Journal Information
Vol. 29. Issue 6.
Pages 546-549 (November - December 2023)
Share
Share
Download PDF
More article options
Vol. 29. Issue 6.
Pages 546-549 (November - December 2023)
Letter to the Editor
Full text access
Poor sleep quality and nocturnal home noninvasive ventilation: Prevalence, risk factors and impact
Visits
1891
J. Suttera,b, A. Cuveliera,c, R. Lukaszewicza, J. Marisa, I. Arnulfb,d, T. Similowskie,b, M. Patouta,b,c,e,
Corresponding author
maxime.patout@aphp.fr

Corresponding author at: AP-HP, Groupe Hospitalier Universitaire APHP-Sorbonne Université, site Pitié-Salpêtrière, Service des Pathologies du Sommeil (Département R3S) Paris, France.
a Service de pneumologie, oncologie thoracique et soins intensifs respiratoires, CHU de Rouen, Université de Rouen, Rouen, France
b AP-HP, Groupe Hospitalier Universitaire APHP-Sorbonne Université, site Pitié-Salpêtrière, Service des Pathologies du Sommeil (Département R3S), F-75013 Paris, France
c Normandie Université, UNIROUEN, UPRES EA3830 (GRHV), Haute Normandie Recherche et innovation biomédicale, Rouen, France
d Institut du cerveau et de la moelle - ICM, Inserm U 1127, CNRS UMR 7225, UPMC, Paris, France
e Sorbonne Université, INSERM, UMRS1158 Neurophysiologie Respiratoire Expérimentale et Clinique, F-75005 Paris, France
This item has received
Article information
Full Text
Bibliography
Download PDF
Statistics
Tables (2)
Table 1. Clinical characteristics according to group (COPD: chronic obstructive pulmonary disease; IQR: interquartile range; mMRC: modified medical research council; OSA: obstructive sleep apnea; NIV: non-invasive ventilation; PSQI: Pittsburgh sleep quality index).
Table 2. NIV settings, effect on PaCO2, tolerance, and data from built-in software according to group (BUR: back-up respiratory rate; EPAP: expiratory positive airway pressure; IQR: interquartile range; IPAP: inspiratory positive airway pressure; NIV: noninvasive ventilation; PaCO2: partial arterial pressure of carbon dioxide; PVA: patient-ventilator asynchrony).
Show moreShow less
Full Text

Chronic respiratory diseases deteriorate sleep, impacting total sleep time, sleep efficiency, and sleep architecture.1 Nocturnal noninvasive ventilation (NIV) can restore sleep quality, as documented by polysomnography (PSG).1 Therefore, current guidelines stipulate that NIV should be initiated under PSG guidance.2 This is not done in practice because of resources limitation. Some evidence supports the hypothesis that, in the absence of PSG guidance, sleep quality can be poor under NIV.3 To further explore this hypothesis, we prospectively administered the Pittsburgh sleep quality index (PSQI)4 to unselected patients placed under NIV without PSG guidance. In addition to prevalence data, we aimed to evaluate risk factors of poor sleep quality under NIV and to assess its impact on quality of life (QoL).

Over 90 days, we consecutively enrolled patients previously established on nocturnal home NIV without PSG guidance, clinically stable for at least 3 months and attending our NIV clinic for a scheduled follow-up. Study received ethics approval (CERNI, Rouen, E2018–06).

Patients answered the PSQI, the Severe Respiratory Insufficiency questionnaire (SRI; 0: poorest QoL; 100: best), and Epworth Sleepiness Scale. NIV side effects were graded from 0 –absent– to 4 –unbearable– according to an in-house questionnaire (eye irritation, rhinorrhea, dry mouth, hoarseness, bloating, mask or harness-related pain, pressure sores, leaks, deventilation dyspnea and perceived patient-ventilator asynchrony). NIV usage was assessed from 30-days ventilator logs. Leaks were considered abnormal if >2 L/min for Resmed® ventilators, >10 L/min for Philips Respironics® ventilators. A PSQI above 5 defined a “poor sleep quality group”,4 mirrored by a “good sleep quality” one.

Results are expressed as number and percentages, median and interquartile range (IQR). Mann-Whitney's U test, the chi-square test and Spearman's test were used for comparisons and correlations. Multivariate analysis (binomial logistic regression) incorporated clinically relevant variables with p <0.15 on univariate analysis. All tests were two-sided with 0.05 significance threshold (GraphPad Prism 6® and IBM SPSS® v26.0).

One hundred and fifty-nine patients attended the clinic during the study period; 32 declined to participate or could not be included (mostly because of staff unavailability), leaving 127 for analysis. The “poor sleep quality” group comprised 64 patients (50.5%) (Table 1). NIV indications, settings, effect on PaCO2 and adherence did not differ between groups (Table 2). Poor sleep quality patients more frequently reported grade 3–4 mouth dryness (59 vs 40% s, p = 0.033) and overnight NIV interruptions: 0.3 [0.0 – 1.1] breaks/night vs 0.1 [0.0 – 0.3], p = 0.026). Multivariate analysis evidenced 3 variables associated with poor sleep quality: benzodiazepine use (OR= 6.83 [1.53–30.65], p = 0.012), airway secretions (OR= 4.66 [1.06–20.43], p = 0.041), and abnormal leaks (OR= 4.52 [1.25–16.39], p = 0.022). SRI-evaluated QoL was lower in poor sleep quality patients (50.0 [39.7 – 62.5] vs. 62.0 [50.0 – 76.3] out of 100, p<0.001). PSQI inversely correlated with SRI (rho = −0.410, p < 0.001).

Table 1.

Clinical characteristics according to group (COPD: chronic obstructive pulmonary disease; IQR: interquartile range; mMRC: modified medical research council; OSA: obstructive sleep apnea; NIV: non-invasive ventilation; PSQI: Pittsburgh sleep quality index).

  Patients with good sleep quality (n = 63)  Patients with poor sleep quality (n = 64)   
  Median [IQR] / n (%)  Median [IQR] / n (%) 
  Demographic data 
Age at NIV initiation (years)  63 [54–68]  63.5 [53–74]  0.308 
Time since NIV initiation (months)  36 [12–82]  34 [7–71]  0.215 
Gender (male)  35 (56%)  25 (39%)  0.076 
Body mass index (kg/m232.4 [25.6–40.7]  30.8 [24.1– 41.5]  0.250 
  Comorbidities 
Apnea-hypopnea index before NIV setup (/h)  17 [7–44.25]  17 [6.75–34]  0.665 
Active or former smoker  37 (79%)  31 (74%)  0.624 
Ischemic heart disease  7 (11%)  7 (11%)  1.000 
Depression  12 (20%)  21 (33%)  0.105 
Mental illness  0 (0%)  3 (5%)  0.240 
  Drug intake 
Benzodiazepines and “Z-drugs”  10 (16%)  25 (40%)  0.005 
Antipsychotics  3 (5%)  12 (19%)  0.025 
  Underlying respiratory disease 
Neuromuscular disease  15 (25%)  17 (27%)  0.972
Obesity hypoventilation syndrome  16 (27%)  16 (25%) 
COPD-OSA  17 (27%)  14 (25%) 
COPD  11 (17%)  12 (19%) 
Chest wall disease  4 (6%)  5 (8%) 
  Daytime symptoms 
Airway secretions (yes)  4 (8%)  26 (41%)  <0.001 
Dyspnea mMRC ≥ 2  40 (64%)  51 (80%)  0.050 
Chronic cough (yes)  22 (36%)  26 (41%)  0.586 
Self-reported daytime sleepiness (yes)  16 (25%)  30 (47%)  0.014 
Epworth Sleepiness Scale (/24)  3 [1–6]  5 [2–9]  0.001 
  Sleep 
PSQI (/28)  3 (1–5)  10 (6–20)  <0.001 
Reported sleep latency (min)  10 [5–15]  20 [5–60]  0.062 
Reported nighttime awakenings  1 [0.5–2]  2 [1–3]  0.002 
Reported snoring (≥ occasional)  13 (21%)  24 (38%)  0.037 
Unrefreshing sleep (yes)  9 (16%)  19 (29%)  0.036 
Daytime nap (yes)  39 (61%)  24 (38%)  0.010 
Table 2.

NIV settings, effect on PaCO2, tolerance, and data from built-in software according to group (BUR: back-up respiratory rate; EPAP: expiratory positive airway pressure; IQR: interquartile range; IPAP: inspiratory positive airway pressure; NIV: noninvasive ventilation; PaCO2: partial arterial pressure of carbon dioxide; PVA: patient-ventilator asynchrony).

  Patients with good sleep quality (n = 63)  Patients with poor sleep quality (n = 64)   
  Median [IQR] / n (%)  Median [IQR] / n (%) 
  NIV settings 
Spontaneous timed mode  52 (84%)  55 (87%)  0.620 
IPAP (cmH2O)  19 [16 – 22]  20 [16 – 22]  0.862 
EPAP (cmH2O)  6 [5 – 9]  6 [4 – 8]  0.083 
BUR (/min)  14 [12 – 14]  14 [12 – 14]  0.801 
Full face mask (yes)  55 (87%)  51 (80%)  0.248 
  Arterial blood gasses 
PaCO2 during unsupported breathing (kPa)  6.02 [5.48 – 6.82]  6.21 [5.67 – 7.07]  0.437 
PaCO2 on NIV (kPa)  5.36 [4.71 – 6.11]  5.76 [4.99 – 6.51]  0.206 
Change in PaCO2 after 1 hour on NIV (kPa)  −0.51 [−0.16 −1.24]  −0.58 [−0.22 −1.15]  0.649 
  NIV tolerance 
Dry mouth ≥ moderate  25 (40%)  38 (59%)  0.033 
Perceived leaks ≥ occasional  21 (33%)  32 (50%)  0.072 
Persistent skin damage related to mask  9 (15%)  9 (14%)  1.000 
Mask-related pain ≥ moderate  3 (5%)  5 (8%)  0.718 
Self-reported PVA ≥ occasional  3 (5%)  6 (9%)  0.492 
Deventilation dyspnea ≥ moderate  3 (5%)  9 (14%)  0.073 
  Data from NIV built-in software 
Adherence (h/day)  8.2 [7.3 – 9.4]  7.7 [5.9 – 9.7]  0.309 
Adherence > 4 h/night  97 [94 – 100]  96 [77 – 100]  0.093 
Number of NIV interruptions overnight  0.1 [0 – 0.3]  0.3 [0 – 1.1]  0.026 
Abnormal leaks (n)  22 (51.2%)  34 (66.7%)  0.144 
Expired tidal volume (ml)  520 [365 – 602]  464 [368 – 653]  0.813 
Respiratory rate (/min)  16.2 [15 – 18]  16 [14 – 18.3]  0.800 
Percentage of triggered breaths  54 [22 – 81]  50 [20 – 78]  0.641 
Estimated residual events (/hour)  2 [0.8 – 5]  2.5 [1.138 – 5.05]  0.456 

Patients receiving benzodiazepines more often had a PSQI above 5 (71 vs. 41%, p = 0.004), had been on NIV for shorter durations (18 [4–50] months vs. 41 [12 – 83], p = 0.014), had poorer daytime PaCO2 control (6.39 [5.89 – 6.99] kPa vs. 6.09 [5.45 – 6.89], p = 0.038), and lower SRI scores (50 [38 – 64] vs. 59 [47 – 72], p = 0.030).

In this real-life study, half of a cohort of patients under long-term NIV initiated without PSG guidance reported poor sleep quality according to the PSQI, irrespective of NIV indication. This was associated with lower QoL, in line with literature.5

The three variables independently associated with poor sleep quality in this study can all be subject to medical intervention . Our data emphasize the known importance of controlling leaks during home nocturnal NIV. Indeed, leaks can fragment sleep and promote mouth dryness, as in our poor sleep quality group. Our observations also highlight the known importance of trying to control airway secretions in NIV patients (e.g., in-exsufflation in neuromuscular patients). Interestingly, 27% of our patients received benzodiazepines or “Z-drugs” that are contra-indicated in chronic respiratory failure where they heighten the risks of respiratory related hospitalization.6 They also increase upper airway resistance during sleep,7 which is deleterious during nocturnal NIV. NIV which may be inappropriately viewed as a “safety net” against deleterious effects of benzodiazepines on breathing control. We could not ascertain the motivation and timing of benzodiazepine prescriptions and causality between benzodiazepines and poor sleep quality cannot be inferred. Nevertheless, our observations call for a reminder that benzodiazepines are inadvisable in chronic respiratory insufficiency: their prescription in this context should systematically be questioned.

We acknowledge study limitations (single-center, observational design, lack of evaluation of the effect of NIV initiation on sleep quality) and the need for larger trials. However, we propose that evaluation of NIV efficacy should not be confined to currently recommended “hard” criteria (e.g. arterial carbon dioxide), but should incorporate the easy and costless evaluation of sleep quality (e.g. PSQI) and QoL (e.g. SRI) that should trigger more in-depth assessments.

Data availability statement

The data that supports the findings of this study are available from the corresponding author upon reasonable request.

Authors contribution

JS, AC, IA, TS, MP: conception, acquisition, analysis, interpretation, drafting the work, revising critically.

RL, JM: acquisition, interpretation, revising critically.

References
[1]
A.J. Piper, D. Wang, B.J. Yee, et al.
Randomised trial of CPAP vs bilevel support in the treatment of obesity hypoventilation syndrome without severe nocturnal desaturation.
Thorax, 63 (2008), pp. 395-401
[2]
C.A. Kushida, M.R. Littner, T. Morgenthaler, et al.
Practice parameters for the indications for polysomnography and related procedures: an update for 2005.
Sleep, 28 (2005), pp. 499-521
[3]
J. Read, A. Simonds, M. Kinali, et al.
Sleep and well-being in young men with neuromuscular disorders receiving non-invasive ventilation and their carers.
Neuromuscul Disord, 20 (2010), pp. 458-463
[4]
D.J. Buysse, C.F. Reynolds 3rd, T.H. Monk, et al.
The Pittsburgh sleep quality index: a new instrument for psychiatric practice and research.
Psychiatry Res, 28 (1989), pp. 193-213
[5]
P. Totterdell, S. Reynolds, B. Parkinson, et al.
Associations of sleep with everyday mood, minor symptoms and social interaction experience.
Sleep, 17 (1994), pp. 466-475
[6]
J. Baillargeon, G. Singh, Y.F. Kuo, et al.
Association of opioid and benzodiazepine use with adverse respiratory events in older adults with chronic obstructive pulmonary disease.
Ann Am Thorac Soc, 16 (2019), pp. 1245-1251
[7]
P.N. Gueye, F. Lofaso, S.W. Borron, et al.
Mechanism of respiratory insufficiency in pure or mixed drug-induced coma involving benzodiazepines.
J Toxicol Clin Toxicol, 40 (2002), pp. 35-47
Copyright © 2023. Sociedade Portuguesa de Pneumologia
Download PDF
Pulmonology
Article options
Tools

Are you a health professional able to prescribe or dispense drugs?