Author + information
- Received August 24, 1998
- Revision received June 15, 1999
- Accepted August 6, 1999
- Published online November 15, 1999.
- Nir Peled, MD∗,†,
- Edward G Abinader, MD†,
- Giora Pillar, MD, DSc∗,
- Dawood Sharif, MD† and
- Peretz Lavie, PhD∗,* ()
- ↵*Reprint requests and correspondence: P. Lavie, Sleep Laboratory, Gutwirth Building, Technion-Israel Institute of Technology, Haifa, Israel 32000
To investigate the occurrence of nocturnal ischemic events in patients with obstructive sleep apnea syndrome (OSAS) and ischemic heart disease (IHD).
Although previous reports documented nocturnal cardiac ischemic events among OSAS patients, the exact association between obstructive apneas and ischemia is not yet clear. It is also not known what differentiates between patients showing nocturnal ischemia and those that do not.
Fifty-one sleep apnea patients (age 61.3 ± 8.3) with IHD participated in the study (after withdrawal of beta-adrenergic blocking agents and anti-anginotic treatment). All patients underwent whole-night polysomnography including ambulatory blood pressure recordings (30 min interval) and continuous Holter monitoring during sleep. A control group of 17 OSAS patients free from IHD were also similarly studied. Fifteen of the 51 patients were also recorded under continuous positive airway pressure (CPAP).
Nocturnal ST segment depression occurred in 10 patients (a total of 15 events, 182 min), of whom six also had morning ischemia (06–08 am). Five additional patients had only morning ischemia. No ischemic events occurred in the control group. Age, sleep efficiency, oxygen desaturation, IHD severity and nocturnal-double product (DP) values were the main variables that significantly differentiated between patients who had ischemic events during sleep and those who did not. Nocturnal ischemia predominantly occurred during the rebreathing phase of the obstructive apneas, and it is characterized by increased heart rate (HR) and DP values. Treatment with continuous positive airway pressure significantly ameliorated the nocturnal ST depression time from 78 min to 33 min (p < 0.001) as well as the maximal DP values (14,137 ± 2,827 vs. 12,083 ± 2,933, p < 0.001).
Exacerbation of ischemic events during sleep in OSAS may be explained by the combination of increased myocardial oxygen consumption as indicated by increased DP values and decreased oxygen supply due to oxygen desaturation with peak hemodynamic changes during the rebreathing phase of the obstructive apnea. Treatment with CPAP ameliorated the nocturnal ischemia.
The relative risk for ischemic heart disease (IHD) among snorers and obstructive sleep apnea syndrome patients (OSAS) is 1.2 to 6.9 higher compared with the general population (1–3). The combination of repetitive hypoxemic events and sympathetic overstimulation in OSAS may increase blood pressure and myocardial oxygen demand (4,5). Considering impaired coronary reserve, it seems that comorbidity of OSAS and IHD may expose these patients to ischemic events during sleep. Recent studies investigating nocturnal ischemia in OSAS have provided inconsistent results (Table 1). Franklin et al. (6)demonstrated OSAS among 9 of 10 ischemic patients who suffered from nocturnal angina pectoris. Schäfer et al. (7)found that 5 out of 14 OSAS + IHD patients had nocturnal ischemia and that 85.4% of the ischemic episodes were concomitant with apneas and oxygen desaturation. Most of them occurred during rapid eye movement (REM) sleep. Moreover, others (8)demonstrated ST segment depression during sleep without IHD. Hillman (3)reported that acute cardiac ischemia was an infrequent finding among postmyocardial infarction (MI) OSAS patients (16 patients) although he also reported that apnea hypopnea index (AHI) is an independent risk factor for MI. Other studies also demonstrated a lack of association between nocturnal hypoxemia and cardiac ischemia (9–11).
In this study, we examined nocturnal ischemia among 51 OSAS + IHD patients and analyzed the association of these events with the obstructive apneas before and during continuous positive air pressure (CPAP) treatment.
Fifty-one patients suffering from OSAS and from IHD (age 61.3 ± 8.3 yr) participated in the study. All had clinical symptoms and positive ergometry for IHD (mean functional capacity 2.06 ± 0.76; Mets for ischemia: 8.08 ± 2.18). Fifteen had coronary angiography, one with four-vessel disease (VD), three with three-VD, seven with two-VD and four with one-VD. Twenty-one patients had hypertension, nine mild hyperlipidemia, five peptic disease, four noninsulin diabetic mellitus, three chronic obstructive lung disease and two had previously mild cerebral vascular accident (CVA). Patients with IHD were selected for the study according to a sleep questionnaire, which indicated a high probability of OSAS. Patients with poorly controlled heart failure, insomnia, wide QRS complex, nonsinus rhythm and digoxin treatment were excluded. Patients being treated with nitrates, calcium-antagonist, ACE-inhibitors and beta-adrenergic blocking agents were weaned gradually 72 h before the sleep study. Patients were under medical supervision throughout the study, as was specified in the informed consent form. A group of 17 OSAS patients (age 47.4 ± 10.9 years) free from IHD symptoms and without ischemia at ergometry underwent the same study, as a control group. The study for both groups was approved by the appropriate Human Rights Committee.
All patients underwent whole night polysomnographic recordings which consisted of horizontal electro-oculogram, submental electromyogram, 2× ECG (lead I, V5), electroencephalogram (C3–A2), air flow (thermistor), chest movements, bed movements and pulse oxymetry (ear/finger tip). All channels were displayed on polygraph (Nihon Kohden, EEG-4214, Japan) with a paper speed of 1 cm/s. In addition, two ECG Holter channels (Dynacord-TM 420, DMS-Diagnostics Monitoring Systems, California) of a modified V5 left ventricular lead and a modified atrial and right ventricular lead and blood pressure (BP) measurements once every 30 min (Circadian 3 portable monitor) were carried out throughout the 24 h period that included the sleep study. All three recorders: polygraph, Holter and BP monitor, were time synchronized. Bedtime was 23:00 to 06:00 (all patients were awakened at 06:00). Data was collected until 08:00 in order to evaluate immediate post-sleep changes.
Sleep architecture was analyzed according to Rechtschaffen and Kales’ criteria (15). The severity of OSAS was determined by AHI which represents the number of apnea and hypopnea per hour of sleep and by oxygen desaturation values: minimal value [oxygen saturation (SaO2 min)] and percent sleep time that saturation was below 90% (SaO2 < 90%). Obstructive apnea was sought visually as an absence of airflow for at least 10 s in the presence of continued respiratory efforts. Hypopnea was defined as a reduction in the amplitude of respiratory effort for ≥50% from the sleeping baseline. The diagnosis of OSAS was based on a finding of at least an AHI of 10 in combination with typical complaints such as excessive daytime sleepiness, restless sleep or unexplained prolonged fatigue. An episode of myocardial ischemia was defined as flat or downsloping ST segment depression >1 mm from baseline for ≥0.08 s after the J point, lasting for at least 1 min and separated from the next episode by at least 1 min. Only one lead showing the largest number of ischemic episodes was used for analysis. The Holter tapes were analyzed without knowledge of the results of the sleep study. Double product (DP) values were calculated based on the 30 min BP values and heart rate data [DP = systolic BP × heart rate (HR)]. Heart rate was determined by the ECG holter every minute.
Continuous positive air pressure treatment
Fifteen patients (nine had nocturnal ischemia at the first study) were monitored under whole night CPAP treatment conducted after one night of titration.
Two sample ttests, ttest for paired samples, Pearson product moment correlations and analysis of variance (ANOVA) were used for data analysis.
Table 2summarized the clinical data for patients with and without IHD. Except for age, there were no significant differences between the two groups.
Heart rate and blood pressure
Systolic BP during sleep correlated with age (r = 0.46, p = 0.001) and with AHI (r = 0.39, p = 0.004), whereas no significant association was found between sleep BP and oxygen desaturation nor with IHD severity. Night and day HR values increased with the increase in AHI (r = 0.58, p = 0.001). Mean nocturnal HR was 65.9 ± 7.9 beats/min for mild OSAS (AHI ≤ 20; n = 27) and 75.2 ± 9.4 for the severe group (AHI > 40; n = 9, p = 0.001). Heart rate did not correlate with oxygen desaturation values. Double product values, which indicate myocardial load, were also associated with AHI (r = 0.62, p = 0.000).
Ten patients had a total of 15 nocturnal ischemic events as defined above. Six of them had also a total of six events during the morning hours (6:00 amto 8:00 am). An additional five patients had only morning ischemia (a total of five events). Thus the temporal distribution of the ischemic events was as follows: four events (46 ischemic min) occurred until 02:00, eleven events (136 min) during the rest of the night until wake up and eleven events (148 ischemic min) occurred during the first 2 h after waking up (Fig. 1). None of the patients in the control group showed any ischemic event. Patients with nocturnal ischemia were on the average 8 years older (age 67.8 ± 7.1 vs. 59.7 ± 7.9; p < 0.002), had worse sleep efficiency (61.5% ± 16.2% vs. 80.5% ± 10.6%, p < 0.001), had more severe IHD (Mets for ischemia at ergometry: 6.6 ± 2.3 vs. 8.4 ± 2.0; p < 0.02), had deeper oxygen desaturation during sleep (SaO2 min: 79.3% ± 3.7% vs. 83.6% ± 7.5%; p < 0.04), had higher nocturnal systolic BP (138.6 ± 22.8 mm Hg vs. 126.8 ± 14.9 mm Hg; p < 0.02) and had a higher nocturnal maximal DP values (14,193 ± 3,468 vs. 12,086 ± 2,721; p < 0.02).
Interestingly, AHI did not differentiate between patients with and without nocturnal ischemia. Across patients, DP values and their components were maximal at the time of ischemia compared with before (−1 h) or after (+1 h) the ischemic period. For this analysis, the BP and HR values measured at the time closest to the ischemic event were considered as the ischemic DP values. Double product increased from 8,319 ± 2,486 (−1 h) to 11,260 ± 3,150 (p < 0.000) and decreased subsequently to 8,681 ± 2,653 (+1 h; p < 0.001). The corresponding values of HR were 62.9 ± 10.2 (−1 h), 76.3 ± 13.6 (ischemic time; p < 0.0001) and 67.1 ± 10.8 (+1, p < 0.0001). The values of SaO2 min were 87.1% ± 5.3% (−1 h), 86.5% ± 5.1% at the time of ischemia NS (not significant) and 91.2% ± 2.5% (+1 h, p < 0.002), respectively. The deepest ST segment deviation was at the rebreathing phase of the obstructive apnea (Fig. 2). Likewise, this phase was also frequently associated with the start of the ischemic event.
Fifteen patients were treated by nasal CPAP of whom nine had nocturnal or morning ischemia during the first study. Overall, CPAP treatment reduced the number of ischemic events from 13 (78 min) to 6 (33 min; p = 0.01, Fig. 3). As expected, OSAS improved significantly under CPAP (AHI: 35.1 ± 20 vs. 10.1 ± 5.7 p < 0.001; SaO2 min: 78.6% ± 6.5% vs. 87.1% ± 5.1%, p < 0.001). Likewise, HR (mean and maximum) and DP values decreased significantly under CPAP (Table 3). However, whenever nocturnal ischemia occurred under CPAP, DP values during the ischemic events remained unchanged by CPAP in comparison with pretreatment values (DP: 12,263 ± 2,456 vs. 13,084 ± 2,781, respectively, NS).
Twenty percent of our patients (10/51) who suffered from IHD and OSAS had ST segment depression during sleep. Moreover, six of them, as well as an additional five patients also had ischemia during the first 2 h after waking up. We should note, however, that patients were studied without beta-blockers and nitrates which would contribute to the high prevalence of ischemic events. Thus, a total of 15 out of 51 patients, i.e., 29%, showed evidence of ischemia during the study. Apart from age and IHD severity, oxygen desaturation and DP values were the main factors that were associated with nocturnal ischemic events. Thus, it can be proposed that an increase in myocardial oxygen demand, and a decrease in its supply, precipitates ischemia in patients with an inappropriate coronary reserve. The control group who had the same severity of OSAS but without any evidence of IHD did not show any sign of ischemia during sleep or during the day. Our results also demonstrate that CPAP treatment significantly ameliorated the ischemic changes during sleep although it did not entirely eliminate them. Based on these results, and previous studies, it appears that the main risk factor for cardiac ischemia during sleep in OSAS is the presence of IHD (6,7,13,16–18).
The prevalence of ischemia found in our study population was similar to that previously reported in similar groups of patients. Koehler et al. (13)showed ischemia in 5 out of 20 patients (OSAS + IHD), mainly during REM sleep; Schäfer et al. (7)showed that 5 out of 14 patients had ischemia during REM sleep and hypoxemia. On the other hand, Olazabal et al. (9)and Hillman (3), who examined similar groups of patients, did not find any significant ischemic changes. Since we did not have a control group of matched IHD patients who were free from OSAS, based on our study we cannot determine if the coexistence of OSAS and IHD is associated with more nocturnal ischemic events than IHD alone. However, the incidence of nocturnal ischemia in IHD patients which is considered to be about 20% (19,20)is probably overestimated. Since the prevalence of OSAS among unselected IHD patients is extremely high (37% – 64%) (9,21,22), it is highly likely that patients with a combination of OSAS and IHD were included in these early studies.
The incidence of ischemia increased through the night and peaked during the first hour after awakening (Fig. 1). This is in agreement with previous findings (23–25), related to late night and early morning increase in sympathetic activity (26)and in platelet aggregability (27). Moreover, the early morning increase in the relative amount of REM sleep, which is associated with both sympathetic activation and prolonged apneas and deeper desaturations, may also trigger the occurrence of ischemic events at this time in OSAS patients. However, in contrast with previous findings (7,13), we did not find any significant association between REM sleep and ischemia.
Continuous positive air pressure treatment
Treatment with CPAP ameliorated ischemia throughout the night. Some factors may explain this improvement: the decrease in HR under CPAP (Table 3), the increase in oxygen saturation and the decrease in AHI which is followed by a decrease in the negative intrathoracic pressure, i.e. decrease of myocardial wall stress (30). However, since monitoring under CPAP was conducted after a single night of treatment, the effects of a longer treatment period should be investigated. The lack of improvement in the ischemic time during the morning hours (6:00 amto 8:00 am) may be due to a short period of treatment. Daytime improvement of hypertension was achieved only by long-term CPAP treatment (31).
The changes in heart rate and DP values before, during and after the ischemic episodes, and the lack of similar changes in the level of oxygen saturation, may suggest that the main cause of ischemia was the increase in oxygen demand and not in its supply. Furthermore, the fact that the deepest ST segment depression occurred at the rebreathing phase of the obstructive apnea, which is characterized by marked tachycardia and sympathetic surge (28,29), also points to the profound impact of oxygen consumption underlying the ischemic event. Likewise, the increase in myocardial wall tension during the obstructive apnea (30)and the elevation of BP and HR during the rebreathing phase (32)reduces the oxygen supply to the myocardium, due to shortening of the diastolic phase and also due to stronger intramural squeezing forces combined with increased oxygen demand. Therefore, at this phase, maximal oxygen consumption coexists with a reduced supply, a condition which could trigger the ischemic events.
The analysis of the effects of CPAP on the ischemic events also supports the above conclusion. Although CPAP treatment significantly ameliorated nocturnal ischemia, it did not eliminate ischemia entirely. The ischemic events that remained under CPAP treatment were associated with DP values that were very similar to the corresponding pretreatment values, although overall DP and SaO2 < 90% improved under CPAP (Table 3). These findings may also support the suggestion that in OSAS patients with inadequate coronary reserve, the main cause of nocturnal ischemia is the increased myocardial oxygen consumption rather than the reduction in oxygen supply. Thus, in order to minimize nocturnal ischemic events in these patients, a combination of CPAP treatment and medical therapy to control HR and BP increases is warranted.
In summary, OSAS may exacerbate ischemia during sleep when inappropriate coronary reserve coexists (and remains untreated). We suggest that the increase in myocardial oxygen consumption is the main cause of nocturnal ischemia although oxygen supply decreases during OSAS as well.
- apnea hypopnea index
- analysis of variance
- blood pressure
- continuous positive air pressure
- double product
- heart rate
- ischemic heart disease
- myocardial infarction
- obstructive sleep apnea syndrome
- rapid eye movement
- oxygen saturation
- vessel disease
- Received August 24, 1998.
- Revision received June 15, 1999.
- Accepted August 6, 1999.
- American College of Cardiology
- Koskenvuo M,
- Kaprio J,
- Telvakivi T,
- Partinen M,
- Heikkila K,
- Sarna S
- Fletcher E.C,
- Lesske J,
- Behm R,
- Miller C.C,
- Stauss H,
- Unger T
- Van Den Aardweg J.G,
- Karemaker J.M
- Keyl C,
- Lamberger P,
- Rodig G,
- Dambacher M,
- Frey A
- Quyyumi A.A,
- Efthimiou J,
- Quyyumi A,
- Mockus L.J,
- Spiro S.G,
- Fox K.M
- Rechtschaffen A,
- Kales A
- Lavery C.E,
- Mittleman M.A,
- Cohen M.C,
- Muller J.E,
- Verrier R.L
- Van den Aardweg J.G,
- Karemaker J.M