American Journal of Respiratory and Critical Care Medicine

Prevention of acute hypercapnia during obstructive events in obstructive sleep apnea requires a balance between carbon dioxide (CO2) loading during the event and CO2 unloading in the interevent period. Earlier studies have demonstrated that acute CO2 retention may occur despite high interevent ventilation when the interevent duration is short relative to the duration of the preceding event. The present study examines the relationship between apnea and interapnea durations and relates this assessment of ventilatory periodicity to the degree of chronic hypercapnia in subjects with severe sleep apnea. A total of 18 subjects with sleep apnea (> 40 apnea/hour; chronic awake PaCO2 36–62 mm Hg) and without underlying lung disease underwent polysomnography. For each event, apnea duration, interapnea duration, and apnea/interapnea duration ratio were determined. No relationship was observed between chronic PaCO2 and mean apnea or interapnea duration (p > 0.1). However, PaCO2 was directly related to apnea/interapnea duration ratio (r = 0.48; p < 0.05) such that with increasing chronic hypercapnia the interapnea duration shortens relative to the apnea duration. The present study suggests that control of the interapnea ventilatory duration relative to the duration of the preceding apnea, is an important component of the integrated ventilatory response to CO2 loading during apnea and may contribute toward the development and/or maintenance of chronic hypercapnia in obstructive sleep apnea/hypopnea syndrome.

Chronic hypercapnia occurs in a subset of patients with obstructive sleep apnea/hypopnea syndrome (14). Analysis of the number, type, and duration of obstructive events reveals no consistent differences in patients with chronic hypercapnia when compared with eucapneic patients (5). However, treatment of these events may result in correction of hypercapnia, suggesting an important contribution of the acute hypercapnia that develops during obstructive events toward the generation/maintenance of the chronic hypercapneic state (6).

Prevention of the acute hypercapnia that results from reduced ventilation during repetitive obstructive events requires a balance between loading of carbon dioxide (CO2) during the event and unloading of CO2 during the interevent period. We have previously studied the acute CO2 kinetics of individual event/interevent cycles during a daytime nap in subjects with severe obstructive sleep apnea/hypopnea syndrome (7). The study indicated that interevent duration could limit CO2 unloading. Specifically, when the duration of the interevent period was short relative to the duration of the event (event:interevent duration ratio ∼ 3:1), there was an obligate net CO2 retention (CO2 load > CO2 unload) for that event, despite a high rate of interevent ventilation. This observation provides the physiologic basis for evaluating the ratio between event duration, which determines CO2 load from ongoing metabolism, and interevent duration, which reflects time for CO2 unloading. The present study examines this measure of ventilatory periodicity, which is a potential mechanism for acute hypercapnia during individual events, and relates it to the degree of chronic hypercapnia in subjects with severe obstructive sleep apnea syndrome (OSAS). We hypothesize that subjects with chronic hypercapnia will have, on average, a shorter interevent duration relative to the duration of the preceding event, and that this phenomenon will be exaggerated with increasing chronic hypercapnia.

Subjects with obstructive sleep apnea and without underlying lung disease were enrolled in the protocol. All subjects complained of excessive daytime sleepiness (Epworth Sleepiness Scale > 15) and loud snoring. Nocturnal polysomnography was performed, and 18 subjects (age, 23–68; 15 men, 3 women) with severe OSAS (apnea index > 40/hour) and awake PaCO2 ranging from 36 to 62 mm Hg were selected for the analysis. Patients with high apnea/hypopnea index were chosen to exclude subjects with extended periods of stable ventilation during sleep that may compensate for apnea; an apnea index greater than 40 per hour dictates that the average interapnea duration is less than or equal to 80 seconds. Subjects with coexisting ventilatory sleep disorders such as central apnea and central hypoventilation were also excluded from the study. This selection maximized the likelihood that hypercapnia was due to the apnea phenomenon. In addition, subjects with predominantly apneic events (> 75% of apnea/hypopnea index) were selected because CO2 accumulation during apnea is dependent only on event duration, in contrast to hypopnea, where CO2 accumulation is also dependent on the ventilation during the event. Subjects with clinical evidence of chronic lung disease, FEV1/FVC less than 70%, hypothyroidism, or currently using respiratory depressants were excluded from the study.

Before the sleep study, subjects had spirometry and an arterial blood–gas measurement during wakefulness while breathing room air. Nocturnal polysomnography was performed at the New York University Sleep Disorders Center. Recordings of central and occipital electroencephalogram, electro-oculogram, and submental electromyogram were used to monitor sleep. An anterior tibialis electromyogram was used to detect leg movements, and unipolar electrocardiogram was used for cardiac monitoring. Oxygen saturation was monitored with a pulse oximeter (Sensormedics, Yorba Linda, CA). Chest wall and abdominal movement were monitored with piezoelectric strain gauges (EPMS, Midlothian, VA). Respiratory airflow was monitored using a nasal/oral thermistor and a nasal cannula connected to a 2-cm water pressure transducer (Validyne, Northridge, CA or Protech PTAF2, Minneapolis, MN) (8). The output of the pressure transducer was connected to a direct current amplifier with a 5-Hz low-pass filter. Sleep was scored in 30-second epochs using the criteria of Rechtshaffen and Kales (9).

Respiratory events were scored on the nasal cannula flow signal. Figure E1 in the online data supplement illustrates the scoring of respiratory events using the nasal cannula flow signal: (1) apnea were identified using the nasal cannula flow signal as periods of more than 10 seconds with flow less than 10% of baseline; (2) interapnea ventilatory periods were defined as the period of breathing after an apnea and before the next respiratory event (apnea/hypopnea); and (3) for each event, the ratio between apnea duration and subsequent interapnea ventilatory duration was determined.

The study was approved by the institutional review boards of the New York University Medical Center and the Health and Hospitals Corporation of New York City. All patients signed informed consent before entering the study.

Statistical Analysis

For each apnea, the apnea duration, interapnea ventilatory duration, and ratio of apnea duration relative to the interapnea ventilatory period were determined. For each subject, the average value of apnea duration, interapnea duration, and average ratio of all events in that subject were calculated. Relationships between variables were investigated using univariate or multivariate regression analyses as indicated (SPSS for Windows version 10.05, Chicago, IL).

Anthropometric, pulmonary function, and arterial blood–gas data for all 18 subjects are illustrated in Table E1 in the online data supplement. All subjects were obese, with body mass index (BMI) ranging from 28 to 57 kg/m2. There was no evidence of obstructive dysfunction, and FEV1/FVC and flow at 50% vital capacity relative to the FVC were within normal limits in the 17 subjects who were evaluated with spirometry. The FVC was greater than 80% of the predicted value in all but four subjects. In these four subjects, the reduction in FVC was due to a reduced expiratory reserve volume with a preserved inspiratory capacity. Thus, spirometry was compatible with obesity, and there was no evidence of parenchymal, chest wall, or neuromuscular disease. The chronic awake PaCO2 ranged from 36 to 62 mm Hg and was greater than 45 mm Hg in eight subjects. The HCO3 levels confirmed the chronic nature of the elevated PaCO2 in hypercapneic subjects. There was a trend toward higher BMI in hypercapneic subjects that did not reach statistical significance (r = 0.41; p = 0.096).

All subjects had severe OSAS, with apnea/hypopnea index greater than 55 per hour (range, 55–144/hour) and respiratory events were predominantly apneas with apnea index greater than 40 per hour (range, 41–137/hour). There was no correlation between apnea/hypopnea index or apnea index and the chronic awake PaCO2 (p > 0.1).

Figure 1A

illustrates the relationship between the mean apnea duration (± SD) in each subject and the level of chronic awake PaCO2. Each point represents the mean value of all apneic events for that subject. The average apnea duration varied from 17 to 44 seconds across subjects. In addition, the wide SD values noted in each subject illustrate the within-subject variability in apnea duration. The mean apnea duration was not significantly longer in subjects with chronic awake hypercapnia (r = 0.22; p = 0.37).

Figure 1B illustrates the relationship between the mean interapnea duration (± SD) in each subject and the level of chronic awake PaCO2. Each point represents the mean value of all events for that subject. The average interapnea duration varied from 7 to 20 seconds across subjects. Hypercapneic subjects demonstrated less variability in interapnea duration compared with eucapneic subjects, as illustrated by the narrower SD values around the mean. The mean interapnea duration was not consistently shorter in subjects with chronic awake hypercapnia (r = −0.35; p = 0.15).

Figure 2

illustrates the mean interapnea ventilatory duration as a function of the mean apnea duration for all subjects. There was no correlation between the apnea duration and interapnea duration. Specifically, long apnea duration did not dictate a short interapnea duration (r = 0.26; p = 0.27).

Figure 3

illustrates the mean apnea/interapnea duration ratio for each subject as a function of the chronic awake PaCO2. The mean apnea/interapnea duration ratio provides a measure of the pattern of ventilatory periodicity and increases as the interapnea ventilatory duration decreases relative to the duration of the preceding apnea. The mean apnea/interapnea duration ratio ranged from 1.6 to 4.7. This apnea/interapnea duration ratio measured during sleep was directly related to the chronic PaCO2 measured during wakefulness (r = 0.48; p < 0.05). Variability of the apnea/interapnea duration ratio was similar within each subject (see Table E2 in online data supplement). Thus, the increased apnea/interapnea duration ratio seen in hypercapneic subjects was driven by a consistent increase in ratio in that individual rather than by a few extreme values.

The effect of sleep stage on these results was analyzed by separating data that were recorded during rapid eye movement sleep from data that were recorded during nonrapid eye movement sleep. The number of apneas that was recorded during rapid eye movement sleep varied across subjects and ranged from 0 to 52 events; these events constituted approximately 7% of the total number of events analyzed. The average apnea duration was significantly longer during rapid eye movement sleep (40.7 ± 10.9 seconds) than during nonrapid eye movement sleep (26.4 ± 6.8 seconds) in all subjects (p < 0.001). The average interapnea duration did not differ during rapid eye movement sleep (11.3 ± 4.1 seconds) compared with nonrapid eye movement sleep (10.8 ± 2.8 seconds; p value was not significant). Therefore, the average ratio between apnea and interapnea duration was significantly larger during rapid eye movement sleep (4.33 ± 1.46) compared with nonrapid eye movement sleep (2.77 ± 0.79) in all subjects (p < 0.001). There was no significant variability in any parameter across the duration of the nocturnal polysomnogram.

Because there was a trend toward higher BMI in hypercapneic subjects, a multivariate analysis was performed relating PaCO2 as the dependent variable to BMI and apnea/interapnea duration ratio as independent variables. Data used for this analysis are shown in the online data supplement (see Tables E1 and E2). Results from the multivariate analysis (Table 1

TABLE 1. Regression analyses with paCO2 as the dependent variable




Independent
 Variable

Slope

Standardized
 β-coefficient

r2

p Value
Univariate analyses
Model 1Ratio4.270.480.230.046
Model 2BMI0.290.410.160.096
Multivariate analysis
Model 3Ratio4.80.540.014

BMI
0.34
0.46
0.45
0.027

Definition of abbreviation: BMI = body mass index.

; Model 3) indicate that both BMI and apnea/interapnea duration ratio have independent effects on PaCO2. An increase in the ratio of 1.0 was associated with a 4.8-mm Hg increase in the chronic PaCO2 in this dataset. The relative effect of BMI on PaCO2 was slightly less than the effect of the ratio, as illustrated by the standardized β-coefficients. No significant interaction between ratio and BMI was found (data not shown). Thus, for a given apnea/interapnea duration ratio, greater degrees of obesity are associated with higher values for PaCO2.

The present study suggests that control of interapnea ventilatory duration relative to the duration of the preceding apnea is an important component of the integrated ventilatory response to CO2 loading during apnea. Previous data on acute CO2 kinetics have demonstrated that shortening of the interapnea period may be a mechanism for impaired CO2 elimination in OSAS independent of the level of interevent ventilation (7). The present study extended the above observation for acute hypercapnia in individual events and examined the ventilatory periodicity in subjects with severe obstructive sleep apnea/hypopnea syndrome across a range of PaCO2 values. The data demonstrated that the pattern of ventilatory periodicity, expressed as the average apnea/interapnea duration ratio, was directly related to the chronic awake PaCO2. Subjects with higher chronic PaCO2 had more events with relative shortening of interevent ventilatory duration.

By definition, transient hypercapnia occurs when ventilation immediately after an apnea is insufficient to balance the CO2 loaded during apnea. The interapnea compensatory ventilation is a function of both the magnitude of ventilation as well as the duration of the interapnea ventilatory period. Previous studies from this laboratory have quantified the CO2 load during individual apneic events and the postapneic ventilation in subjects with severe OSAS (10). The postapnea ventilation was uniformly high after all events, even in hypercapneic subjects. However, when this postapnea ventilation was related to the CO2 load of the preceding apnea, hypercapneic subjects demonstrated decreased ventilation for a given CO2 load compared with eucapneic subjects (10). This finding implies that maintenance of CO2 balance in subjects with chronic hypercapnia might require a longer duration of ventilation between apneas. However, the present data demonstrate that increased chronic PaCO2 is associated with a reduced duration of interapnea ventilation relative to the length of the preceding apnea. Thus, subjects with chronic awake hypercapnia are likely to have more apnea/interapnea cycles with impaired CO2 unloading when compared with eucapneic subjects, due to both a reduced magnitude of interapnea ventilation relative to the CO2 load, as previously shown (10), and a reduced interapnea ventilatory duration relative to the duration of apnea, as shown in the present study.

Mechanisms for the relative shortening of the interapnea duration are of physiologic interest. In the present study, the relative shortening of interapnea duration was not dictated by the duration of apnea or frequency of apneic events. All subjects had severe OSAS and apnea occurred with onset of sleep, dictating that interapnea duration was determined by the ability to maintain wakefulness between obstructive events. Greenberg and coworkers (11) demonstrated that interapnea duration increased after blockade of opioid receptors, suggesting a role for endogenous opioids as a mechanism for control of duration of the interapnea period. Furthermore, increased cerebrospinal fluid β-endorphin activity has been reported in subjects with OSAS, with return to normal values after treatment of apnea (12). Although it is not known if endogenous opioid levels are different in hypercapnia compared with eucapnia, opioid receptor blockade has resulted in the normalization of arterial Pco2 in a subject with obesity hypoventilation syndrome (13) and has improved nocturnal hypoxemia and hypercapnia in eucapneic subjects with OSAS (14).

The presence of chronic hypercapnia in OSAS has previously been attributed to a variety of factors including underlying lung disease, presence of coexisting central hypoventilation, impaired ventilatory response to CO2, and obesity (6, 1518). Subjects in the present study were specifically selected to have normal lung function and no ventilatory sleep disorder other than OSAS to control for some of these confounding variables. In this group of subjects, an independent effect of obesity on PaCO2 was demonstrated, such that for a given apnea/interapnea duration ratio, greater degrees of obesity are associated with higher values for PaCO2. The additive effect of obesity on the relationship between apnea/interapnea duration ratio and chronic PaCO2 may be due to the effect of increased metabolic rate (high BMI) on CO2 load during apnea, as suggested by Javaheri and coworkers (17), or due to the impaired interapnea elimination of CO2 due to mass loading and/or reduced FRC.

The link between acute CO2 loading during apnea/interapnea cycles and the development of chronic awake hypercapnia is complex. Once chronic hypercapnia has been established, control of the pattern of ventilatory periodicity would be a potential mechanism for the maintenance of the chronic hypercapneic state by impairing unloading of CO2 during sleep. This impairment of CO2 unloading is counterbalanced by the increased gradient for CO2 elimination when blood Pco2 is elevated, thereby allowing for the stability of the chronic hypercapneic state. The relevance of this phenomenon to the period during which chronic hypercapnia develops in patients with OSAS is speculative and remains to be further explored.

1. Resta O, Foschino-Barbaro MP, Bonfitto P, Talamo S, Legari G, De Pergola G, Minenna A, Giorgino R. Prevalence and mechanisms of diurnal hypercapnia in a sample of morbidly obese subjects with obstructive sleep apnoea. Respir Med 2000;94:240–246.
2. Rapoport DM, Garay SM, Epstein H, Goldring RM. Hypercapnia in the obstructive sleep apnea syndrome: a reevaluation of the “Pickwickian syndrome.” Chest 1986;89:627–635.
3. Berthon-Jones M, Sullivan CE. Time course of change in ventilatory response to CO2 with long-term CPAP therapy for obstructive sleep apnea. Am Rev Respir Dis 1987;135:144–147.
4. Verin E, Tardif C, Pasquis P. Prevalence of daytime hypercapnia or hypoxia in patients with OSAS and normal lung function. Respir Med 2001;95:693–696.
5. Garay SM, Rapoport D, Sorkin B, Epstein H, Feinberg I, Goldring RM. Regulation of ventilation in the obstructive sleep apnea syndrome. Am Rev Respir Dis 1981;124:451–457.
6. Berger KI, Ayappa I, Chatr-aryamontri B, Marfatia A, Sorkin IB, Rapoport DM, Goldring RM. Obesity hypoventilation syndrome as a spectrum of respiratory disturbances during sleep. Chest 2001;120:1231–1238.
7. Berger KI, Ayappa I, Sorkin IB, Norman RG, Rapoport DM, Goldring RM. CO2 homeostasis during periodic breathing in obstructive sleep apnea. J Appl Physiol 2000;88:257–264.
8. Norman RG, Ahmed MM, Walsleben JA, Rapoport DM. Detection of respiratory events during NPSG: nasal cannula/pressure sensor versus thermistor. Sleep 1997;20:1175–1184.
9. Rechtschaffen A, Kales A. A manual of standardized terminology, techniques, and scoring system for sleep states of human subjects. Washington, DC: U.S. Government Printing Office; 1968. NIH Publication No. 204.
10. Berger KI, Ayappa I, Sorkin IB, Norman RG, Rapoport DM, Goldring RM. Post-event ventilation as a function of CO2 load during respiratory events in obstructive sleep apnea. J Appl Physiol 2002;93:917–924.
11. Greenberg HE, Rapoport DM, Rothenberg SA, Kanengiser LA, Norman RG, Goldring RM. Endogenous opiates modulate the postapnea ventilatory response in the obstructive sleep apnea syndrome. Am Rev Respir Dis 1991;143:1282–1287.
12. Gislason T, Almqvist M, Boman G, Lindholm CE, Terenius L. Increased CSF opioid activity in sleep apnea syndrome: regression after successful treatment. Chest 1989;96:250–254.
13. Orlowski JP, Herrell DW, Moodie DS. Narcotic antagonist therapy of the obesity hypoventilation syndrome. Crit Care Med 1982;10:604–607.
14. Ferber C, Duclaux R, Mouret J. Naltrexone improves blood gas patterns in obstructive sleep apnoea syndrome through its influence on sleep. J Sleep Res 1993;2:149–155.
15. Han F, Chen E, Wei H, He Q, Ding D, Strohl KP. Treatment effects on carbon dioxide retention in patients with obstructive sleep apnea-hypopnea syndrome. Chest 2001;119:1814–1819.
16. Bradley TD, Rutherford R, Lue F, Moldofsky H, Grossman, RF, Zamel N, Phillipson EA. Role of diffuse airway obstruction in the hypercapnia of obstructive sleep apnea. Am Rev Respir Dis 1986;134:920–924.
17. Javaheri S, Colangelo G, Lacey W, Gartside PS. Chronic hypercapnia in obstructive sleep apnea-hypopnea syndrome. Sleep 1994;17:416–423.
18. Resta O, Foschino BM, Bonfitto P, Talamo S, Mastrosimone V, Stefano A, Giliberti T. Hypercapnia in obstructive sleep apnoea syndrome. Neth J Med 2000;56:215–222.
Correspondence and requests for reprints should be addressed to Indu Ayappa, Ph.D., New York University School of Medicine, 550 First Avenue, NBV7W54 New York, NY 10016. E-mail:

Related

No related items
American Journal of Respiratory and Critical Care Medicine
166
8

Click to see any corrections or updates and to confirm this is the authentic version of record