Circadian blood pressure pattern in sarcoidosis patients

Abstract

Objectives. Autonomic dysfunction (AD) is frequent in sarcoidosis and considered a result of small fiber neuropathy. A non-dipper blood pressure (BP) pattern, which is also linked to AD, is associated with increased risk of cardiovascular and renal diseases. The aim of the present study was to evaluate the non-dipping BP pattern in normotensive patients with pulmonary sarcoidosis (PS). Methods. Sixty-three normotensive patients with PS (group 1) and 49 healthy subjects (group 2) were prospectively enrolled. Ambulatory BP monitoring was performed in all participants over a 24-h period. Results. The non-dipping BP pattern was significantly more frequent in patients with PS compared with the control group (80% vs 53%, respectively, p = 0.002). More advanced PS (grade 2) was an independent predictor of non-dipper BP pattern (odds ratio = 10.4, 95% confidence interval 1.1–95.4, p = 0.03). Masked hypertension and body mass index were also found to be other predictors of non-dipping BP pattern. Conclusions. The present study showed that non-dipping BP pattern is frequently observed in normotensive patients with PS. The probable mechanism underlying the non-dipping BP in PS is autonomic nervous system dysfunction. PS represents an independent risk factor for non-dipping BP and these patients have increased cardiovascular risk.

Key Words::

• Ambulatory blood pressure monitoring
• cardiovascular risk
• non-dipping blood pressure
• pulmonary sarcoidosis

Introduction

Sarcoidosis is a multisystem disease of unknown etiology that is characterized by the non-caseating granulomas in involved tissues. It may involve any tissue in the human body. The granuloma consists of mononuclear phagocytes and lymphocytes, and is thought to be triggered by infectious and environmental antigens (1). Cardiovascular involvement is fairly common in sarcoidosis, and patients with cardiac sarcoidosis may present with arrhythmias, conduction abnormalities, congestive heart failure (HF) and sudden cardiac death (1). In addition to direct cardiac involvement, sarcoidosis may involve small nerve fibers, resulting in cardiac autonomic dysfunction (AD) (2). Cardiac AD is demonstrated in patients with systemic sarcoidosis, and is even more prominent with those with cardiac involvement and may contribute to sarcoidosis-associated morbidity and mortality (3).

Blood pressure (BP) has a circadian variation that decreases during the night and increases at the morning (4). On the basis of percentage reduction in the circadian BP, patients could be classified into several distinct patterns including dipper pattern (average systolic BP [SBP]> 10% higher when awake), non-dipper pattern average (SBP < 10% higher when awake) and reverse-dipper pattern (nocturnal rise in SBP) (5,6). A non-dipping BP pattern is particularly important, as it is associated with hypertension (HT) development, obstructive sleep apnea syndrome (OSAS), obesity, diabetes mellitus (DM), renal dysfunction and diseases characterized by altered autonomic cardiovascular regulation (5,7–9). It is suggested that a non-dipper BP pattern is related to AD (10).

As AD is a known feature of sarcoidosis, and is involved in the genesis of certain abnormal BP patterns, particularly the non-dipper pattern, we hypothesized that a non-dipper BP pattern could be more frequent in these patients. Therefore, we aimed to investigate the BP patterns in pulmonary sarcoidosis (PS) patients without known systemic HT or cardiovascular involvement.

Materials and methods

Patients

Between June and December 2013, 63 patients with biopsy-proven grade 1 or 2 PS (mean age: 44.7 ± 10.3 years; 22 males) and 49 healthy subjects (mean age 40.8 ± 14.4 years, 21 males) were prospectively enrolled. Patients with known HT, DM, HF, atrial fibrillation (AF), coronary artery disease (CAD), more than mild valvular disorder, and/or thyroid, liver (ALT/AST ≥ 3 × upper value) and renal disease (glomerular filtration rate [GFR] < 30 ml/min/1.73 m²), dyslipidemia (a total cholesterol level > 240 mg/dl or taking lipid-lowering agents), obesity (body mass index [BMI]≥ 30 kg/m²), collagen vascular diseases, any cardiovascular drug use, current or recent (≥ 3 months) steroid use were excluded. Also, patients who had a history of a respiratory disease such as asthma, chronic obstructive pulmonary disease, emphysema or chronic bronchitis were excluded from the study. Patients with current or previous (within the last 6 months) use of oral contraceptives, hormone replacement therapy, non-steroidal anti-inflammatory drugs, methotrexate, hydroxyurea, mycophenolate mofetil, azathioprine, oral retinoids, ciclosporin or adalimumab were excluded from the study. Participants who reported that their sleep was severely disturbed when wearing the ambulatory BP monitoring (ABPM) device were also excluded.

Study protocol

All patients underwent physical examination including BP and BMI measurements. PS patients also underwent pulmonary function testing. PS was graded using standard radiographic criteria (11). The study protocol was approved by the Ethics Committee of the Bezmialem Vakif University Faculty of Medicine.

Laboratory data

Blood samples were collected after 12-h overnight fasting and analyzed by the hospital clinical laboratory using standard methods. Biochemical and hematological parameters were measured by an Olympus AU 600 autoanalyzer (Olympus Corp., Tokyo, Japan) and a Bayer Advia 120 Cell CBC Counter Hematologia autoanalyzer (Bayer Advia 120 CBC counter, NJ, USA).

Office blood pressure measurement

Office BP was measured from the brachial artery using a mercury sphygmomanometer (ERKA D-83646 Bad Tölz, Kallmeyer Medizintechnik GmbH & Co. KG, Germany). Measurements were taken twice with 1-min intervals in a quiet environment with the patient in a sitting position after 5 min of rest. BP was not measured if the patient had consumed tobacco, ingested caffeine or eaten within 30 min. An appropriate cuff size was chosen for each subject. Systolic and diastolic BPs were taken as the first and fifth phases of Korotkoff sounds. The average of these measurements was used to determine office BP. If the readings differed by 5 mmHg, an extra reading was obtained. The average of these three readings was defined as office BP. None of the patients’ office BP was ≥ 140/90 mmHg.

Ambulatory blood pressure monitoring

Twenty-four-hour ABPM was recorded using a validated non-invasive automatic device (Tonoport V, PAR Medizintechnik GmbH, Berlin, Germany) and the cuff of the device was fitted on the non-dominant arm. The frequency of measurements was set to once every 15 min throughout the 24 h. Patients were instructed to maintain their daily routines and sleep patterns. Strenuous physical activity was discouraged in all patients during the monitoring period. BP reading was edited by a computer and rejected if the systolic BP was less than 80 mmHg or greater than 250 mmHg, or if the diastolic BP was less than 40 mmHg or greater than 140 mmHg. Recordings for each subject were accepted if more than 80% of the raw data were valid. Daytime and night-time ambulatory BPs (ABPs) were defined through the use of a diary reporting the time of awakening and retiring. Average 24-h ABP of less than 130/80 mmHg, daytime ABP of less than 135/85 mmHg and average night-time ABP of less than 120/70 mmHg were considered normotensive (12,13). Patients with ABP above these limits but normal office BP were considered masked hypertensive (14). Non-dippers were defined as those who showed a reduction in BP of less than 10% between the average daytime and night-time SBPs. Dippers were defined as those SBP decrease was ≥ 10% (15). Patients with increased BP at night were defined as reverse-dipping. The average of BPs during the first 4 h after waking was defined as morning BP (16). Sleep-trough morning BP surge was defined as the difference between the average BP during the 2 h after awakening and the lowest night-time BP (i.e. the average of the lowest BP and the two readings immediately preceding and after the lowest value). Pre-awakening morning BP surge was defined as the difference between the average BP during the 2 h after awakening and the average BP during the 2 h before awakening (17,18). The 24-h standard deviation (SD) BP was defined as a mean of day and night SD values corrected for the number of hours included in each of these subperiods (19).

Statistical analysis

Qualitative variables were expressed as percentages (%) and quantitative variables as means± SD. Among-group features were compared by analysis of variance for continuous variables and using χ² statistics for categorical variables. Comparisons of parametric values between two groups were performed using the two-tailed Student's t-test. A backward stepwise multivariate logistic regression analysis, which included variables with p-value less than 0.1 on univariate analysis, was performed to identify independent predictors of non-dipper pattern. A p-value less than 0.05 was accepted as statistically significant. All statistical studies were performed using SPSS software (version 15.0, SPSS, Chicago, IL).

Results

Demographic, clinical and laboratory characteristics of study group were shown in Table I. Baseline characteristics of patients such as age, sex, BMI, laboratory values and smoking status were not different between the groups. Table II gives the ABPM findings in the study group. Most of the ABPM parameters and office BP levels were not significantly different between PS and control patients, but the non-dipper pattern was more common in the PS group compared with controls (80% vs 53%, p = 0.002). Day–night SBP difference was lower in PS patients compared with controls (p = 0.03). There were no significant differences between groups regarding minimum, maximum or mean heart rate, although both minimum and mean heart rate were slightly higher in the sarcoidosis group. Univariate analysis determined six variables as significant predictors of a non-dipper BP pattern (Table III). Multivariate analysis indicated that the best predictive variables for non-dipping pattern were a BMI > 25 (odds ratio [OR] = 8.94, 95% confidence interval [CI] 1.1–71.4, p = 0.04), masked HT (OR = 13.6, 95% CI 1.7–103.2, p = 0.01) and stage 2 PS (OR = 10.4, 95% CI 1.1–95.4, p = 0.03).

Table I. Demographic, clinical and laboratory characteristics of study group.

Parameter	Sarcoidosis, n = 63	Control, n = 49	p
Age, years	44.7 (10.3)	40.8 (14.4)	0.20
Body mass index, kg/m^2	28.8 (5.2)	28.9 (3.3)	0.92
Male	22 (35)	21 (42)	0.39
Current smoking	14 (22.2)	7 (14.2)	0.65
Creatinine, mg/dl	0.74 (0.14)	0.78 (0.24)	0.49
Low density lipoprotein, mg/dl	128.9 (34.9)	130.3 (28.4)	0.90
Triglyceride, mg/dl	137.2 (51.4)	157.3 (67.8)	0.31
Glucose, mg/dl	101.5 (14.4)	97.2 (21.3)	0.37
Serum calcium, mg /dl	9.4 (0.47)
Stage
Stage 1	34 (54%)
Stage 2	29 (46%)
Disease duration (month)	48.3 (43.9)
Angiotensin converting enzyme, U/l	66.7 (5.9)
FEV1,%	91.9 (16.7)
FEVC, %	82.1 (22.6)
Diffusion capacity of lung for carbon monoxide, %	98.9 (17.6)
Extrapulmonary sarcoidosis	15 (23%)

Table II. Ambulatory blood pressure monitoring findings in study group.

Parameter	Pulmonary sarcoidosis, n = 63	Control, n = 49	p
Office blood pressure
Systolic	120.4 (15.9)	116.7 (17.1)	0.24
Diastolic	77.5 (9.4)	76.2 (8.7)	0.45
Average 24-hour blood pressure, mmHg
Systolic	119.4 (11.9)	120.3 (9.76)	0.68
Diastolic	75.3 (7.8)	74.4 (7.4)	0.53
Mean	89.4 (8.5)	89.4 (7.9)	0.98
Average day-time blood pressure, mmHg
Systolic	121.5 (11.4)	122.6(9.9)	0.63
Diastolic	77.6 ((7.8)	76.5(7.4)	0.47
Mean	91.3 (9.5)	91.2 (8.1)	0.99
Average night-time blood pressure, mmHg
Systolic	114.6 (13.6)	112.9 (11)	0.49
Diastolic	70.1 (8.7)	67.5 (8.5)	0.11
Mean	84.5 (9.7)	82.6 (8.9)	0.3
Average morning blood pressure, mmHg
Systolic	123.3 (16.6)	123.7 (11.9)	0.86
Diastolic	78.1 (11.6)	77.4 (9.8)	0.74
Average pre-awakening blood pressure, mmHg
Systolic	112 (14.2)	109.3 (12.7)	0.31
Diastolic	68.2 (10.7)	64.3 (10.6)	0.06
Minimum blood pressure, mmHg
Systolic	101.5 (13.9)	97 (13.9)	0.09
Diastolic	57.6 (10.2)	54.4 (9.4)	0.09
Sleep-trough blood pressure surge, mmHg
Systolic	22.3 (11.4)	26.1 (12.3)	0.11
Diastolic	21.2 (10.3)	23.1 (9)	0.33
Pre-awakening blood pressure surge, mmHg
Systolic	12.2 (7.8)	14.5 (9.7)	0.18
Diastolic	11 (7.5)	13.6 (7.7)	0.08
24-h SD blood pressure, mmHg	10.7 (4.2)	10.8 (4.1)	0.9
Non-dipping pattern	51 (80)	26 (53)	0.002
Day–night SBP difference	6.9 (6.2)	9.6 (6.7)	0.03
Reverse dipper pattern	7 (11)	2 (4)	0.28
Masked hypertension	34 (53.9)	21 (42.8)	0.24
24-h heart rate
Minimum	54.7 (10.5)	52.4 (10)	0.2
Maximum	107.7 (20.7)	107 (14.4)	0.9
Mean	77.9 (8.7)	76.4 (8.7)	0.3

Mean values (SD) and n (%) are reported for continuous and categorical variables, respectively. SD, standard deviation.

Table III. A binary logistic regression model for the association of clinical variables with non-dipping hypertension in the whole study population (n = 112).

	Univariate			Multivariate
Variables	OR	95% CI	p	OR	95% CI	p
BMI > 25 kg/m^2	3.79	1.19–12	0.02	8.94	1.1–71.4	0.04
Masked HT	6.62	2.56–17.1	< 0.001	13.6	1.7–103.2	0.01
Sleepthrough SBP	0.89	0.86–0.94	< 0.001	0.93	0.83–1.4	0.15
Pre-awakening SBP	0.91	0.86–0.96	< 0.001	0.95	0.87–1.6	0.11
Stage 1 PS	2.88	1.09–7.6	0.03	1.1	0.42–9.8	0.26
Stage 2 PS	5.53	1.67–18.27	0.005	10.4	1.1–95.4	0.03

OR, odds ratio; CI, confidence interval; BMI, body mass index; HT, hypertension; SBP, systolic blood pressure; PS, pulmonary sarcoidosis.

The frequency of the non-dipping pattern was 53.1% in healthy controls, whereas this ratio increased to 76.5% in patients with grade 1 PS and to 86.2% in patients with grade 2 PS. The ratio of patients with non-dipping pattern was significantly higher in grades 1 and 2 PS compared with healthy controls (p = 0.03 for grade 1 PS vs controls and p = 0.003 for grade 2 PS vs controls). A non-dipping pattern was more frequent in grade 2 PS compared with grade 1 PS, but this finding did not reach statistical significance (86.2% vs 76.5%, p = 0.32).

Discussion

In this study, we compared the circadian rhythm of BP in patients with PS and healthy controls. We demonstrated that a non-dipper BP pattern was more frequent in PS patients, compared with healthy volunteers. Grade 2 PS was also a predictor of non-dipper BP pattern, demonstrating a relationship of disturbed circadian BP rhythm with severity of PS. To the best of our knowledge, this is first study to demonstrate a blunted nocturnal BP decrease in patients with PS.

In normal conditions, BP declines on average by 10–20% with sleeping, this termed a dipping BP pattern (20). Alterations in this normal decline patterns are known as a non-dipper pattern (less than 10% decline at night-time), an extreme-dipper pattern (more than 20% decrease during night-time) and a reverse-dipper pattern (nocturnal BP rises above the wake-time level). Physiologically, BP peaks twice during the daytime, when the first peak is observed just before the beginning of daily activities around 08:00–10:00 h and the latter is seen at 19:00–21:00 h. Normally, no increase in BP is expected during sleep-time, where the BP should be lower than in the daytime hours (21). These circadian rhythm patterns are determined by numerous factors in the body and exogenous environment factors such as neurohormonal regulation, mental activity and emotional state, cigarette smoking, alcohol consumption, physical activity and dietary sodium intake (22,23). Non-dipper and reverse- dipper BP patterns were shown to increase the risk of cardiovascular and kidney injury independent of mean BP values (24,25). Lurbe et al. (26) found that in patients with type 1 DM, an increase in systolic BP during sleep precedes the development of microalbuminuria. Thus, the risk of microalbuminuria was very low in patients who remain normotensive in sleep and an increase in night-time systolic BP could be the earliest sign of altered BP regulation in patients with type 1 DM. In addition, subjects with a non-dipper BP pattern have more severe renal and cardiovascular disease, and the risk was most profoundly increased in those with both a nocturnal HT and non-dipper BP pattern (27). Also, non-dipping BP was associated with cerebrovascular events, insulin resistance, left ventricular hypertrophy, carotid intima-media thickening, HT development and secondary HT (8,28,29). Markers associated with adverse cardiovascular outcomes, including red cell distribution width, mean platelet volume, inflammatory markers (CRP), fibrinogen and asymmetric dimethyl arginine, were all increased in those with non- dipping BP pattern (30–34). These studies emphasize the importance of detecting a non-dipping BP pattern for cardiovascular risk stratification, reduction and prognosis.

It is considered that autonomic nervous system (ANS) and catecholamine levels play a major role in circadian BP regulation (35). Norepinephrine and epinephrine concentrations have a significant diurnal variation, with peak levels typically in the morning, and lowest levels during the late evening and initial hours of night-time sleep (36). Nielsen et al. (37) measured the plasma noradrenaline and adrenaline levels for the sympathetic nervous activity in 31 hypertensive and diabetic nephropathy patients and found that sleep-time changes in BP were closely related to nocturnal noradrenaline levels. Similarly, another study has demonstrated that the night-time fall in urinary norepinephrine and epinephrine excretion was reduced in those with non-dipper BP, and non-dippers demonstrate a heightened alpha 1-AR responsiveness compared with dippers (10).

Sarcoidosis commonly involves the lung, although any organ may be involved in the body. AD, which is related to small fiber neuropathy in autonomic nerves, can be observed in sarcoidosis patients with neural involvement (2). Previous observations regarding abnormal heart rate reflexes in sarcoidosis patients are considered a manifestation of AD in cardiovascular system. Uslu et al. (3) evaluated the heart rate variability (HRV), which is a useful method to measure autonomic activity, in patients with systemic sarcoidosis. They found the decreased HRV values in patients with systemic sarcoidosis reflect a reduction of vagal activity with a concomitant sympathetic dominance. In addition, this reduction was more obvious in patients with cardiac sarcoidosis. Similarly, the heart rate recovery index, which is an indicator of post-exerional parasymphatetic activity, was altered in patients with sarcoidosis (38). These studies confirm the altered symphato-vagal balance in patients with sarcoidosis. We found that a non-dipper BP pattern was more common in PS patients compared with healthy controls. This condition may be explained by alterations in the ANS activity in PS patients, as a non-dipper BP pattern is considered a reflection of disturbed ANS activity.

Sudden cardiac death, which could be caused by atrioventricular block, severe ventricular, arrhythmia or asystolic arrest, is a prominent cause of mortality in sarcoidosis patients (39). It has been shown that hypertensive patients with a non-dipper profile have a prolonged ventricular repolarization throughout the 24-h period, which is absent either in dippers or normotensives (40). Another report showed that non-dipper hypertensive patients are likely to experience supraventricular and ventricular arrhythmias more frequently than dippers (41). This tendency to arrhythmias observed in a non-dipper BP pattern most probably reflects underlying cardiac AD, which is frequently observed in sarcoidosis patients as stated before. Therefore, a non-dipper BP pattern may also serve as a marker of arrhythmic tendency in sarcoidosis patients, which could be used for risk stratification of these patients.

While AD remains the most probable cause for the non-dipper BP pattern observed in sarcoidosis, our study was not adequate to prove this relationship, as it is not designed to demonstrate AD. Inflammation could be a common link between systemic HT and all autoimmune diseases, including sarcoidosis (42,43). Although inflammation is associated with increased BP and inflammatory biomarkers are elevated in non-dipper BP pattern, to date no study has shown a causative link between inflammation and non-dipper BP (44,45). Non-caseating granulomas in sarcoidosis patients produce angiotensin-converting enzyme (ACE), whose over-production may result in HT. In our study, we failed to show a significant increase in ACE levels in PS patients with a non-dipping pattern compare with a dipping pattern. Therefore, our findings could not be explained with a concomitant increase in ACE.

A final consideration that should be emphasized is the potential contribution of OSAS to AD and the non-dipping BP pattern. Obstructive sleep apnea is a well-known cause of AD and non-dipper BP pattern (46,47). Patients with sarcoidosis have increased incidence of OSAS compared with the general population, and OSAS tended to be more severe in these patients. The exact contribution of OSAS to the increased frequency of non-dipping pattern cannot be assessed in this present study, since no polysomnographic data was collected from these patients, and none of the patients had known OSAS prior to enrollment. Since only one-fifth of sarcoidosis patients have co-existent OSAS, the degree of contribution should be limited, while further studies were needed to ascertain this effect.

Study limitations

This study had several limitations. First, the size of our patient group was relatively small. This study was solely designed to investigate BP patterns in patients with PS, so we did not look for long-term implications of BP patterns or relationship with arrhythmias. This is particularly important, not as the presence of nocturnal HT per se, but the detrimental effects of non-dipping BP on target organs have clinical significance and a long-term prognosis (14,24,48). Therefore, the lack of information regarding clinical events or target organ damage (such as left ventricular hypertrophy or atherosclerotic changes in large arteries) limits the value of the current study, though the potential risks associated with the presence of non-dipping BP pattern can be deduced from the previous literature (14,24,48). Also, no work-up was done to investigate the potential causes of nocturnal HT, such as an evaluation of ANS function by measuring the catecholamine levels or assessing the presence and severity of inflammation with inflammatory markers. As this study was preliminary and observatory in nature, further studies in patients with PS are needed in order to elucidate the relationship between the non-dipping BP pattern and the clinical outcomes, target organ damage or ANS function. Additionally, some variables that could affect circadian BP profile, such as the potential contribution of associated diseases, including OSAS, and exposure to noise or decreased sleep quality were not controlled.

We found that the day and night SBPs were similar between the groups; the frequency of non-dipping pattern was significantly higher in PS group. Although mean SBP did not differ between groups, patients with PS had significantly lower circadian BP variation (i.e. day–night BP difference) compared with healthy controls, which may explain this apparent discrepancy. Also, a relatively low sample size in both groups, which limits the power of study, might have contributed to these discrepant results.

Conclusion

In conclusion, this study demonstrates that a non-dipping BP pattern is highly prevalent in normotensive patients with PS. Although the underlying mechanisms linking these two diseases are unclear, ANS dysfunction/sympathovagal imbalance seems a possible cause of a non-dipping BP pattern. As non-dipper BP pattern is associated with unwanted cardiovascular outcomes, this study calls attention to the importance of ABPM in patients with PS. Also, high-grade PS is a predictor of a non-dipper BP pattern, which could translate into a higher frequency of CV complications in advanced PS, even when direct cardiac involvement is not present. This readily available non-invasive measurement may be clinically useful in the identification of high-risk patients with PS. Further prospective studies are warranted to elucidate the prognostic implications of our findings in PS patients.

Conflicts of interest and grant support: There are no conflict of interest issues. We have no grant support.