Abnormal Blood Pressure Dipping Pattern in Smokers

Introduction

Hypertension is a major risk factor for cardiac disorders, stroke, renal diseases, dementia, and many other severe comorbidities.^(1-4) World Health Organization (WHO) estimates that globally 1,28 billion people 30 to 79 years of age have hypertension.⁽⁵⁾ It is actually the cause of death for 10,4 million people annually.⁽⁶⁾ Therapeutic guidelines of hypertension currently focus on average Blood Pressure (BP) levels.^(1-4) However, research points out the importance of particular sequential BP values and BP fluctuations beyond average BP values.^{(7, 8)}

BP is not a static physical feature; instead, it is a highly dynamic biological aspect of life exhibiting regional and temporal variability in the body. As the body has two diurnal phases, one awake and the other one asleep, it tries to maintain two different levels of dynamic stability, with these two periods having their own normals.^(2-4) Thus, BP exhibits a diurnal variation, and normally declines during sleeping by 10% to 20% compared to the awake period.^{(9, 10)}

Normal BP dipping phenomenon reflects an end result of the normal functioning of circadian rhythm, water and sodium regulation, and multiple interplaying neuroendocrine mechanisms.^{(7, 11)} On the other hand, genetic and environmental factors may influence the BP dipping pattern.⁽¹¹⁾ In fact, deviant diurnal variations are shown to be associated with an increased risk of diabetes mellitus, cardiovascular autonomic neuropathy, and overall cardiovascular morbidity and mortality.⁽¹²⁾

Abnormal dipping patterns are extreme dipper, non-dipper, and reverse dipper patterns. If the BP decline is above 20%, it is called as extreme dipper pattern. In the case of a non-dipper pattern, BP declines between 0-10% compared to the awake period. The reverse dipper pattern exhibits a rise in BP during sleeping compared to the awake period.^{(11, 12)} These phenomena occur both in normotensive and hypertensive individuals.⁽¹³⁾ Hypertension types with non-dipper or reverse dipper patterns of BP are associated with increased cardiovascular and overall morbidity and mortality^{(10, 12)}, increased risk of stroke⁽¹⁰⁾, heart failure⁽¹⁴⁾, cognitive impairment, and dementia.⁽¹⁵⁾ Furthermore, DNA damage markers that are shown in hypertensive patients are higher in non-dipper hypertension.^{(16, 17)}

On the other hand, smoking is a major causative risk factor for a variety of comorbidities like many types of cancers^{(18, 19)}, chronic obstructive pulmonary disease⁽²⁰⁾, coronary heart disease, atherosclerosis, cerebrovascular disease, and hypertension^{(4, 21)}. Nevertheless, due to nicotine, tobacco smoking is highly addictive and, thus, comprises another serious and widespread concern of public health.^{(22, 23)} According to WHO data, 22,3% of the total world population, 36,7% of men and 7,8% of women smoke.⁽²⁴⁾ Currently, over 8 million deaths annually are linked to smoking, of which 1.3 million are attributed to secondhand exposure.⁽²⁴⁾

As abnormal dipping patterns are associated with increased morbidity and mortality^(9-14), it is critical to know if there are further associations between smoking and abnormal dipping patterns, which may shed light on the effects of smoking and how to tackle them. Some studies identify a correlation between smoking and hypertension with non-dipper and reverse dipper patterns.^{(25, 26)} Others relate no such correlation^(27-32); some even indicate a negative correlation.^{(13, 33-36)} There are some discrepancies regarding the definitions and classifications of BP dipping patterns, the inclusion criteria of the studies, and the grouping of the patients among these studies, which may account for the conflicting results.⁽¹¹⁾ Therefore, we have primarily aimed to achieve a consistent and clear view regarding the particular relationship between tobacco smoking and abnormal BP dipping in essential hypertension.

Methods

Results

The patients (n=178) were aged between 30-83 years old and consisted of 84 (47,2%) females and 94 (52,8%) males (Table 1). The average age of male and female participants was similar: 54,33±10,61 years for females, and 54,69±12,20 years for males (p=0,711). Anthropometric measurements of patients were distributed normally. The average values were 165,30±9,65 cm for height; 80,41±15,18 kg for weight; 29,35±4,81 kg/m2 for BMI; 99,63±12,68 cm for waist circumference; and 38,69±3,72 cm for neck circumference. BMI values were similar between male and female groups. 78 (43,8%) patients have never smoked, 37 (20,8%) patients had smoked but quit, and 63 (35,4%) patients have smoked actively. Median smoking years among patients with a history of smoking (active or quit; n=100) were 25 (6-50) years. The median smoking amount in terms of pack.years was 23 (1-135). The median time of abstinence among those defined as quit was 15 (1-43) years.

There were 66 (37,1%) dipper, 3 (1,7%) extreme dipper, 76 (42,7%) non-dipper, and 33 (18,5%) reverse dipper patients. 77 (43,3%) of the study population were newly diagnosed with the current ABPM, and 101 (56,7%) of them were already diagnosed with essential hypertension. 102 (57,3%) patients were under medical antihypertensive treatment. Median number of antihypertensives used was 2 (1-5). Newly diagnosed patients’ distribution, antihypertensive use status, and antihypertensive substance counts were similar between the female and male groups. The median score of FNDT was 5 (0-10). There were 30,2% very low; 19,1% low; 7,9% medium; 22,2% high; 20,6% very high degrees of nicotine dependence among smokers. Patient population characteristics are further detailed in Table 1.

Patient characteristics according to BP dipping patterns are detailed in Table 2. There were no significant differences between the normal dipper group and abnormal dipping group with regard to their anthropometric specifications, comorbidities, time of hypertension diagnosis (new versus established earlier), antihypertensive use (present versus absent), antihypertensive count and antihypertensive timing (night versus others). Remarkably, we also checked for an association between abnormal dipping and waist or neck circumference threshold values; however, we did not find significant differences. Dipper BP rates (52,6%) in patients without any history of smoking (n:78) were significantly higher than the rates of dipper BP (25,0%) in the patients with a history of smoking (active or past) (n:100), (p<0,001). Overall, abnormal BP dipping patterns were significantly more frequent among patients with a history of smoking (75,0%) than among those without any history of smoking (47,4%), (p<0,001). Logistic regression analysis showed that patients with a history of smoking were 3,484 (1,693-7,171 CI:0,95 p=0,001) times more likely to have an abnormal dipping pattern (Table 4).

Abnormal dipping patients had significantly higher FNDT Scores (5 (0-10) versus 2 (0-9), p=0,046), longer smoking years (25 (6-50) versus 20 (5-50), p=0,017), and more tobacco smoking exposure in terms of pack.years (25 (3-135) versus 15 (1-75), p=0,023). More specifically, actively smoking and abnormal dipping BP patients were significantly more likely to score equal to or higher than 2 points in the first question (which questions the timing of first cigarettes upon awakening) of FNDT compared to dipper patients, which in turn indicates some further association between more specific traits of nicotine dependence and abnormal dipping BP pattern (Table 2).

Abnormal dipping patients had also higher rates of isolated nocturnal HT (86,2% versus 13,8%, p=0,009) and higher mean nocturnal systolic (129,23±15,38 mmHg versus 117,27±10,30 mmHg, p<0,001) and diastolic (82,09±12,28 mmHg versus 73,48±9,51 mmHg, p<0,001) BP values. 24-hour and awake-time averages of both systolic and diastolic BP levels did not differ significantly (Table 2). Isolated nocturnal hypertension patients (patients with hypertension during sleeping but normal BP levels during awake period) were significantly at greater risk for abnormal dipping BP patterns in the multivariate logistic regression analysis; OR= 4,905( CI: 1,502-16,019) (Table 4). Isolated nocturnal hypertension is defined as hypertension with normal daytime and office BP.⁽⁹⁾ Nocturnal hypertension is not synonymous with non-dipper or reverse dipper hypertension.⁽⁹⁾ Patients with normal dipper patterns may have nocturnal hypertension.⁽⁴⁵⁾

We have found that patients with a history of smoking had significantly higher nocturnal systolic (128,10±15,54 versus 120,56±12,84 mmHg, p=0,001), nocturnal diastolic (80,82±12,60 versus 76,44±10,89 mmHg, p=0,016), and 24-hour mean systolic (133,15±13,52 versus 127,90±12,03 mmHg, p=0,008) BP values; and more blunted systolic (5 ((-11)-21) % versus 10 ((-9)-22) %, p=0,004) and diastolic (6 ((-9)-27) % versus 9 ((-11)-36) %, p=0,020) dipping ratios than those patients without any history of smoking (Table 3).

History of smoking was associated significantly with being male, having diabetes mellitus, dyslipidemia, and metabolic syndrome (Table 3). Remarkably, this association was detected between history of smoking and non-smokers; however, it was not detected between active smoking and not smoking actively (ie quit and never-smokers together).

Discussion

In this study, patients with a history of smoking showed significantly more blunted dipping patterns both for systolic BP [5 ((-11)-22)% for smokers versus 10 ((-9)-22) % for non-smokers, p=0,004] and diastolic BP [6 ((-9)-27) % for smokers versus 9 ((-11)-36) % for non-smokers, p=0,020]. Patients with a history of smoking were 3,484 (1,693-7,171 CI:0,95 p=0,001) times more likely to have an abnormal dipping pattern (Table 4). Age, sex, neck circumference, and BMI showed no significant differences in regression analysis. These parameters were similar between our groups. Other studies reported significantly higher age^{(33, 35)} and BMI⁽³³⁾ values for non-dipping groups. However, we did not detect such differences possibly because of homogeneity in our study population. This homogeneity in terms of anthropometric values may help us observe the effect of other variables (eg. smoking) upon BP dipping pattern.

Some studies reported contrasting results with our findings, e.g., Hermida et al⁽¹³⁾, de la Sierra et al⁽³³⁾, Kaya MG et al⁽³⁴⁾, Chotruangnapa et al⁽³⁵⁾, Boggia et al.⁽³⁶⁾ In fact, these studies reported an inverse relationship between smoking and non-dipping in their study groups. However, there were a number of important limitations in those studies. For example, although Hermida et al had accepted the diagnosis of secondary hypertension as an exclusion criterion, they did not exclude obstructive sleep apnea patients.⁽¹³⁾ Additionally, their non-dipper group had been significantly older, which could have affected the results because it was reported that the non-dipper pattern increased with aging.⁽²⁵⁾ Furthermore, they included only patients under treatment with multiagent antihypertensives, which in turn could bias the results. Lastly, they had exercised their study with 48 hours of ABPM, and this relatively long testing time might have caused discomfort and stress in patients, which in turn might have influenced and masked the actual dipping pattern of the patients.^{(38, 46)}

The study of de la Sierra et al was retrospective, and they did not report exclusion of secondary hypertension patients.⁽³³⁾ In addition, the retrospective character of the study makes it very hard to evaluate the sleep quality of the patients. Also, they reported significant differences between their study groups concerning comorbidities and age, which may have affected the results.⁽²⁵⁾ Again, Kaya MG et al⁽³⁴⁾, Chotruangnapa et al⁽³⁵⁾, and Boggia et al⁽³⁶⁾ did not report any explicit effort to exclude secondary hypertension patients and sleep disturbances, making the results of their studies problematic from our perspective.⁽¹¹⁾ In addition, the retrospective nature of the study of Chotruangnapa et al⁽³⁵⁾ brings about uncertainty about the evaluation of sleep disturbances and the smoking status of patients. Clinical records about smoking may be incomplete because they are taken with different intentions, and sometimes, smoking may not be the primary concern at the clinical meeting. Hence, retrospective studies about smoking have practical limitations. Again, Boggia et al also reported significant differences in age and antihypertensive use between dipping groups.⁽³⁶⁾

Borrelli et al⁽²⁷⁾ reported no significant association between smoking status and dipping status of the patients. However, their study was performed in a CKD setting which has a very different nature in itself compared to essential hypertension, that is determining the patterns of patients’ BP. They reported significant differences in GFR levels of their patients. Besides, they did not exclude other types of secondary hypertension. These factors might have concealed the effect of smoking itself.^{(11, 27)}

Gunebakmaz et al⁽²⁸⁾, Ozcan et al⁽³⁰⁾, Sunbul et al⁽³¹⁾, and Demir⁽³²⁾ have also reported no significant difference. However, these studies were retrospective, and their hypertensive patient groups were heterogeneous, with no report of exclusion of secondary hypertension or sleep disturbances. The heterogeneity of their groups might have concealed the effect of smoking on dipping patterns. Palatini et al⁽²⁹⁾ also reported an absence of association between smoking and non-dipping. However, their study did not report any exclusion of sleep disturbances, which might have affected their results.⁽¹¹⁾

Brotman et al⁽²⁵⁾ and Cai et al⁽²⁶⁾ reported results similar to our study regarding the association between smoking and non-dipping BP patterns. However, their studies were retrospective, and secondary hypertension patients were not excluded, which caused uncertainty about their results.⁽¹¹⁾

Our findings suggest a link between abnormal dipping status and nicotine dependency beyond smoking per se. According to our results, higher FNDT scores (5 (0-10) versus 2 (0-9), p=0,046), longer smoking years (25 (6-50) versus 20 (5-50), p=0,017), and more tobacco smoking exposure in terms of pack.years (25 (3-135) versus 15 (1-75), p=0,023) were associated with abnormal BP dipping patterns. To our knowledge, there have been no studies evaluating the relationship between nicotine dependency and BP dipping patterns. This result invites further research on this topic. The higher scores on the first question of the FNDT and the higher total FNDT scores or higher smoking exposure should prompt an ABPM of the patient to check for any nocturnal hypertension or abnormal dipping status.

Smokers have a relatively long nicotine-free period during sleeping. Patients reported even awakening during the night due to nicotine cravings in a study.⁽⁴⁷⁾ Nicotine deprivation causes stress and increases the stress-related hormones and neurotransmitters. The noradrenergic system is activated and reinforces the craving symptoms.⁽⁴⁸⁾ These neurohormonal changes due to nicotine deprivation may be a possible mechanism explaining the increase in nocturnal BP values and blunted BP dipping status in our study. One study found that alpha blockage restored the normal dipping pattern.⁽⁴⁹⁾ Another study found that alpha blockage assisted patients in coping with nicotine deprivation symptoms.⁽⁴⁸⁾ Our findings are in line with this evidence, suggesting a link between higher nicotine dependence, ie, higher FNDT scores and higher rates of nocturnal hypertension, non-dipper, and reverse dipper patterns. Other studies reported melatonin deficiency in smoking groups.⁽⁵⁰⁾ Melatonin is an important factor in maintaining normal diurnal rhythm. Melatonin deficiency may be another mechanism behind the abnormal dipping.

We observed that there was a link between smoking and isolated nocturnal hypertension, higher 24-hour systolic, nocturnal systolic, and nocturnal diastolic BP levels in our study population. Isolated nocturnal hypertension and nocturnal hypertension patients were shown to be at increased risk for target organ damage due to hypertension, and their risks for cardiovascular events and mortality were higher.^{(9, 45, 51, 52)}

There is evidence to recommend the treatment of isolated nocturnal hypertension; otherwise, these patients would be at higher risk of cardiovascular morbidity and mortality.^{(53, 54)} Thus, it is of note that clinicians should be aware of the probability that a smoker may present with nocturnal hypertension despite being normotensive in office BP measurements. Due to their increased risk of nocturnal hypertension, patients with a history of smoking may be at higher risk for cardiovascular events and mortalities, even if they show normotensive values in office or daytime home BP measurements. It is unpractical to perform ABPM for general population screening purposes; however, performing ABPM on selected at-risk patients might be of benefit with applicability also. History of smoking may be proposed as a criterion of predictive value for determining such at-risk patients.

We did not find any association between the dipping status of the patients and whether the patients took their antihypertensive medications at night or not. HYGIA and MAPEC trials, among many other studies, have found that chronotherapy or administering antihypertensive medications at night was effective for reestablishing the normal dipping status.^(55-57) These trials also reported significant cardiovascular benefits with night administration of antihypertensive medications.^(55-57) However, TIME study found that there was no significant change in cardiovascular adverse event rates with the ingestion of the antihypertensive medication at night.^{(58, 59)} Additionally HARMONY trial laid proof that there were no significant changes in dipping pattern or cardiovascular events upon nighttime administering of antihypertensive medications.⁽⁶⁰⁾ Our findings are in line with the results of HARMONY trial. Antihypertensive drug amounts in the blood reach a plateau level after a period of regular use, which may explain the indifference in effects.⁽⁶¹⁾

Our findings agreed with the majority of the current scientific research in pointing out the association of a history of smoking with diabetes mellitus^{(62, 63)} and dyslipidemia^{(63, 64)}. Visceral obesity, which is a strong indicator of metabolic syndrome, is increased in smoking patients.⁽⁶²⁾ A meta-analysis found that MetS was increased with smoking.⁽⁶⁵⁾ Our findings demonstrated an increased rate of MetS in the groups with a history of smoking; however, we did not find any significant difference neither in waist circumference nor in neck circumference as continuous variables and as variables with cutoff values. The main association was between history of smoking and non-smokers; however, it was not detected between active smoking and not smoking actively (ie quit and never-smokers together). This suggests a remaining link between smoking and comorbidities even after quit.

Limitations

Despite the fact that many automatic devices that are used for ABPM can give false results if the patient has cardiac arrhythmias, we and many other researchers didn’t exclude those patients. The number of such patients was 6 in our study, so we have deemed this relatively small number as not changing the results. However, it should be reported as a limitation of our study. In addition, there was an insufficient number of extreme dipper patients (n=3), which prevented us from reaching any significant conclusions about this condition. Our classifications rested upon only one ABPM test, but research suggests some limitations regarding the reproducibility of BP dipping patterns on repeat testing at individual level.⁽⁴⁶⁾ Thus, our one-time testing should be noted as another limitation of our study. More experimental and observatory studies with longitudinal follow-up and repeat testing plans are needed. Nevertheless, at the population level, ABPM has been shown to hold a valuable consistency with similar results on the overall ratios.⁽⁴⁶⁾ Therefore, even one-time testing preserves an important significance for the population level. Additionally, there is a need to implement a broader multidisciplinary team approach to avoid positive outcome bias in further studies. Furthermore, ESC guidelines recommend evaluation for secondary hypertension under the age of 40.⁽⁴⁾ Our age cutoff of 30 might have brought about letting in some more secondary hypertension patients than a cutoff of 40 years of age. In addition, we did not perform polysomnography, renal ultrasonography, or endocrinological blood tests such as renin, aldosterone, or cortisol. We relied on the current results and diagnosis of the patients. Absolute exclusion of secondary hypertension patients might have not been achieved. Another limitation of our study was not describing antihypertensives in detail and taking each one as equals, however, evidence suggests there are differences in effect for dipping among anti-hypertensive medications.⁽¹¹⁾

Conclusion

Smoking was associated with more blunted diurnal systolic and diastolic BP dipping levels, in addition to its association with higher absolute values of 24-hour systolic, nocturnal systolic, and diastolic BP. Higher nicotine dependency scores of FNDT were found to be related to abnormal BP dipping, largely due to higher rates of non-dipper and reverse dipper patterns. Patients with a history of smoking were more likely to have diabetes mellitus, dyslipidemia, and metabolic syndrome.

Conflict of Interests

☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Disclosure

The authors report no conflicts of interest in this work.