Sleep quality among relatively younger patients with initial diagnosis of hypertension: Dippers versus non‐dippers

Abstract

Background. Sleep is a basic physiological process. Normal sleep yields decrease in sympathetic activity, blood pressure (BP) and heart rate. Those, who do not have expected decrease in their BP are considered “non‐dippers”. We aimed to determine if there was any association between the non‐dipping status and sleep quality, designed a cross‐sectional study, and enrolled and evaluated the sleep quality of relatively young patients with an initial diagnosis of hypertension. Methods. Seventy‐five consecutive patients, diagnosed to have stage 1 hypertension by their primary physicians, were referred to our study. Patients had newly diagnosed with stage 1 hypertension. Patients with a prior use of any anti‐hypertensive medication were not included. Eligible patients underwent the Pittsburgh Sleep Quality Index (PSQI), which has an established role in evaluating sleep disturbances. All patients underwent ambulatory BP monitoring. Results. There were 42 non‐dipper patients (mean age = 47.5±11.9 years, 24 male/18 female), as a definition, 31 dipper hypertensive patients (mean age = 48.5±12.8 years, 21 male/10 female) and two with white coat hypertension. Daytime systolic and diastolic mean BPs were not significantly different between the two groups. Night‐time mean systolic and diastolic BPs were significantly higher in non‐dippers compared with dippers. PSQI scores, globally, were significantly higher in non‐dippers compared with dippers. Total PSQI score was not correlated with body mass index. It was noticed that, individually, sleep quality, sleep efficiency and sleep disturbance scores were significantly higher in non‐dippers. Being a poor sleeper in terms of high PSQI score (total score>5) was associated with 2.955‐fold increased risk of being a non‐dipper (95% confidence interval 1.127–7.747). Conclusion. We showed that the risk of having non‐dipping hypertension, a risk factor for poor cardiovascular outcomes among hypertensive individuals, was tripled (odds ratios) among poor sleepers. We think that evaluating sleeping status and sleep quality among the hypertensive population may help unmask non‐dipper hypertension, enabling physicians to treat appropriately.

• Hypertension
• non‐dipping
• Pittsburgh sleep quality index
• sleep quality

Introduction

Sleep, a basic physiological process, is an essential part of healthy life. It provides restorative, adaptive and energy conserving functions [1]. Normal sleep yields decrease in sympathetic activity, blood pressure (BP) and heart rate. Systolic BP on average decreases 10–20 mmHg at night in both normotensive and hypertensive individuals. This is referred to as BP “dipping”. Those who do not have expected decrease in their BP are considered “non‐dippers”. These definitions are somewhat arbitrary; however, they have shown a remarkable relationship to cardiovascular outcomes. Non‐dipping appears to be associated with the progression of renal disease and increased cardiovascular morbidity [2]. Non‐dipping BP at night among hypertensive individuals may cause the heart no respite from an increased workload. Night‐time BP was shown to correlate better with left ventricular mass than daytime BP [3]. Left ventricular hypertrophy was present in 23% of non‐dippers versus 5% of dippers [3]. The incidence of cardiovascular events increased linearly with increasing left ventricular mass, and was almost three times more in non‐dippers compared with dippers [4]. Sleep apnea is a common problem, associated with hypertension, renal failure and adverse cardiovascular outcomes [5]. It is most frequently observed in elderly obese men. When a non‐dipping BP is found in an elderly, overweight man, it will often be caused by sleep apnea. However, among young individuals and women, the likelihood of sleep apnea is lower. Non‐dippers were thought to have a worse sleep profile than dippers with less deep sleep and more micro‐arousals [1]. In order to understand what percentage of patients with the initial diagnosis of hypertension had disturbed sleep quality and non‐dipping BP status, and if there existed any association between the two, we designed a cross‐sectional study, and enrolled and evaluated the sleep quality of relatively young patients with the initial diagnosis of hypertension.

Material and methods

Seventy‐five consecutive patients, who were diagnosed to have stage 1 hypertension by their primary physicians, were referred to our study (Table I), and all patients gave informed consent. Patients were men and non‐pregnant women aged>18 years, but <65 years of age, who were newly diagnosed with stage 1 hypertension, defined as average of seated systolic BPs>140 mmHg, but <160 mmHg, and seated diastolic BP>90 mmHg, but <100 mmHg. Since, some antihypertensive drugs might have disturbed the sleep, such as diuretics, patients with a prior use of any anti‐hypertensive medication were not included.

Table I. Flow chart of study population.

•  102 consecutive patients with the initial diagnosis of hypertension, considered for the study by primary physicians

•  Patients with valvular heart disease (both mitral stenosis, 1 had primary insomnia problem) n = 2, night‐time workers n = 6, patients with irregular sleep due to work n = 4, patients with history of chronic obstructive pulmonary disease n = 5 (2 of them had BMI>30 kg/m^2), habitual snorers n = 3, patients with paroxysmal atrial fibrillation n = 4, patients with heart failure (ischemic) n = 2, pregnancy n = 1 were excluded according to study protocol

•  Remaining n = 75 was referred for the study

•  Evaluation by the chest physician for sleep apnea/hypopnea

•  Remaining n = 75

•  PSQI was performed on all patients

•  Remaining n = 75

•  ABPM was performed on all patients

•  Excluded due to presence of white coat hypertension n = 2, Remaining n = 73

•  42 patients with non‐dipper, 31 patients with dipper hypertension

Once patients had signed the informed consent form, the screening phase began. A complete physical examination was performed, including a 12‐lead electrocardiogram and standard laboratory tests. Patients were asked to define the quality of their sleep and only those with adequate recordings who reported a normal sleep or a sleep like that in the previous nights were included both in terms of ambulatory BP recordings and Pittsburgh score.

Eligible patients underwent the Pittsburgh Sleep Quality Index (PSQI), which has an established role in evaluating the sleep disturbances [6], by an author, who was blinded to ambulatory BP results. PSQI is made up of individual scores (S1: sleep quality, S2: sleep latency, S3: sleep duration, S4: sleep efficiency, S5: sleep disturbance, S6: use of sleep medication and S7: day time dysfunction) with higher scores indicating poorer sleep parameters. The higher the PSQI score is, the poorer the sleep quality. Total score of more than five designates poor sleeping [6].

All patients underwent ambulatory BP monitoring (ABPM) by a previously validated system [7]. The monitor was mounted on the non‐dominant arm and removed 24 h later. A mercury sphygmomanometer was initially attached to the monitor through a Y‐connector to ensure agreement between the two modes of measurement (within a range of 5 mmHg). Cuff size was according to arm circumference. The average of two to three initial measurements, taken by a trained technician after the subject was in a sitting position for 5 min, was considered the subject's manual BP. Monitoring was performed in a day of typical activity. Ambulatory BP readings were obtained for 24 h at 20‐min intervals from 06.00 to 22.00 h, and at 30‐min intervals from 22.00 to 06.00 h. Average daytime (awake period), average night‐time (asleep period, defined as the period from falling asleep to awakening and not as just the time in bed based on the patient's diary), and average 24‐h systolic and diastolic BP measurements were evaluated. Daytime and night‐time borders were considered according to the evaluation of patients during the removal of ABPM. Each patient and his/her partner were asked for the time the patient fell asleep, and woke up at what time in the morning, and hence, dipping status was defined accordingly. Recordings were automatically registered. Ambulatory hypertension was defined as average daytime systolic BP⩾135 mmHg or diastolic BP ⩾85 mmHg. Patients were defined as non‐dippers when the night‐time systolic and diastolic BP decrease was <10% [8], [9] and as dippers when decrease in night‐time systolic BP was ⩾10% of the mean daytime values [9], as shown previously.

Patients with any of the following were excluded from participation: night‐time workers; valvular heart disease, due to probable arrhythmic events (rheumatic heart disease); chronic obstructive pulmonary disease, which was identified up on discretion of a chest physician; patients with a history of sleep apnea/hypopnea; habitual snorers, recent acute coronary syndrome (within the last 2 months); significant systemic disease; history of any psychiatric disease including primary insomnia; symptomatic cerebrovascular disease (including previous transient ischemic attack within 6 months); obstructive coronary artery disease; any atrioventricular conduction disturbance more than first‐degree block; history of any arrhythmia; history of congestive heart failure; hepatic failure; severe hypertension (>210/120 mmHg); serum creatinine>1.4 mg/dl; history of diabetes mellitus (or fasting blood sugar of >125 mg/dl); pregnant women and women of childbearing potential who were not using an appropriate method of contraception; patients with hypo‐hyperthyroidism; patients having body mass index>25 kg/m²; and patients who consume alcohol regularly. Hence the study was conducted on non‐overweight and relatively young participants (exclusion of elderly population having>65 years). Patients who were using medications for other reasons including beta‐blocking agents (which are known to influence sleep quality negatively), diuretics (which disturb sleep quality because of increased urine output), major psychotropic agents, oral steroids or daily non‐steroidal anti‐inflammatory drugs, and high‐dose acetylsalicylic acid were not included into the study. None of the patients had sleep apnea/hypopnea, which was judged by the thorough evaluation of the chest physician, blinded to our study.

Statistical analysis

Parametric data were expressed as mean±standard deviation, and categorical data as percentages. SPSS 10.0 was used to perform statistical procedures. Parametric data were compared by Mann–Whitney U‐test, and categorical data via the chi‐square test. Spearman's correlation test was used to search for correlation. A p‐value ⩽0.05 was accepted as significant.

Results

There were 42 non‐dipper patients (mean age = 47.5±11.9 years, 24 male/18 female), as a definition, and 31 dipper hypertensive patients (mean age = 48.5±12.8 years, 21 male/10 female) included in the analysis. No patient in our study group had only diastolic non‐dipper or dipper status, neither there was a patient with reverse or extreme dipper status. Two patients who were diagnosed white coat hypertension were excluded from the analysis. Both groups were similar in terms of age and sex distribution (Table II). Daytime systolic and diastolic mean BPs were not significantly different between the two groups. Night‐time mean systolic and diastolic BPs were significantly higher in non‐dippers compared with dippers (Table II). PSQI scores, globally, were significantly higher in non‐dippers compared with dippers (Table II). Total PSQI score was not correlated with body mass index. It was noticed that individually, sleep quality, sleep efficiency and sleep disturbance scores were significantly higher in non‐dippers (Table II). When patients were classified according to their total PSQI scores (above total score of five indicating impaired sleep quality, poor sleepers), it was found that impaired total PSQI score was significantly associated with non‐dipping BP such that 70.3% (26/37) of poor sleepers were non‐dippers, whereas 44.4% (16/36) of good sleepers were non‐dippers (p = 0.026). Being a poor sleeper in terms of high PSQI score had 2.955‐fold more frequent association with being a non‐dipper (95% confidence interval 1.127–7.747).

Table II. Comparison of non‐dippers versus dippers.

	Non‐dippers (n = 42)	Dippers (n = 31)	p
Age (years)	47.5±11.9	48.5±12.8	0.723
Sex (male/female)	24/18	21/10	0.498
Body mass index (kg/m^2)	22.66±2.31	23.23±2.30	0.306
Daytime systolic BP (mmHg)	141±12	143±13	0.344
Daytime diastolic BP (mmHg)	87±12	90±9	0.238
Night‐time systolic BP (mmHg)	136±15	123±12	<0.001
Night‐time diastolic BP (mmHg)	84±12	76±9	0.004
Average systolic BP (mmHg)	140±12	141±12	0.683
Average diastolic BP (mmHg)	87±12	88±8	0.692
Percent dipping in systolic BP	1.4±5.9	14.3±4.6	<0.001
Percent dipping in diastolic BP	2.3±7.4	16.1±6.5	<0.001
Total PSQI score	6.76±3.11	5.16±2.92	0.029
Mean heart rate during recording (beats/min)	71.5±6	69.8±8.5	0.334
Sleep quality (S1)	1.24±0.76	0.87±0.72	0.040
Sleep latency (S2)	1.17±1.01	1.06±1.03	0.673
Sleep duration (S3)	0.81±0.80	0.84±0.78	0.877
Sleep efficiency (S4)	1.00±0.94	0.52±0.81	0.024
Sleep disturbance (S5)	1.26±0.63	0.81±0.65	0.004
Use of sleep medication (S6)	0.33±0.79	0.16±0.45	0.243
Day time dysfunction (S7)	0.95±1.06	0.90±1.22	0.855

BP, blood pressure; PSQI, Pittsburgh Sleep Quality Index.

Discussion

Hypertension is an important cardiovascular risk factor with considerable morbidity and mortality if left untreated [10]. On the other hand, non‐dipping high BP brings about further increased risk in the context of hypertension [2], though there are contrary reports in the literature stating that only non‐dipping, but not hypertension, was not associated with target organ damage [11]. Hence, it should be properly recognized and treated, if possible, appropriately within the presence of hypertension [12]. Though sleep apnea in association with obesity currently serves as a novel risk factor for non‐dipping hypertension, it tends to be more frequent in elderly individuals [12]. Besides, our study not only excluded obese (>30 kg/m²) but also overweight (BMI>25 kg/m² but <30 kg/m²) individuals, and hence, mean BMI, as a whole group, was 22.90±2.31 kg/m², not significantly different between the two groups. Furthermore, none of the patients in our study group had sleep apnea/hyponea or pulmonary disease or any kind of disease status, which could have disturbed the sleep quality and confounded the results. In our relatively homogenous study group of patients with newly diagnosed hypertension, there was a significant association between non‐dipping status and poor sleep quality. Since no patient in both groups had left ventricular hypertrophy, it was relatively a early phase of hypertension being associated with sleep status. However, it was difficult to judge which of them was the cause or the result, upon showing the significant association between the non‐dipping hypertension and poor sleeping quality, though it was stated that non‐dipping hypertension was associated with volume overload [12]. High BP at night might be disturbing sleep quality simply because of headache, or poor sleep status itself might be inducing non‐dipping pattern via several unknown mechanisms, but not due to sleep apnea/hypopnea‐associated mechanisms. Besides, since we did not evaluate other issues of sleep (number of naps, positions, etc.), we were limited to evaluations by PSQI. This issue remains to be established. Several relatively milder confounding factors, such as mild alcohol consumption, exercise, marital status [13], along with relatively small sample size and exclusion of many potential confounders, might have an impact on the results of our study, particularly in the setting of evaluation of the sleeping status. On the other hand, reproducibility of non‐dipping BP could be said to represent a problem; though it was shown that in newly diagnosed and uncomplicated hypertension, as in our study group, that reproducibility was quite high [14]. Besides, it is not known whether the sleep quality measured at the time of ABPM is directly associated with dipping or non‐dipping status irrespective of reproducibility. However, our study could be said to have a fairly good design in terms of control of established factors, and PSQI provides a well established method for characterizing and quantifying the sleep quality among different health problems [15], [16]. As a limitation, it might be speculated whether non‐dipping hypertension causes a sleep disorder, or if it is the other way around, since our study was not specifically designed to differentiate such a relationship. On the other hand, exclusion of many confounders rendered us limited instead of being generalized in drawing conclusions. However, narrow selection criteria helped avoid multiple comparisons and significance inflation in such a small study. Furthermore, the possible impact of therapies that restore dipping status onto sleep is a matter for further research. One interesting and potentially important issue might be that poor sleep quality might be predominantly related to relatively high BP levels at night rather than non‐dipping status. However, this issue remains a matter for another study.

In conclusion, although our study group did not represent the general population as a whole, within the light of the findings of our study we think that the relative risk of having non‐dipping hypertension, a risk factor for poor cardiovascular outcomes among newly diagnosed hypertensive individuals, was tripled (odds ratio) among poor sleepers. Therefore, it might be critical to evaluate sleeping status and sleep quality among the newly diagnosed hypertensive population in order not to miss the opportunity to treat it appropriately [17].