the control group (control: 83.3% vs HIIT: 50%, P=.048). However, a larger (although nonsignificant) proportion of AEs in the HIIT group might have been related to the intervention (HIIT: 3 vs control: 1 P=.3): recurrent episodes of nonsymptomatic hypoglycemia or respiratory exacerbation.

Table 2.

Adverse event list (a) in HIIT group and (b) in control group

Number	Event/Adverse Effect	Intensity	Causality Link With Intervention
(a) Adverse event list in HIIT group
8	Airway congestion, exacerbation or pulmonary infection	Mild to moderate	No
2	Digestive symptoms (diarrhea, stoma pain)	Mild	No
2	Asymptomatic hypoglycemia episodes	Mild	No/possible
2	Dizziness on exertion	Moderate	Probable
1	Asthenia	Mild	No
1	Mycosis	Mild	No
1	Renal function impairment	Moderate	No
1	Road traffic accident	Mild	No
(b) Adverse event list in control group
6	Airway congestion, exacerbation or pulmonary infection	Mild to moderate	No
5	Musculoskeletal disorders	Mild to moderate	No
2	Digestive symptoms	Moderate	No
2	Allergy	Moderate	No
2	Ophthalmic symptoms	Mild	No
1	Anxiety aggravation	Moderate	No
1	Febrile episode	Moderate	No
1	Mycosis	Mild	No
1	Dehydration	Moderate	Possible
1	Sleep disorders	Moderate	No
2	Dermatologic symptoms	Moderate	No

HIIT did not increase dyspnea (modified Borg index; P=.73), muscle fatigue at rest (P=1), or muscle damage (CPK levels) (P=.34) compared with the control group (supplementary data 1).

Compliance and tolerance with HIIT as a function of participant profile

Respiratory severity (FEV1), CFTR modulators treatment, or diabetes did not affect compliance of HIIT. In the control group, compliance was poorer for those taking CFTR modulators (P=.035; 80% compliance with modulators vs 100% without).

Respiratory severity (FEV1≤40%) influenced tolerance in HIIT, with slight increases in resting dyspnea (Borg index) (P<.0001) (fig 3A). There was no such effect in the control group. In addition, in HIIT group, patients without modulator therapy experienced greater increases in dyspnea (P=.0036) (fig 3B) and muscle fatigue (fig 3C), although numbers were small. More AEs occurred in those without modulator therapy in the HIIT group (P=.032). In the control group without modulator therapy (6 participants), Borg index improved at discharge (difference −2.5 [−3.1; −1.5], P=.0088) compared with patients with modulator therapy.

Fig 3.

Box plots illustrating (A) change in Borg Dyspnea Index from admission to discharge as a function of respiratory impairment severity (FEV1) in the HIIT group, (B) change in Borg index in the HIIT group, and (C) change in muscle fatigue from admission to discharge as a function of CFTR modulator treatment in the HIIT group.

Nutritional status could not be analyzed in relation to compliance and tolerability of HIIT because of the small number of malnourished participants (body mass index<18.5) in each group.

Effectiveness

6MWT distance increased from admission to discharge in both groups: +15 [1; 35] meter in the control group (P=.013) vs +23 [1; 38] meter in the HIIT group (P=.003) (table 3); the magnitude of change did not differ between groups (P=.68). No desaturation or altered cardiac rhythm occurred. The analysis adjusted on initial distance and session compliance confirmed the absence of a group effect.

Table 3.

Analysis of the 6MWT distance, muscle strength, bioelectrical impedance, spirometry, and the multidimensional dyspnea profile at admission and discharge for the control and HIIT groups

Variable	Control Group			HIIT Group			Change
	Admission	Discharge	Intragroup Difference	Admission	Discharge	Intragroup Difference	Between-Group Difference
6MWT distance (m)	640 (570; 686)	620 (590; 698)	15 (1; 35)*	613 (570; 682)	635 (576; 692)	23 (1; 38)†	P = .68
Borg scale before 6MWT	0 (0; 1)	0 (0; 0)	0 (0; 0)	0 (0; 0)	0 (0; 0)	0 (0; 0)	P = .44
Borg scale after 6MWT	6 (5; 7)	5 (4; 7)	−1 (−2; 1)	5 (3; 7)	5 (3; 8)	0 (0; 1)	P = .3
Spirometry (%FEV1)	53.9±20.4	55.6±21	1.7±3.8*	55±20.9	57.2±21.7	1.3±4.9	P = .75
Left quadriceps strength (N.m)	93.5 (74.5; 158.8)	101.5 (80.8; 159.8)	12 (1; 22.5)†	103 (83; 152.5)	117 (89.5; 156.5)	5 (−11; 20)	P = .22
Right quadriceps strength (N.m)	110.5 (77.2; 145.8)	113 (87; 157)	15 (−5; 23)*	132 (96; 166)	129 (95.5; 162)	4 (−11.5; 14.5)	P = .096
Left quadriceps endurance (s)	39 (30.5; 47.8)	42.5 (31.2; 55)	0.5 (−11.2; 6)	37 (27.5; 45.5)	38 (30; 53.5)	3 (−4.5; 9)	P = .34
Right quadriceps endurance (s)	41 (29.8; 49.8)	39 (31; 45)	−5 (−14; 7)	40 (32; 49.5)	38 (29; 50)	4 (−6; 8)	P = .22
Impedancemetry
Weight (kg)	66.4 (56.2; 71.8)	66.6 (58; 70)	0.5 (−0.2; 1.3)*	60.7 (55.3; 73)	60.6 (57.5; 73.3)	0.1 (−0.8; 2.3)	P = .92
% lean mass	77.8±16.6	77.4±17	−0.3±2.6	78.5±8	78.4±8.1	0.5±3	P = .31
% fat mass	20.4±9.8	20.7±9.6	0.4±2.6	21.5±8	21.6±8.1	−0.5±3	P = .24
% body water	60±8.2	59.3 (50.5; 63.7)	−1.1 (−1.8; 1.1)	58.8±6.9	57.2 (53.3; 63.8)	−0.1 (−0.7; 2.3)	P = .057
Basal metabolism (kcal/j)	1582 (1379; 1723)	1549 (1394; 1724)	2.5 (−28.2; 19.2)	1530 (1351.5; 1748)	1579 (1377; 1785)	16 (−5; 34)	P = .22
Multidimensional dyspnea profile
Discomfort	3.9±2.7	3±2.7	−0.4±2.5	2.9±2.4	3.2±2.7	0.4±2.4	P = .26
Sensory component	15.9±10	12.3±10	−1.9±10.3	11.7±10.9	11.9±11.2	−0.2±6.6	P = .51
Emotional component	1.5 (0; 10.8)	1 (0; 5)	0 (−1.8; 0.8)	2 (0; 5)	1 (0; 3.5)	0 (−3.5;0.5)	P = .4

Intragroup difference, change from admission to discharge.

^⁎P<.05.

^†P<.01.

Quadriceps strength increased significantly and met the minimal clinically important difference criteria only in the control group (left +12 [1; 22.5] Nm [P=.002], right +15 [−5; 23] Nm [P=.044]).

FEV1 improved in both groups, but the change was not considered clinically relevant (control +1.7±3.8, P=.023; HIIT +1.3±4.9, P=.16). Body composition (lean mass, fat mass, and basal metabolic rate) did not change in either group. Weight remained stable, although there was a slight, significant increase of 0.5 [−0.2; 1.3] kg in the control group. Between-group difference in change (table 3) was not significant for quadriceps strength, FEV1, or body composition. This was confirmed by the sensitivity analysis adjusted for baseline measures and session compliance (table 3).

For dyspnea, no change from admission to discharge was found for the 3 multidimensional dyspnea profile subscores (discomfort, sensory, and emotional components) in either group. Similarly, there was no between-group difference in scores.

Quality of life, anxiety, and depression

The physical, psychological, and energy subscores of the Cystic Fibrosis Quality of Life Questionnaire improved in both groups and are clinically relevant. The treatment burden was significantly lower at discharge in the HIIT than in the control group (P=.03). HAD anxiety score decreased in both groups (control P=.014, HIIT P=.004) but the between-group difference in change was not significant (P=.66). HAD depression score improved significantly in the control group (P=.027) but not the HIIT group (P=.076); the between-group difference in change was not significant (P=.45) (supplementary data 2).

Perception of physical activity and satisfaction

The perception of physical activity (STARFISH) improved from admission to discharge in all 12 domains in the HIIT group and in 6 domains in the control group (supplementary data 2). There were no between-group differences for any subdomains (supplementary data 2).

Total PACES score at discharge was significantly lower in the HIIT group (HIIT 37.8±5.8 vs control 41±5.7, P=.038) (fig 4). Enjoyment of physical activity was not affected by respiratory impairment severity, CFTR modulator therapy, or diabetes (P>5%).

Fig 4.

Box plot of the satisfaction score (PACES) assessed at discharge for the control and HIIT groups.

Blood glucose levels

Ten participants in the control group and 9 in the HIIT group had diabetes. One patient in the control group had no glucose measurement during the follow-up period. Blood glucose levels decreased over the 3 in-session measurements, but with different dynamics between the groups, because the group × time interaction was significant (P<.0001). Indeed, after the session, blood glucose levels rose more rapidly in the HIIT group (fig 5).

Fig 5.

Blood glucose levels at the beginning of the session, at the end of the session, and after 15 minutes of recovery in participants with diabetes in the control (red n=9) and HIIT groups (blue n=9). As blood glucose levels were recorded at 3 time points in participants with diabetes, changes were compared between groups using a mixed linear model with group × time interaction, and adjusting on initial level and compliance. A bootstrap method was used to calculate the P values.

Discussion

This study demonstrated the feasibility of HIIT program for people with CF. Compliance with the physical activity sessions was excellent (>90%), and the sessions were well tolerated. HIIT appeared to be as effective as conventional training for most participants and was shorter to perform (10min vs 30min). However, it does not appear to be suitable for those with severe respiratory impairment, and also those who are ineligible for modulator therapy.

Although we included 62 individuals, fewer than the planned 100, this cohort is relatively large compared with other studies regarding HIIT, which included 1-24 participants.^9,11,14,30 Recruitment was impacted by the COVID-19 pandemic.³¹ In addition, with the expanded indications for CFTR modulator therapy,³² most participants were receiving this treatment, which can substantially improve CF condition,^33,34 potentially reducing the perceived need for intensive physical rehabilitation.

Few studies have assessed HIIT in people with CF. Although HIIT improves lung function, exercise capacity, and quality of life in chronic obstructive pulmonary disease,³⁵ its effects on cardiorespiratory fitness and functional outcomes appear comparable with those of continuous training.^13,36 In our study, HIIT did not provide additional clinical benefits over conventional rehabilitation in terms of 6MWT distance, quadriceps strength, or FEV1, likely due to the short duration of the program. Longer interventions may yield different results, as a 12-week outpatient HIIT program demonstrated improvements in exercise capacity, oxygen saturation, and dyspnea compared with continuous exercise, although gains in 6MWT distance were similar between methods.¹¹ These findings suggest that HIIT should be further evaluated over extended outpatient programs after initial hospital-based implementation for safety assessment.

Anxiety and quality of life improved in both groups, with a greater reduction in perceived treatment burden in the HIIT group, suggesting that high-intensity training may enhance acceptance of treatment constraints. However, enjoyment measured by PACES was significantly lower in the HIIT group, contrary to previous findings^30,37 and despite positive changes in physical activity perception assessed by STARFISH. Participants reported frustration during the low-intensity sessions scheduled between HIIT sessions, perceiving them as unproductive despite their recognized role in preventing overtraining and injury.³⁸ Although elevated CPK levels have been reported in clinical trials of CFTR modulators such as IVACAFTOR/TEZACAFTOR/ELEXACAFTOR, typically transient and asymptomatic and sometimes associated with physical activity,³⁹ we observed no increase in CPK in either training group, regardless of modulator therapy. These findings support good muscular tolerance of exercise in patients receiving IVA/TEZ/ELX under real-life conditions.

Our study confirms previous findings³⁷ that HIIT is compatible with diabetes, with no symptomatic hypoglycemia and better glycemic stability during exercise. Modulator therapy can cause considerable weight gain and metabolic changes, including disruptions in lipid and carbohydrate profile.^40,41 The value of HIIT in these cases could be questioned. However, 2 recent meta-analyses show that HIIT is beneficial in people with obesity⁴² and type 2 diabetes⁴³ for improving body composition and lipid and carbohydrate profiles.

Given the clinical improvements with modulator therapy, increased physical capacity, and new sociologic concerns (participation in sports clubs or competitions), professional practices must be reviewed. It is crucial to develop treatment approaches tailored to these new profiles. HIIT could address this need, given its good tolerance and acceptance, although adjustments may be needed in more severely affected individuals. HIIT may also contribute to reconditioning individuals, helping them regain fitness for sports.

Study limitations

This study was unblinded and the duration of rehabilitation was relatively short.

Second, maximal heart rate or heart rate reserve was not accounted for in the prescription of HIIT session intensity. Indeed, Intensity was established using calculated equation and adjusted to monitored clinical vital signs, that is routinely performed in clinical practice.

In addition, the 6MWT was used to evaluate the effectiveness of HIIT instead of peak oxygen uptake (VO₂peak) using cardiopulmonary exercise testing that is considered as the criterion standard in CF.⁴⁴ However, according to the European Cystic Fibrosis Society Exercise Working Group, cardiopulmonary exercise testing cannot always be performed in routine clinical practice (eg, because of lack of time, personnel, equipment, or because of infection-control constraints). It should also be noted that VO₂max is difficult to improve and is not necessarily the most sensitive marker for detecting changes in physical fitness after exercise-training interventions; in some circumstances, alternative field tests may therefore be more appropriate.⁴⁵

Field-based exercise tests, including walking tests, offer a practical means of assessing aerobic capacity in individuals with CF. In Europe, the 6MWT remains one of the most commonly used functional tests in clinical practice. Recent data in adults suggest that the 6MWT is sensitive to changes after initiation of CFTR modulator therapy.⁴⁶ Nonetheless, to our knowledge, no minimal clinically important difference has been established for the 6MWT in adults with CF, and certainly none exists in the era of CFTR modulators.

Finally, the sample size estimation was heuristic; however, this was a pilot study, and the number of participants included was considerably larger than in previous studies. Subgroup analyses would require larger and more comprehensive investigations.

Conclusion

This study demonstrated the feasibility and tolerance of a 3-week, inpatient HIIT program in people with CF, with clinical benefits similar to conventional rehabilitation. The results reaffirm the effectiveness of physical training, including for people on CFTR modulator therapy.

Suppliers

a. Cycle ergometer (Bike Reha EN-Cardio (230v/5hz) manufacturer ENRAF-NONIUS, Rotterdam, Netherlands).

b. Bioelectrical impedance analysis (Quadscan 4000, Bodystat Ldt).

c. microFET2; HogganScientific LLC.

d. FreeStyle Libre Glycemic Holter; Abbott.

e. Statistical Software R version 4.3.0.

Disclosures

None.