Blood pressure, snoring, obesity, and nocturnal hypoxaemia

Blood pressure, snoring, obesity, and nocturnal hypoxaemia

Summary

Subjects and methods

The association between snoring and blood pressure is still a matter for debate, partly because of uncertainty about the definition of snoring and partly because confounding factors may affect systemic blood pressure such as obesity, sleep apnoea, and nocturnal hypoxaemia. To isolate the contribution of each of these factors, 1415 patients (389 females, 1026 males) referred to a sleep disorders centre were studied. A full history was obtained with particular attention to cardiovascular disease and medications. The patients had nocturnal polysomnography including objective measurement of snorning, and blood pressure was measured in the morning. 18% of non-snorers had hypertension as did 20% of heavy snorers (not significantly different). Multivariate linear regression analysis showed that snoring was not a significant determinant of blood pressure. Only age, male sex, apnoea/hypopnoea index, and body mass index contributed significantly to the variability of blood pressure. We conclude that snoring in the absence of sleep apnoea is not associated with raised blood pressure.

Study population

Introduction Snoring has been reported by some,l-6 but

not all’-11 with be associated increased risk of investigators hypertension, coronary artery disease, and cerebrovascular accidents. This difference of opinion may be due in part to methodological problems regarding detection of snoring, apnoea, and hypertension. In none of the studies was snoring measured at home or in the sleep laboratory, but was inferred from questionnaires, resulting in a subjective classification of patients into snorers and non-snorers based on the bedpartner’s perception of snoring. Furthermore, because snoring is a common sign of sleep apnoea, and because sleep studies were not done consistently, it is possible that populations of snorers included patients with sleep apnoea, thus modulating the relation between snoring and hypertension with the unaccounted effects of apnoea and associated nocturnal oxygen desaturation. In addition, other factors that influence blood pressure, such as age, obesity, smoking, and alcohol consumption, were not always controlled for. In view of these limitations, the present study was designed to separate the effects of obesity, snoring, apnoea, and nocturnal oxygen desaturation on blood pressure. to

Department of Medicine, St Michael’s Hospital, 30 Bond Street, Toronto, Ontario M5B 1W8, Canada (V Hoffstein MD) Correspondence to: Dr Victor Hoffstein

1415 consecutive patients referred to a sleep disorders clinic were studied. All patients had standard clinical history and physical examination with particular attention to cardiovascular disease, medications that may affect blood pressure, and smoking; nocturnal polysomnography including measurements of snoring; and measurement of blood pressure. 752 patients were referred because of snoring and suspicion of sleep apnoea; other reasons for referral were excessive daytime sleepiness, tiredness, or fatigue (257); difficulty in initiating or maintaining sleep (88); bedpartner’s observations of cessation of breathing (71); insomnia (60); nocturnal choking (29); assessment of nocturnal oxygenation (18); gasping for breath (13); morning headaches (11); nocturnal cough (9); and leg twitching (9). The remaining 98 were referred because of a variety of complaints such as sudden cardiac arrest at night, suspicion of narcolepsy, night screams, and night terrors.

Nocturnal polysomnography included Standard nocturnal polysomnography electroencephalogram, submental and anterior tibial electromyogram, measurements of oronasal airflow, chest wall and abdominal excursions, oxygen saturation, and single-lead electrocardiogram. Snoring was measured12 with a microphone

placed above the nasion, connected to a sound meter (model SL120, Pacer Industries, Chippewa Falls, WI, USA) calibrated between 40 and 100 dB. Sound intensity was displayed along with other polysomnographic variables on the strip-chart recorder. The sound signal was also sampled by analogue-to-digital converter and stored for subsequent analylsis. With this setup, normal breathing registers less than 50 dB; spikes in sound intensity higher than 50 dB are counted as snores. This definition of snoring was validated in previous studies dealing with perception of snoring by an independent observer." Polysomnograms were scored in the usual manner. Apnoeas and hypopnoeas were identified as either complete or incomplete (greater than 50% reduction in tidal volume) episodes of cessation of breathing lasting longer than 10 s. The number of such episodes per hour of sleep is the apnoea/hypopnoea index (AHI); the number of snores per hour, the snoring index (SI); maximum nocturnal sound intensity (dBmax) was also recorded. Lowest nocturnal oxygen saturation (Lo02), mean nocturnal oxygen saturation (Mn02), and percentage of total sleep time with oxygen saturation lower than 90% (TST90%) were also measured. Blood pressure measurements

Systolic and diastolic blood pressures were measured with an automated system (Critikon Dinamap model 1846 SX, Critikon Inc, Tampa, FL, USA). The cuff appropriate for patient’s body size was used. Three consecutive measurements were done over 10 min in the morning, before patients’ arousal from bed and the average used for subsequent analysis.

Table 1:

Proportions of hypertensive and normotensive patients according to categories of snoring severity 643

TST= total sleep time, d8max= maximum nocturnal sound intensity, Aw0,= awake oxygen saturation, BPsys, BPdia = systolic and diastolic blood pressures.

Table 2:

Anthropometric, sleep, and blood pressure

Patients were classified as hypertensive if their blood pressure measured in the laboratory was 160/95 mm Hg or more, or if they were currently on antihypertensive medications.

Analysis study possible associations between snoring and blood pressure, we examined, first, the proportion of patients with and without hypertension among mild snorers (SI < 100), moderate snorers (SI= 100-00), and severe snorers (SI > 400) by X2 tests. Second, we used one-way analysis of variance to compare blood

To

pressures among the three category of snorers. Patients on antihypertensive medications were excluded from this analysis. Third, regression analysis was done to determine the relative contributions of age, sex, obesity, snoring, apnoea, and nocturnal oxygenation to variability in blood pressure. Univariate analysis was done to determine separate contributions of each of the above variables, then full model multiple linear regressions analysis was done to determine the influence of each variable after adjusting for all of the other variables in the model. Finally, stepwise regression was carried out to select significant predictors of blood pressure.

Results Table 1 shows proportions of patients with hypertension among different categories of snoring severity; there was no significant difference (X2 4-2, p 0-12). Table 2 summarises the anthropometric data, relevant sleep information, and morning blood pressures in all patients. There was a wide individual variability in blood pressures. Within-individual variability in blood pressure was small (coefficient of variation of 2-9%). The figure shows how blood pressure, body mass index (BMI), AHI, and nocturnal oxygenation vary with the =

=

For abbreviations see text and table 2. *Significant contribution to final model

by stepwise multiple regression. Table 3: Associations between blood pressure and other variables by univariate and multivariate regression analysis

severity of snoring (260 patients

on

antihypertensive

medications were excluded from this analysis). Although the differences in blood pressure (mean [SD]) between severe snorers (122/74 [16/10] mm Hg), moderate snorers (120/73 [16/11] mm Hg), and non-snorers (119/71 [17/11]

Hg) were trivial, they were statistically significant (p < 0-005). However, the differences in all other variables (AHI, BMI, TST90%) were also statistically significant (p < 0-005); therefore we cannot ascribe even these minor differences in blood pressure to snoring alone. Results of the regression analysis are shown in table 3. Although univariate analysis showed statistically significant contributions for all variables when considered individually, full-model multivariate analysis showed that only male sex, age, BMI, and AHI contribute significantly to a final model. These variables in fact were the only ones selected by the stepwise multiple regression analysis which

mm

resulted in =0-181 for the diastolic blood pressure and r2 0-209 for systolic blood pressure. The above analysis indicates that snoring is not a significant determinant of blood pressure. To further illustrate this, we selected a subgroup of non-apnoeic =

TST90% (AHI < 10), non-desaturating (Mn0>90%, < 5%), non-snorers (SI < 50) and matched them one-forfor BMI with a similar group of severe snorers (SI > 500). This selection process resulted in 110 patients, 55 in each group. There was no significant difference in blood pressures (mean [SD]) between them (118/72 [16/12] mm Hg in non-snorers vs 116/71 [14/9] mm Hg in severe

one

snorers). Discussion

1-11 S1UU

l.VV<::>1

,4UL)

::>1>4VV

Severity of snoring Figure: Mean (SD) of systolic (BPsys) and diastolic (BPdla) blood pressure (mm Hg), AHI, BMI, and TST90% according to severity of snoring 644

This study shows that snoring is not associated with increases in systemic blood pressure, but sleep apnoea is. Features of our study, not found in previous investigations, are direct measurement of snoring in the sleep laboratory and standardised measurement of blood pressure, always done immediately upon awakening with patients supine and resting. It is probably lack of consideration of the above factors that led early investigators1-S to report a positive association between snoring and blood pressure, although Waller et al14

reviewing these studies concluded in 1989 that "it is premature to describe snoring as a risk factor for cardiovascular disease". Once the confounding effects of age and obesity were taken into account in the more recent investigations 7-10 snoring was eliminated as an independent variable affecting blood pressure. Furthermore, when sleep studies were done in a subset of habitual snorers with hypertensionmany of them were found to have sleep apnoea, thus raising a possibility that it is sleep apnoea, and not snoring, which is an independent predictor of raised blood pressure. It could be said that our division of patients into snorers and non-snorers based on objective measurement of snoring is inaccurate if the night-to-night variability of snoring is large. If that is the case, then single-night polysomnography may classify some patients incorrectly. This argument cannot be resolved at present due to lack of data. One study1S suggested that night-to-night variability of snoring is about 20%, although the method used to measure snoring was different from ours. An additional feature of the present study, which precludes a meaningful comparison of our results with others, is the type of population examined. Whereas previous investigations examined a large unselected segment of the population, our sample consisted of patients referred to the sleep disorder centre, although not always because of snoring. Nevertheless, it is possible that this group as a whole may be predisposed to hypertension and that is why we did not find a significant additive effect of snoring or nocturnal hypoxaemia on blood pressure. However, it is surprising that we did not find any significant differences in blood pressure between patients at the two extremes of the snoring spectrum. Acute measurements of blood pressure during sleep in patients with sleep apnoea show cyclic increases in blood pressure during apnoeic episodes.16-18 However, it is not clear whether this rise in blood pressure is due to hypoxaemia, apnoea, or snoring. Some human 17 and animah9 studies present evidence suggesting that hypoxia has a specific pressor effect in sleep apnoea, whereas others2o suggest that hypoxaemia alone does not explain rises in blood pressure. Our findings support the latter hypothesis and point toward apnoea as being an important factor in the pathogenesis of nocturnal increases in blood pressure.

Why should intermittent episodes of cessation of breathing during sleep be more important in terms of their association with increases in blood pressure than other transient phenomena such as snoring and hypoxaemia? One possibility is that the association between blood pressure and apnoea represents a response to stimulus that is a consequence of apnoea, rather than a response to apnoea itself. Apnoea is a complex event, associated with seemingly conflicting alterations in parasympathetic and sympathetic activity. However, one stimulus that is almost invariably seen at the termination of apnoea is an arousal. This may lead to sympathetic activation and rise in blood pressure.21,22 Apnoeic events are much more commonly

associated with arousals than snoring or intermittent non-apnoeic hypoxaemia, and that is why they may exert a stronger influence on blood pressure. We conclude that direct measurements of snoring and blood pressure, after adjustment for apnoea, hypoxaemia, and obesity, do not support an independent positive correlation between these two variables.

References 1

Lugaresi E, Cirignotto F, Coccagna G, Piana C. Some epidemiological data on snoring and cardiocirculatory disturbances. Sleep 1980; 3:

221-24. Koskenvuo M, Kaprio J, Partinen M, Langinvainio H, Sarna S, Heikkila K. Snoring as a risk factor for hypertension and angina pectoris. Lancet 1985; i: 893-96. 3 Partinen M, Palomaki H. Snoring and cerebral infarction. Lancet 1985; ii: 1325-26. 4 Koskenvuo M, Kaprio J, Telakivi T, Partinen M, Heikkila K, Sarna S. Snoring as a risk factor for ischaemic heart disease and stroke in man. BMJ 1987; 294: 16-19. 5 Gislason T, Aberg H, Taube A. Snoring and systemic hypertensionan epidemiological study. Acta Med Scand 1987; 221: 415-21. 6 Gislason T, Benediktsdóttir B, Björnsson JK, Kjartansson G, Kjeld M, Kristbjarnarson H. Snoring, hypertension, and the sleep apnea syndrome: an epidemiologic survey of middle-aged women. Chest 1993; 103: 1147-51. 7 Schmidt-Nowara WW, Coultas DB, Wiggins C, Skipper BE, Samet JM. Snoring in a Hispanic-American population: risk factors and association with hypertension and other mortality. Arch Intern Med 1990; 150: 597-601. 8 Rausher H, Popp W, Zwick H. Systemic hypertension in snorers with and without sleep apnoea. Chest 1992; 102: 367-71. 9 Jennum P, Sjol A. Snoring, sleep apnoea, and cardiovascular risk factors: the MONICA II study. Int J Epidemiol 1993; 22: 439-44. 10 Jennum P, Hein HO, Suadicani P, Gyntelberg F. Cardiovascular risk factors in snorers: a cross-sectional study of 3,323 men aged 54 to 74 years: the Copenhagen male study. Chest 1992; 102: 1371-76. 11 Stradling JR, Crosby JH. Relation between systemic hypertension and sleep hypoxemia or snoring: analysis of 748 men drawn from general practice. BMJ 1990; 300: 75-78. 12 Mateika JH, Mateika S, Slutsky AS, Hoffstein V. The effect of snoring on mean arterial blood pressure during non-REM sleep. Am Rev Respir Dis 1992; 145: 141-46. 13 Hoffstein V, Mateika S, Anderson D. Snoring: is it in the ear of the beholder? Sleep (in press). 14 Waller PC, Bhopal RS. Is snoring a cause of vascular disease? An epidemiological review. Lancet 1989; i: 143-46. 15 Series F, Marc I, Atton F. Comparison of snoring measured at home and during polysomnographic studies. Chest 1993; 103: 1769-73. 16 Tilkian AG, Guilleminault C, Schroeder JS, Lehrman KL, Simmons FB, Dement WC. Hemodynamics in sleep-induced apneas: studies during wakefulness and sleep. Ann Intern Med 1976; 85: 714-19. 17 Hedner JA, Wilcox I, Laks L, Grunstein RR, Sullivan CE. A specific and potent effect of hypoxia in patients with sleep apnea. Am Rev Respir Dis 1992; 146: 1240-45. 18 Stoohs R, Guilleminault C. Cardiovascular changes associated with obstructive sleep apnea syndrome. J Appl Physiol 1992; 72: 583-89. 19 Iwase N, Kikuchi Y, Hida W, et al. Effects of repetitive airway occlusion on O2 saturation and systemic and pulmonary arterial pressure in anesthetized dogs. Am Rev Respir Dis 1992; 146: 1402-10. 20 Ringler J, Basner RC, Shannon R, et al. Hypoxemia alone does not explain blood pressure elevations after obstructive apneas. J Appl Physiol 1990; 69: 2143-48. 21 Marrone O, Riccobno L, Salvaggio A, Mirabella A, Bonanno A, Bonsignore AR. Catecholamines and blood pressure in obstructive sleep apnea syndrome. Chest 1993; 103: 722-27. 22 Ali NJ, Davies RJO, Fleetham JA, Stradling JR. The acute effects of continuous positive airway pressure and oxygen administration on blood pressure during obstructive sleep apnea. Chest 1992; 101: 526-32. 2

645