Outcomes of Impaired Objective Daytime Vigilance in Obesity-Hypoventilation Syndrome

1 May
2016

Sleep

Baseline Anthropometric, Functional, Sleep, and Vigilance Data

Fifteen consecutive patients (10 men), with a mean age of 55 ± 9 years were prospectively included (Tables 1-3). They were morbidly obese, had moderate-to-severe daytime hypercapnia without abnormal ventilatory function. They presented with a combination of OSAS (ie, apnea-hypopnea index [AHI], 62 ± 32 events per hour of sleep) and REM hypoventilation. The average sleep time spent in hypoventilation exceeded one third of REM sleep (mean duration, 35 ± 33% [corresponding to a mean duration of 19.2 ± 17.4 min per night]). Subjective daytime sleepiness was impaired, with a mean Ep-worth sleepiness scale score of 11 ± 4. An objective sleepiness assessment showed reduced sleep latency during the OSLER test in six patients,

CO2 Ventilatory Response Status, Sleep Abnormalities, and Daytime Vigilance

Seven patients were included in the low CO2 responder group (< 1.5 L/min/mm Hg), whereas eight patients had a normal CO2 sensitivity (Table 4). There was a significant relationship between CO2 sensitivity and the amount of hypoventilation in REM sleep (r = 0.54; p = 0.037) [Fig 2]. Patients who were low responders had higher objective daytime sleepiness, which was measured in terms of shorter mean sleep latency periods during the OS-LER test (23 ± 14 min vs 37 ± 8 min, respectively; p = 0.05), although they exhibited the same amount of sleep fragmentation.

NIV Efficacy Assessment

NIV settings were 12 to 22 millibars for inspiratory pressure and 6 to 12 millibars for expiratory pressure (Tables 2, 3). Short-term NIV use improved diurnal PaC02 significantly (from 47.3 ± 2.3 to 41.3 ± 3 mm Hg; p < 0.0001). Inspiratory pressure and expiratory pressure may be increased or decreased with medications of Canadian HealthCare Mall.

AHI decreased significantly from 62 ± 32 to 11 ± 13 events per hour (p < 0.0001). We did not find any residual hypoventilation with the use of NIV during REM sleep. Sleep architecture changed significantly. Stage 1 decreased (p = 0.005), while stages 3 and 4 and REM sleep significantly improved (p = 0.007 and 0.02, respectively) [Fig 3]. Whereas the AHI normalization was associated with a major reduction in the number of respiratory-related mi-croarousals, the number of non-respiratory-related microarousals increased significantly.

For the whole group, there was a nonsignificant increase in ventilatory responses to CO2 (Table 1). Low responders at baseline increased their ventilatory responses by 47%, but the mean value remained close to 1.5 L/min/mm Hg, which is the lower limit for normal values (Table 5). The values of only two patients returned to the normal range.

Using NIV, the vigilance assessed by the Epworth sleepiness scale score was significantly improved by the treatment (p = 0.05) for the whole group. The Epworth sleepiness scale score was significantly improved in low CO2 responders (p = 0.02), whereas it did not improve in normal CO2 responders (Table 3). Moreover, only low CO2 responders who significantly improved their vigilance, according to the number of sleep latency periods during the OSLER test, reached an average level that was comparable to the values of the normal CO2 responders (p = 0.04) [Fig 4].

Fig2
Figure 2. Correlation between the percentage of REM sleep spent in hypoventilation and initial CO2 sensitivity (n = 15). REM HypoVA = time spent in hypoventilation during REM sleep, expressed as a percentage of REM sleep time.
Fig3
Figure 3. Sleep architecture before and after NIV for the whole group (n = 15). % TST = percentage of total sleep time; SB = spontaneous breathing. * = statistical significance with p < 0.05.
Fig4
Figure 4. Sleep latency during the Osler test before and after NIV for patients with low CO2 sensitivity (n = 7) and normal CO2 sensitivity (n = 7). NS = no statistical significance; § = this patient decreased his sleep latency after NIV; an analysis of his individual data showed that his total sleep time spent using NIV was 30% less than during the first PSG session. Only seven patients with normal sensitivity had been considered in this figure because one patient was excluded from the study owing to technical problems occurring during the final OSLER test. His baseline latency period was 39.7 min.

Table 1—Anthropometric Parameters and Functional Data Before and After NIV Treatment (n = 15)

VariablesBefore NIVAfter NIVp Value
Patients
Female5
Male10
Age, yr 9±55
BMI, kg/m238.7 ± 6.1
Paco2, mm Hg47.3 ± 2.341.3 ± 3 < 0.0001
Pao2, mm Hg77.3 ± 6.874.3 ± 6.8NS
VC2NS
L3.6 ± 1.03.5 ± 1.1
% predicted9089
TLCNS
L6.0 ± 1.55.8 ± 1.7
% predicted9997
CO2 sensitivity,2.0 ± 1.32.4 ± 1.9NS
L/min/mm Hg

Table 2—Sleep Parameters Before and During NIV (n = 15)*

Variables WithoutNIV WithNIVp Value
Sleep architecture
TST, min343 ± 68312 ± 66NS
Sleep latency, min20 ± 1921 ± 18NS
Stage 1, % TST26 ± 1515 ± 70.005
Stage 2, % TST50 ± 1350 ± 7NS
Stage 3-4, % TST3 ± 37 ± 50.007
REM sleep, % TST20 ± 927 ± 40.02
RMAI, No. events/h of sleep61 ± 2411 ± 12< 0.001
Respiratory events
Non-RMAI, No. events/h of sleep4 ± 614 ± 110.006
AHI, No. events/h of sleep62 ± 3211 ± 13< 0.0001
REM HypoVA, % REM sleep35 ± 33
Mean sleep Sao2, %89 ± 393 ± 1< 0.001
Nadir sleep Sao2, %65 ± 1487 ± 6< 0.001
Time spent with Sao2 < 90%, % TST38 ± 325 ± 100.002

Table 4—Patients With Normal CO2 Sensitivity Compared to Those With Low CO2 Sensitivity (n = 15)

VariablesLow CO2 Sensitivity (n = 7) Normal CO2 Sensitivity(n = 8) pValue
CO2 sensitivity, L/min/mm Hg1.1 ± 0.32.8 ± 1.30.001
BMI, kg/m238.5 ± 738.8 ± 6NS
PaCO2, mm Hg47.3 ± 2.347.3 ± 2.3NS
PaO2, mm Hg78.0 ± 8.375.0 ± 6.0NS
Sleep parameters
RMAI, No. events/h of sleep54 ± 2767 ± 20NS
Non RMAI, No.6 ± 93 ± 2NS
events/h of sleep
AHI, No. events/h of sleep50 ± 3873 ± 240.16
REM HypoVA, %REM sleep46 ± 3725 ± 290.16
Mean sleep SaO2, %89 ± 389 ± 3NS
Nadir sleep SaO2, %63 ± 1467 ± 15NS
Time spent with SaO2 <90%, % TST35 ± 3039 ± 34NS
Epworth scale12 ± 211 ± 4NS
OSLER test23 ± 1437 ± 80.05
Sleep latency, min
EP 3-6, n2.7 ± 5.12.7 ± 4.7NS

Table 3—Daytime Sleepiness Before and After 5 Nights of Efficient NIV (n = 14)

Variables BeforeTreatment AfterTreatment pValue
Epworth sleepiness scale score
Whole group11 ± 48 ± 30.05
Low CO2 sensitivity 12 ± 27 ± 20.02
Normal CO2 sensitivity11 ± 58 ± 5NS
OSLER test
Sleep latency, min30 ± 1435 ± 100.08
Total errors, No.11 ± 158 ± 3NS
EP 3-6, No.2.0 ± 3.90.7 ± 1.70.14

Table 5—O2 Sensitivity Before and After 5 Nights of Efficient NIV (n = 15)

Variables Low CO2 Sensitivity(n = 7) Normal CO2 Sensitivity(n = 8)p Value
Baseline CO2 sensitivity, L/min/mm Hg1.1 ± 0.32.8 ± 1.30.001
Post-NIV CO2 sensitivity, L/min/mm Hg1.5 ± 0.973.2 ± 2.20.09-NS
ACO2 sensitivity, L/min/mm Hg0.44 ± 1.1 (47)0.47 ± 1.9(18)NS
top