Impact of nocturnal oxygen saturation on sleep quality and exercise performance in adult Fontan patients

Abstract

Introduction

A decreased peripheral oxygen saturation (SpO₂) is common in Fontan patients. We aimed to determine whether these patients experience greater nocturnal desaturation (ΔSpO_2,noct), and to which extent this may impact sleep quality and exercise performance, compared to healthy controls.

Methods

16 adult Fontan patients and 16 healthy, individually matched controls completed questionnaires on quality of life, sleep quality, and daytime sleepiness, performed measurements of lung function and diffusion capacity (DLCO), and cycled in a cardiopulmonary exercise test to volitional exhaustion. Sleep quality was assessed with a wrist-worn actigraph and nocturnal SpO₂ was measured with a pulse oximeter during six nights.

Results

Fontan patients experienced greater ΔSpO_2,noct (−3.4 ± 0.8%-points, n = 14) compared to their controls (−2.5 ± 0.6%-points, n = 14, p = 0.001). In patients, ΔSpO_2,noct correlated with baseline SpO₂, DLCO, and exercise performance. Fontan patients had a longer sleep onset latency (22.8 ± 9.6 vs. 10.8 ± 6.0 min, p < 0.001) and lower self-perceived physical functioning (91 ± 7 vs. 100 ± 0 points, p < 0.001), but similar subjective sleep quality (4.2 ± 1.5 vs. 3.4 ± 1.4 points, p = 0.121) and daytime sleepiness (6.1 ± 4.3 vs. 6.3 ± 2.8 points, p = 0.884) compared to controls. In patients, daytime sleepiness correlated with the total number of desaturation events (r = 0.74, p = 0.003).

Conclusion

Indices of nocturnal SpO₂ were related to daytime sleepiness and reduced exercise performance, and may therefore be a contributing factor to exercise limitation in adult Fontan patients. Enhancing nocturnal SpO₂ may thus be a promising target to improve daytime sleepiness and exercise performance in adult Fontan patients.

Highlights

• •Larger nocturnal desaturation in Fontan patients compared to matched controls.
• •Nocturnal desaturation was associated with reduced exercise capacity.
• •In Fontan patients, desaturation events correlated with daytime sleepiness.
• •Interventions to mitigate nocturnal desaturation might improve exercise capacity.

1. Introduction

Since its first description in 1971 [1], Fontan palliation has undergone numerous modifications [[2], [3], [4]] and has evolved as the standard palliation operation for patients born with univentricular physiology, leading to markedly improved survival rates among affected patients [5]. In brief, the principle of the Fontan operation involves separation of systemic and pulmonary circulation by direct connection of systemic venous return to pulmonary arteries, leading to passive pulmonary blood flow. Its main advantages are a reduction of the volume overload of the single ventricle and alleviation of chronic cyanosis [6]. Decreased peripheral oxygen saturation (SpO₂) at rest and during exercise is, however, commonly found, caused by Fontan fenestrations, veno-venous collaterals and potentially pulmonary factors [7]. Lower SpO₂, in turn, was shown to contribute to these patients' reduced exercise capacity [8]. To which extent a further decrease in SpO₂ during sleep (nocturnal desaturation [ΔSpO_2,noct]) affects sleep quality, daytime sleepiness and exercise performance, has not yet been assessed in detail within this patient group. This may, however, be of importance, since good sleep quality has already been associated with better athletic performance, health, and quality of life in healthy individuals [[9], [10], [11]].

Some sleep-related aspects have already been shown to be altered in Fontan patients, i.e. sleep onset latency was increased in one population [12], while in others, sleep disturbance was identified in the context of obstructive sleep apnea, sleep disordered breathing (SDB), and nocturnal hypoxemia [13,14]. Furthermore, subjective sleep disturbance was found to be associated with reduced quality of life [15], though subjective and objective measures of sleep do not always correspond [16].

The aim of the present study was therefore to assess: a) to which extent indices of nocturnal SpO₂ affects Fontan patients' sleep quality, daytime sleepiness, and exercise capacity; and b) whether these relations are similar or different to healthy, individually matched controls when assessed via ambulatory monitoring at home.

2. Methods

2.1. Ethical approval

The study was approved by the Cantonal Ethics Committee of Zurich (ID 2019-01320) and registered on clinicaltrials.gov (NCT04151004). The study followed the ethical guidelines of the Declaration of Helsinki for research involving human subjects.

2.2. Participants

16 adult Fontan patients (all in New York Health Association [NYHA] class I; two with surgical fenestrations) and 16 healthy matched controls participated in this study. Controls were matched for age, sex, body mass index and self-reported physical activity levels. Baseline characteristics are shown in Table 1. All patients were in clinically stable condition and none had contraindications to exercise.

Table 1.

Participant characteristics.

Descriptive characteristics	Fontan			Control
Sex	9m	:	7f	9m	:	7f
Age, yr	28.3	±	7.0	28.6	±	7.8
Height, cm	169.8	±	10.1	171.1	±	8.8
Weight, kg	64.6	±	8.9	65.9	±	8.8
Body mass index, kg ⋅ m−2	22.4	±	1.8	22.5	±	2.0
Physical activity, MET-min ⋅ wk−1	2570 [6426]			2430 [1474]
CPET
$\dot{\mathrm{V}}$O2peak, ml ⋅ kg−1 ⋅ min−1	29	±	5	46	±	9 ***
$\dot{\mathrm{V}}$O2peak, % predicted	67	±	11	108	±	16 ***
Spirometry & pulmonary function
FVC, l	4.0	±	0.8	5.2	±	1.0 **
FVC, % predicted	87	±	9	109	±	9 ***
FEV1, l	3.3	±	0.7	4.3	±	0.8 **
FEV1, % predicted	85	±	11	108	±	10 ***
FEV1/FVC, %	82	±	6	83	±	6
FEV1/FVC, % predicted	97	±	7	98	±	7
DLCO, ml ⋅ min−1 ⋅ mmHg−1	17.8	±	3.7	28.9	±	7.0 ***
DLCO, % predicted	66	±	15	103	±	12 ***
Respiratory muscle strength
MIP, cmH2O	100	±	36	117	±	28
MIP, % predicted	105	±	38	124	±	24
MEP, cmH2O	119	±	57	163	±	43 *
MEP, % predicted	91	±	34	131	±	33 **

Clinical	Number (%) of patients
Fontan type
Atriopulmonary	7	(44)
Lateral tunnel Fontan	1	(6)
Extra-cardiac Fontan	6	(38)
Norwood palliation	2	(13)
Predominant ventricular morphology
Left ventricular type	13	(81)
Right ventricular type	3	(19)
Previous pacemaker implantation	4	(25)
Fenestrations	2	(13)
Cardiac medications
Oral anticoagulation	9	(56)
ACE/AT2 inhibitors	5	(31)
Beta blocker	5	(31)
Low dose aspirin	5	(31)
Dronedarone	2	(13)
Flecainide	1	(6)

Values are means ± SD, except for physical activity that is given as the median [interquartile range]. MET, metabolic equivalent of task; CPET, cardiopulmonary exercise test; $\dot{\mathrm{V}}$O_2peak, peak oxygen uptake; FVC, forced vital capacity; FEV₁, forced expiratory volume in 1 s; DLCO, diffusion capacity of the lung for carbon monoxide; MIP, maximal inspiratory pressure; MEP, maximal expiratory pressure; ACE, Angiotensin-converting-enzyme; AT2, Angiotensin 2. Predictions for $\dot{\mathrm{V}}$O_2peak are from Wagner et al. [17], FVC and FEV₁ predictions are from Quanjer et al. [18], DLCO predictions are from Stanojevic et al. [19,20], and MIP and MEP predictions are from Wilson et al. [21]. High interquartile range for physical activity level of the patients results from work-related high scores in slow walking in four patients. *p < 0.05, **p < 0.01, ***p < 0.001 significantly different between groups.

Patients and controls refrained from intense exercise 48h before test days and did not exercise at any intensity 24h prior to test days. They slept at least 7h the two nights before testing, refrained from alcohol the day before and on the day of a test, and also refrained from caffeinated food or drink on the test day. Participants kept their training constant throughout the study period.

2.3. Experimental overview

Upon arrival, all study equipment was shown and explained to the participants and questions were discussed before participants provided their written informed consent. Then, participants completed several questionnaires which was followed by lung function testing. Subsequently, participants performed a symptom-limited cardiopulmonary exercise test (CPET) to volitional exhaustion. Finally, participants were shown how to wear two wrist-worn devices, an actigraph and a pulse oximeter which they wore for the following six nights.

2.4. Questionnaires

Participants filled out a questionnaire on their daily physical activities (International Physical Activity Questionnaire, IPAQ; [22]), one assessing physical and mental aspects of their quality of life (SF-36; [23]), one on daytime sleepiness (Epworth Sleepiness Scale, ESS; [24]), and one on sleep quality (Pittsburgh Sleep Quality Index, PSQI; [25]).

2.5. Lung function and respiratory muscle strength

Spirometry and lung diffusion capacity for carbon monoxide (DLCO) were performed with a commercially available pulmonary function system (Quark PFT/Q-Box, Cosmed, Rome, Italy) according to guidelines [26,27] and absolute values were subsequently calculated as percent predicted [[18], [19], [20]]. Respiratory muscle strength was assessed via maximal inspiratory pressure (MIP) and maximal expiratory pressure (MEP) assessments according to current guidelines [28,29] from residual volume and total lung capacity, respectively, using a respiratory pressure meter (RP Check, MD Diagnostics LTC., Kent, United Kingdom). Values were calculated as percent predicted according to Wilson et al. [21].

2.6. Cardiopulmonary exercise test

All participants performed a CPET on an electronically braked cycle ergometer (Ergoline 900, Ergoline, Blitz, Germany). They were equipped with a face mask connected to a metabolic cart. The test was performed in an upright position. After a 3 min resting measure, the starting load for the patients was set at 20 W with a ramp-wise increase of 10, 15, or 20 W⋅min⁻¹ based on their previous exercise tests to reach peak exercise between 8 and 12 min. Controls started at an individualized higher load to likewise reach peak exercise between 8 and 12 min. Participants kept cycling cadence between 60 and 80 rpm and the test was terminated either if the cycling cadence dropped below 60 rpm or the participants volitionally stopped despite verbal encouragement. Ventilation and gas exchange were measured breath by breath with the Oxycon pro (Oxycon pro, Jaeger, Höchberg, Germany) which was calibrated before each test. Patients’ heart rate and rhythm were recorded continuously and live monitored by a cardiologist throughout the whole test duration with a mobile 12-lead electrocardiogram device (MS-12 blue, Schiller AG, Switzerland). Peak oxygen consumption ($\dot{\mathrm{V}}$O_2peak) was defined as the highest 15 s moving average within the last minute of the examination. Reference values (L⋅min⁻¹ and ml⋅kg⁻¹⋅min⁻¹) were calculated according to Wagner et al. [17].

2.7. Sleep measures

Sleep variables were measured using wrist-worn actigraphy (Actiwatch score, Cambridge Neurotechnology, Cambridge, UK) and pulse oximetry (WristOx2, Nonin medical, Plymouth Minnesota, USA). Actigraphy and pulse oximetry data were recorded with a sampling frequency of 32 Hz and 1 Hz, respectively, and stored on the devices. Both devices were worn on the non-dominant arm. Participants were requested to go to bed and get up at their usual times, and to behave (eat, drink, perform physical activity) as they normally do. They were asked to set an event marker on the actigraph when they intended to fall asleep, and again when they woke up in the morning. These markers were then used to define the time in bed start and end. Additionally, each individual reported “in bed time” and “out of bed time” in a sleep diary as a complement to the actigraph-derived analyses. For actigraphy analysis, data were downloaded as 60-s epochs to the Cambridge Neurotechnology Ltd. analysis software (Version 5.32, Cambridge Neurotechnology, Cambridge, UK). Time in bed, sleep duration, sleep onset latency, sleep efficiency and fragmentation index were used for further analyses. For pulse oximetry analysis, the nVision software was used (nVision, Nonin medical, Plymouth Minnesota, USA). Baseline SpO₂, lowest nocturnal SpO₂, ΔSpO_2,noct (lowest - baseline SpO₂), total number of desaturation events (defined as a drop in SpO₂ by at least 4% for a minimum duration of 10 s), total duration desaturated, average duration of desaturation events, and Desaturation Index (number of desaturation events per hour) during the night were used for further analyses. Whole night sleep data was averaged over the available number of nights (all patients recorded for at least three nights, except for one patient with technical problems with the SpO₂ sensor, where data was available for one night only).

2.8. Statistics

Distribution of the data was first checked for normality using the Shapiro-Wilk test. Between-group comparisons of patient characteristics, variables of lung function, CPET, sleep, SpO₂ and questionnaires were performed using unpaired t-tests. If normality was not present, the Mann-Whitney-U test was used. Pearson or Spearman correlation coefficients (chosen dependent on the presence of normality) were calculated to explore the relationships between ΔSpO_2,noct and the baseline SpO₂, DLCO, respiratory muscle strength, desaturation during exercise and exercise performance of Fontan patients and controls (n = 14 per group). Pearson or Spearman correlation coefficients were also calculated to quantify the relationships between subjective (ESS and PSQI) and device-derived sleep measures (actigraph and pulse oximetry), the relationship between the ESS and PSQI, as well as the relationship between general health and self-reported physical activity in both groups (n = 16). P < 0.05 was considered statistically significant. If not stated otherwise, all values are mean ± SD. All analyses were performed using GraphPad Prism 9 (GraphPad Software Inc., San Diego, CA, USA).

3. Results

3.1. Sleep variables and nocturnal saturation

Two patients (2m; age: 26.7 ± 5.7 yr, height: 184.8 ± 8.1 cm, weight: 70.6 ± 2.1 kg, physical activity: 5425 [1965] MET-min⋅wk⁻¹) forgot to wear the wrist-worn actigraph and pulse oximeter, and were thus excluded from these analyses along with their respective controls. Thus, sleep and nocturnal SpO₂ measures are given for 14 participants in each group (Table 2). The available number of night-time recordings with adequate data quality was similar between groups for actigraphy (5.6 ± 0.8 vs. 5.8 ± 0.4, p > 0.999) and pulse oximetry (4.4 ± 1.6 vs. 4.9 ± 1.3, p = 0.524). Fontan patients had a similar sleep duration and efficiency, but a longer sleep onset latency (Table 2). In Fontan patients, ΔSpO_2,noct was correlated with baseline SpO₂ (r = 0.54, 95% confidence interval [95% CI] = 0.02 – 0.83, p = 0.046), % predicted DLCO (r = 0.62, 95% CI = 0.13 – 0.87, p = 0.018), and % predicted $\dot{\mathrm{V}}$O_2peak (r = 0.71, 95% CI = 0.29 – 0.90, p = 0.004) (Fig. 1), but not in controls (baseline SpO₂: r = 0.50, 95% CI = −0.05 – 0.81, p = 0.072; % predicted DLCO: r = 0.15, 95% CI = −0.41 – 0.63, p = 0.601; % predicted $\dot{\mathrm{V}}$O_2peak: r = 0.31, 95% CI = −0.27 – 0.72, p = 0.285). ΔSpO_2,noct was not correlated with exercise-induced desaturation (r = 0.24, 95% CI = −0.33 – 0.68, p = 0.408), MIP (r = −0.13, 95% CI = −0.62 – 0.43, p = 0.657), or MEP (r = −0.18, 95% CI = −0.65 – 0.39, p = 0.262) in Fontan patients. In controls, no correlation was found between ΔSpO_2,noct and any actigraph-derived sleep variable, or respiratory muscle strength.

Table 2.

Sleep variables and nocturnal saturation.

	Fontan			Control
Sleep variables
Time in bed, min	484.8	±	46.2	501.0	±	55.2
Sleep duration, min	403.8	±	44.4	425.4	±	55.2
Sleep efficiency, %	83.3	±	3.4	85.1	±	5.7
Sleep onset latency, min	22.8	±	9.6	10.8	±	6.0 ***
Fragmentation Index	38.1	±	34.6	41.3	±	20.4
Nocturnal SpO2
Baseline SpO2, %	91.2	±	2.0	94.9	±	1.1 ***
Lowest SpO2, %	87.8	±	2.5	92.4	±	1.5 ***
ΔSpO2,noct, %-points	−3.4	±	0.8	−2.5	±	0.6 **
Duration below 88%, min	27.8	±	81.6	0.0	±	0.0 ***
Number of desaturation events	12.1	±	11.3	12.8	±	11.7
Average duration of desaturation events, s	50.3	±	16.8	43.8	±	20.6
Total duration of desaturation events, min	8.4	±	5.9	8.0	±	6.1
Desaturation Index, events ⋅ hour−1	1.6	±	1.4	1.8	±	1.5

Values are means ± SD. SpO₂, peripheral oxygen saturation; ΔSpO_2,noct, nocturnal desaturation. **p < 0.01, ***p < 0.001 significantly different between groups.

Fig. 1.

Correlation of nocturnal desaturation (ΔSpO_2,noct) with baseline peripheral oxygen saturation (SpO₂;left), diffusion capacity of the lung for carbon monoxide (DLCO; middle), and peak oxygen consumption ($\dot{\mathrm{V}}$O_2peak; right) within Fontan patients (n = 14) and controls (n = 14). Solid black line represents a regression line for the Fontan patients, while the dashed black line represents a regression line for the controls.

3.2. Subjective sleep quality and daytime sleepiness

PSQI global score was not different between Fontan patients (4.2 ± 1.5 points) and controls (3.4 ± 1.4 points, p = 0.121). Three patients and two controls had pathological PSQI scores (>5 points; [25]). PSQI global score was not related to any actigraph-derived sleep variable. Daytime sleepiness (ESS scores) was not different between Fontan patients (6.1 ± 4.3 pts) and controls (6.3 ± 2.8 pts, p = 0.884). Three patients and one control were symptomatic for daytime sleepiness with an ESS score ≥ 10 [24]. ESS scores correlated with the total number of desaturation events in Fontan patients (r = 0.74, 95% CI = 0.33 – 0.92, p = 0.003), but not in controls (r = 0.10, 95% CI = −0.47 – 0.61, p = 0.745).

3.3. Physical and mental aspects of quality of life

Fontan patients rated their physical functioning significantly lower than controls (91 ± 7 vs. 100 ± 0 points, p < 0.001) and there was a tendency towards perception of lower general health in patients compared to controls (77 ± 15 vs. 86 ± 10 points, p = 0.063). Higher general health rating correlated with higher scores of self-reported physical activity in Fontan patients (r = 0.64, 95% CI = 0.20 – 0.87, p = 0.009), but not in controls (r = −0.20, 95% CI= −0.64 – 0.34, p = 0.395).

4. Discussion

In this study, adult Fontan patients were compared with healthy, individually matched controls. Fontan patients were found to have a small, but significant ΔSpO_2,noct which was significantly larger compared to their matched controls. In patients but not in controls, ΔSpO_2,noct was associated with baseline SpO₂, DLCO, and exercise capacity, which were all significantly reduced compared to controls. This cohort of stable Fontan patients showed no evidence of SDB, similar to controls. Also, sleep duration, sleep efficiency, and subjectively perceived daytime sleepiness were similar between patients and controls. Only sleep onset latency was significantly increased in patients. Interestingly, the level of daytime sleepiness perceived by patients was associated with their total number of nighttime desaturation events.

The mild ΔSpO_2,noct of Fontan patients is in line with previous data in asymptomatic Fontan patients [13]. In the present study, ΔSpO_2,noct was significantly associated with the level of baseline SpO₂, DLCO, and exercise capacity. The association to baseline SpO₂ is not astonishing. During sleep, respiratory drive is reduced and consequently, also alveolar O₂ partial pressure, reducing driving pressure for O₂ diffusion and thus resulting in a reduction in SpO₂, in particular at levels around or below 90% where the sigmoid oxygen-hemoglobin dissociation curve becomes steeper. Thus, the lower baseline SpO₂ in these patients, the larger the drop in SpO₂ for any given decrease in arterial O₂ partial pressure. In healthy controls, this change is less prominent with a baseline SpO₂ of 95% in the present study, where the dissociation curve is still in the flat part. To which extent the significantly reduced DLCO contributes to this desaturation remains speculative since the measure of reduced DLCO may partially result from a reduced lung volume. As for healthy, also in Fontan patients, DLCO was shown to increase from sitting to supine position where lung perfusion is increased. Specifically, DLCO increased in Fontan patients by 52% from sitting to supine with a 48% increase in pulmonary capillary blood volume, indicating a preserved alveolar-capillary membrane diffusing capacity [30], and thus the reduced DLCO not being the main driver of the larger drop in SpO₂ during sleep. However, a limitation of the lung diffusion capacity cannot be fully excluded in the present cohort of Fontan patients. Also, shunting such as veno-venous collaterals, commonly found in Fontan patients [31], and fenestrations might play a role for the hypoxemic state and ΔSpO_2,noct in Fontan patients. In the present cohort, however, shunting may not have been a primary driver of desaturation given the lack of relationship between ΔSpO_2,noct and exercise-induced desaturation that would allude to such a case. Indeed, two patients have documented fenestrations, but both displayed the same ΔSpO_2,noct as the average of the cohort (3.4%) indicating that the presence of such fenestrations was not confounding the relationship seen between ΔSpO_2,noct and exercise capacity. Finally, respiratory muscle weakness, commonly seen in Fontan patients [32], may also contribute to ΔSpO_2,noct. However, the present cohort displayed normal respiratory muscle strength, and no associations were present between both MIP and MEP with ΔSpO_2,noct.

The association of ΔSpO_2,noct with exercise capacity may seem astonishing on first sight. Though speculative, the association between ΔSpO_2,noct and reduced exercise capacity might be related to a negative impact of the low SpO₂ on the systemic vasculature. With the nocturnal desaturations, SpO₂ might reach levels that promote inflammation and oxidative damage as shown during obstructive sleep apnea [33,34], resulting in impaired vascular endothelial function. Impaired endothelium-dependent vasodilation in Fontan patients was shown to play a crucial role in the blood supply to the working skeletal muscle and was related to reduced functional capacity [35,36]. Besides this negative impact on blood supply to peripheral muscles, endothelial dysfunction has recently been reported to correlate with pulmonary vascular disease [37,38] with negative impact on the pulmonary vascular reserve in Fontan patients, which was shown to contribute to an impaired exercise capacity in Fontan patients [38].

In the present cohort, we observed no evidence of SDB, which is known to affect sleep quality [39]. This is in line with findings by Nanayakkara et al. [13], but not with others [14,40]. Increased prevalence of SDB was found to be associated with NYHA class ≥ II and body mass index ≥ 25 [40], which both was not the case in our cohort and represents a selection bias of our sample. Notably, in patients with SDB, targeted therapies such as continuous positive airway pressure should be explored.

Our finding of similar sleep duration and efficiency, but higher sleep onset latency compared with controls is consistent with the recent finding in pediatric Fontan patients in stable medical condition [12]. Interestingly, these authors found no overall improvement of any sleep variable after a 12-week aerobic endurance training program, but patients with low sleep efficiency and long sleep onset latency at baseline significantly improved both. Given the similar sleep onset latencies of our Fontan patients with this latter subgroup, one could speculate that sleep of our patients could similarly benefit from an endurance training protocol, supported by the positive association between physical activity and sleep in healthy adults [41]. Also, physical activity is known to promote general health [42], an association also seen in our cohort of Fontan patients.

Preserved overall sleep quality was reflected in actigraph-derived and subjective sleep measures, with PSQI and ESS scores being in the normal range [24,25] and with no difference compared to controls. Although it may seem obvious that a good sleep quality results in a low daytime sleepiness, subjective sleep quality only tended to correlate with daytime sleepiness, which is in line with previous findings [43]. Given the time and costs required for polysomnography, the ESS was proposed as an initial approach to screen for sleep quality and disturbance of Fontan patients [14]. However, this approach may be suboptimal in face of our and previous findings [44].

5. Limitations

All Fontan patients were in functional NYHA class I. Whether our findings can be extrapolated to patients with more advanced Fontan-failure remains to be determined. We used wrist-worn actigraphs and pulse oximeters to objectively assess sleep variables and nocturnal SpO₂ instead of polysomnography in a sleep laboratory regarded as the gold standard for sleep analyses. However, high agreements between wrist-worn devices and traditional polysomnography at home or in sleep laboratories have been demonstrated [[45], [46], [47], [48]]. Given the relatively small sample size, present findings need to be confirmed by studies with a larger number of Fontan patients and controls. Despite the small number of participants, the correlation between ΔSpO_2,noct and the reduced exercise capacity was relatively strong. However, the clinical relevance of this statistically significant relationship requires further investigations, but may open a new field for optimization of care in adult Fontan patients, likely one of the most vulnerable and complex patient cohorts in adult congenital heart disease.

6. Conclusion

Compared to carefully selected, individually matched controls, asymptomatic Fontan patients displayed more pronounced ΔSpO_2,noct. This was associated with their (already low) baseline SpO₂, lung diffusion capacity, and reduced exercise performance. Subjective sleep quality and daytime sleepiness were similar between groups, but in patients, the latter was related to the number of nocturnal desaturation events. ΔSpO_2,noct is a common phenomenon in asymptomatic Fontan patients caused by a multitude of factors including respiratory weakness and shunts. Given that indices of nocturnal SpO₂ might be a factor contributing to the exercise limitation and daytime sleepiness of adult Fontan patients, understanding these specific mechanisms for tailored interventions to enhance nocturnal SpO₂ in home settings may be promising to improve quality of life in this patient cohort.

CRediT authorship contribution statement

Patrick Schön: Writing – review & editing, Writing – original draft, Visualization, Validation, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Kyle G.P.J.M. Boyle: Writing – review & editing, Investigation, Formal analysis. Fernando G. Beltrami: Writing – review & editing, Investigation. Matthias Greutmann: Writing – review & editing, Methodology, Investigation, Conceptualization, Funding acquisition. Christina M. Spengler: Writing – review & editing, Validation, Resources, Methodology, Funding acquisition, Conceptualization.

Ethics declaration

Written informed consent to take part in the study and to publish the article has been obtained from all participants or their legal representatives. The privacy rights of participants have been observed.

This study was performed in compliance with relevant laws, regulatory frameworks and guidelines where the research took place. This study was approved by the Cantonal Ethics Committee of Zurich. (Approval No. ID 2019-01320)

The results of this clinical trial and any associated work have been posted in a registry. This clinical trial was registered with number NCT04151004 (clinicaltrials.gov).

Funding

This study was supported by the Swiss National Science Foundation (grant no. 32003B_146481).

Declaration of competing interests

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Christina Spengler reports financial support was provided by Swiss National Science Foundation. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.