Portal de Periódicos da CAPES

Seasonal variations in blood pressure

2012; Lippincott Williams & Wilkins; Volume: 30; Issue: 7 Linguagem: Inglês

10.1097/hjh.0b013e328355d7f9

1473-5598

Cesare Cuspidi, Juan Eugenio Ochoa, Gianfranco Parati,

Birth, Development, and Health

Seasonal variations in cardiovascular morbidity and mortality show a winter peak and summer nadir, as reported since decades across different latitudes, ethnic groups and age strata. Mortality excess in winter months is mostly related to cardiovascular events including acute myocardial infarction, sudden death, stroke and pulmonary thromboembolism [1–3]. In a landmark study based on 300 000 cardiovascular deaths from the Canadian Mortality Database, Sheth et al.[4] observed a 19 and 20% increase in mortality, respectively, from acute myocardial infarction and stroke, in January compared with September. In a community-based study carried out in Minnesota, USA from 1979 to 2002, a 17% increase in sudden cardiac death was found in winter compared with summer [5]. The association between cold weather and sudden death was similar over years, across age and sex groups and was stronger for individuals without a previous history of coronary heart disease. Available data are mostly derived from countries exposed to cold/temperate climates characterized by large variations in seasonal temperature. Similar findings, however, have been also reported by studies performed in countries with smaller variations in environmental temperature. In a population of almost 1 million persons in Israel, mortality from ischemic heart disease and stroke in men was, respectively, 51 and 48% higher in mid-winter than in mid-summer; the corresponding figures in women were 48 and 40%, respectively [6]. Seasonal variations have been also reported for nontraumatic rupture of thoracic and abdominal aortic aneurysms [7]. Exposure to winter weather conditions has been hypothesized to induce physiological and clinical changes including sympathetic activation, hemoconcentration, hypercoagulability, increase in plasma lipids and in infections rate that, overall, may increase the incidence of cardiovascular diseases [8,9]. As a direct effect of cold weather on the cardiovascular system, reflex coronary and systemic vasoconstriction may occur. Platelet activation may take place during viral infections; increased platelet stickiness and thrombus formation may provide a rationale for the association between ischemic heart disease, stroke and influenza. Short-term and long-term effects of cold exposure on blood pressure (BP) have also been consistently demonstrated in both normotensive and hypertensive individuals. Thus, weather-related increments in BP are seen as major determinants of cardiovascular morbidity excess in the cold season. These seasonal BP fluctuations represent one of the long term components of the variations that characterize BP behavior in daily life. BP variability is indeed a dynamic and complex phenomenon including short-term and long-term fluctuations as a result of intricate interactions between behavioral, humoral and neural central or reflex influences. Although often overlooked, also environmental factors may importantly contribute to BP variations, as they may potentially influence physiological mechanisms involved in BP regulation. Particularly, seasonal climatic changes have been reported to be associated with BP variations in humans, SBP and DBP levels presenting significant increases during winter months [10,11]. Seasonal influences on BP were first described in the early 1960s by Rose [10], who analyzed BP measurements in 56 middle-aged men affected by ischemic heart disease during a 1–3 years follow-up period and found a clear seasonal trend, with a peak in spring and a trough in late summer. Additional evidence on this phenomenon has been provided by large population studies including both normotensive and hypertensive individuals, showing that seasonal BP changes are not only limited to conventional BP measurements performed in the clinic but also affect out-of-office BP levels. Either when considering clinic BP values, the average of self-BP measurements performed by individuals at home or the mean of the 24-h BP values collected by ambulatory BP monitoring (ABPM), BP levels have been reported to be lower during summer and higher during winter [12]. In the last decades, studies performed in different settings (i.e. clinical trials, general population, hypertension clinics, transplant recipients) have also shown that BP undergoes seasonal changes, and that higher values of either clinic BP, home BP or ABP are generally recorded in winter months [13–17]. A series of studies, investigating specific climatic conditions responsible for the seasonal variation in ABP levels, have found fluctuations in outdoor air temperature to be a major independent determinant for this phenomenon [11,18,19]. The Medical Research Council treatment trial including more than 17 000 individuals with mild hypertension from 190 centers throughout England, Scotland and Wales reported that pressor effects of low outdoor temperatures were greater in older than in younger individuals [11]. Men aged 55–64 years had office SBP by 6–7 mmHg and DBP by 3–4 mmHg higher on cold winter days than on warm summer days; in men aged 35–44 years, the comparable figures were 2–4 mmHg SBP and 2–3 mmHg DBP. Similar effects of temperature on BP were observed in women. The first large-scale population-based study focusing on seasonal variations in BP measured in and out of the office conditions was published by Sega et al.[12] in the late 1990s. Office SBPs and DBPs were lower in summer and higher in winter, whereas intermediate BP values were observed in spring and autumn both in normotensive and in hypertensive individuals. Home BP values showed a similar pattern and this was also the case for 24-h average, daytime and nighttime BPs. Similar climate-related changes in office and mean 24-h BP were found by Modesti et al.[18] in 6404 individuals referred to two hypertension clinics in Italy. Office and mean 24-h SBP, as well as morning BP surge were significantly lower in hot days (136 ± 19, 130 ± 14 and 33 ± 16 mmHg, respectively), and higher in cold days (141 ± 12, 133 ± 11 and 37 ± 9 mmHg) when compared with intermediate temperature days (138 ± 18, 132 ± 14 and 35 ± 15 mmHg). A higher prewaking morning surge (21 ± 1 versus 15 ± 1 mmHg) and a greater nocturnal BP decline (16 ± 1 versus 14 ± 1 mmHg) in coldest compared with warmest days was observed by Murakami et al.[20] who examined short-term BP variations in 158 individuals undergoing ABPM for 7 consecutive days. Finally, a large amount of information on the link between ambient temperature and BP has been recently obtained from 21 894 individuals, aged 45 years and older, participating in the Reasons for Geographic And Racial Differences in Stroke (REGARDS) study [21]. Within this large biracial population-based sample, both SBP and DBP were significantly related with seasons: mean SBP and DBP was approximately 2 and 1 mmHg higher, respectively, in winter than in summer. Decreasing environmental temperatures may affect BP regulation through several mechanisms: first, alterations in skin vasomotor tone resulting in a marked increase in vascular peripheral resistances; and, second, activation of the sympathetic nervous system accompanied by secretion of catecholamines as well as substances involved in heat production is considered to play a central role, as it produces arteriolar constriction and a subsequent increase in peripheral vascular resistance. In due course this may decrease sweating and, therefore, salt loss, increasing the load of sodium on the kidneys, thus further contributing to the increase in BP. Evidence that an increased sympathetic activation may contribute to the rise in BP during winter time has been provided by several studies both in normotensive and essential hypertensive individuals, showing an increased norepinephrine and epinephrine concentrations in plasma and urine accompanying the BP increase during the cold season [22,23]. Third, marked seasonal variations in loss of fluids and electrolytes through sweating and breathing accompanying seasonal BP changes have also been observed by some studies in which 24-h urinary volume was shown to be significantly higher and inversely correlated with mean ambient temperature during the winter season [22]. Fourth, cold temperature may also increase erythrocyte deformability and blood viscosity, a major determinant of systemic vascular resistances. Fifth and finally, a reduced intensity in ultraviolet light during winter has been shown to reduce the epidermal photosynthesis of vitamin D3 and parathyroid hormone, which was shown in turn to be associated with elevated BP levels [24]. As vitamin D is a negative regulator of the renin–angiotensin system, decrements of its blood levels in the winter period may lead to BP increments via an activation of the renin–angiotensin system [25]. In addition, data obtained in a retrospective analysis of 424 732 pregnant women from year 2011 through year 2005 in Australia support the view that a higher sunlight intensity before delivery is associated with a decreased incidence of pregnancy-related hypertension [26]. Several studies investigating the potential determinants of seasonal BP variations have indicated that the magnitude of seasonal BP rise from summer to winter may be significantly increased by the presence of a number of interfering factors such advancing age [11,18,19], increasing body height [14], decreasing body weight [27], cigarette smoking [28], reduction in physical activity and in vegetable dietary intake. Conversely, the availability of indoor temperature control at home or in the working place (which generates a higher gradient between outdoor and indoor temperature while reducing the seasonal changes in living ambient temperature) has been shown to significantly attenuate the BP increase from summer to winter [13] (Fig. 1). It has also been suggested that in treated hypertensive patients, a downward titration of antihypertensive drug regimen (which is often performed on the basis of clinical variations of clinic BP) may contribute to the nighttime increase in BP levels reported during summer in some studies. Evidence supporting this latter concept was provided by a study by Modesti et al.[18], exploring the combined effects of age and antihypertensive treatment in temperature-induced BP variation, in more than 6000 individuals referred to hypertension clinics in two large Italian cities. Apart from confirming the previously reported inverse relationship between 24 h and daytime ambulatory SBP and outdoor temperature, this study reported significantly higher nighttime SBP levels during hot days in the subgroup of elderly treated hypertensive individuals in whom the number of antihypertensive drugs assumed per day was significantly lower in hot than in cold days (1.71 ± 0.86 versus 2.30 ± 1.31 mmHg, P < 0.001).FIGURE 1: Complex relation between ambient temperature and blood pressure (BP) levels and its known determinants. *Changes in systolic and diastolic office BP, home BP and ambulatory BP levels. **Significant increases in nighttime BP with increasing outdoor temperatures during summer have been reported in treated hypertensive patients. CV, cardiovascular; T, temperature.The study by Lewington et al.[29] published in the current issue of the Journal of Hypertension provides further evidence on the occurrence of seasonal BP variations as well as on their determinants in the frame of a large-scale study in an Asian population from China [30,31] where such phenomenon was never explored before. The main finding of this study is the demonstration of a significant mean increase in SBP/DBP by 10/4 mmHg between summer (June to August) and winter (December to February), and of an inverse relationship of outdoor temperature with BP for outdoor air temperatures above 5°C (with a mean SBP/DBP rise of 5.7/2.0 mmHg for each 10°C decrease in outdoor temperature). Such seasonal BP changes appear to be more pronounced than those previously reported in other large population studies [32,33]. The study by Lewington et al. has several merits. First, it is the largest study to date assessing seasonal BP variations on a population basis (including data from more than 500 000 men and women aged 30–79 years) across 10 different geographical areas of China with a wide range of climatic conditions. Second, at variance from most previous studies that have separately assessed either the relationship between BP and seasonal changes or that between BP and air temperature fluctuations, the study by Lewington et al. has provided a comprehensive assessment of the complex interrelationships among all these factors, that is seasonal changes, outdoor temperature changes and BP variations. Moreover, Lewington et al. not only have accounted for previously known modifiers of this phenomenon (i.e. age, adiposity, presence of antihypertensive treatment, alcohol intake and smoking status) but they have also assessed whether certain demographic factors (i.e. living in rural versus urban areas, or in low-income versus middle-income areas) and other climatic variables such as humidity and indoor climate conditions (i.e. availability or not of central heating) might modify the magnitude of the effect of outdoor temperature on seasonal BP variations. Indeed, this study has for the first time provided consistent evidence over a very large scale that the effects of outdoor temperature on seasonal BP variation may be substantially attenuated (and even abolished) by availability of indoor temperature control, which may prevent BP from rising during winter months. This factor, according to the authors, might be a likely explanation for the reduced increase in BP during winter months observed in individuals living in urban and middle-income areas, where home heating is more frequently available as compared with rural areas. It is worth emphasizing that, in line with previous reports but for the first time in such an extremely large population, the study by Lewington et al. has shown that the magnitude of BP rise from summer to winter may be significantly increased by advancing age (from 7.8 mmHg at age 30–39 years to 11.2 mmHg at 60–69 years), smoking status and decreasing BMI (from 9.3 mmHg with a BMI of 30–39 kg/m2 to 12.2 mmHg with a BMI of 15–18.4 kg/m2). For this latter factor, and at variance from previous studies conducted in selected groups, the study has the merit to show for the first time on a population basis that the seasonal variation in BP is significantly wider in lean subjects. Finally, the changes in BP associated with season or outdoor temperature were not modified by outdoor relative humidity or by drinking status. Despite the consistent weight of the evidence provided by this study, a major limitation is the method employed for BP measurement that consisted on the average of two BP measurements performed in a clinic setting only. Given the highly dynamic behavior of BP over time, isolated clinic BP measurements have been repeatedly shown to be unable to provide a reliable assessment of an individual's prevailing BP level in a given condition, including seasons. This limitation is particularly important if we consider that most centers participating in the study by Lewington et al. were provided with electric heaters to keep rooms for patients’ evaluation warm during winter, so that the spot clinic BP measurements considered for the study could have been importantly influenced by indoor temperature control in the clinic, thus failing to reflect the BP effects of daily climate conditions in subjects’ home. Although it is understandable, given the size of the population investigated, why ABPM could not be implemented in the study by Lewington et al., use of home BPM might have been usefully implemented at least in selected subgroups, thus providing a more reliable assessment of the actual seasonal BP changes in the daily conditions of individuals living in different geographical and social environments. Moreover, the unavailability of 24-h ABPM even in selected subgroups of the populations investigated did not allow to separately assess the effects of seasonal changes on daytime and nighttime ABP, which were reported in other studies to show differential changes with warmer weather [18,34,20]. In this issue of the Journal of Hypertension, Fedecostante et al.[35] offer an additional contribution to our understanding of the association between environmental temperature and ABP levels in a large sample of individuals referred to a single outpatient hypertension center. In this cross-sectional survey, the authors compared average 24-h, daytime and nighttime BP values and nocturnal BP decline observed in hypertensive individuals undergoing 24-h ABPM in the winter period (January to February) (n = 867, 52% men, mean age 56 years) with the corresponding parameters obtained in hypertensive individuals undergoing ABPM in the summer period (July to August) (n = 528, 55% men, mean age 56 years). In the presence of a mean difference in outdoor temperature between the coldest and warmest months of approximately 18.5°C, daytime SBP and DBP were higher in winter than in summer (mean differences 2.4 and 2.2 mmHg, respectively, P = 0.001 for both), whereas an opposite trend was seen for nighttime SBP and DBP values that were higher during the summer period (2.3 mmHg, P = 0.005; 1.0 mmHg, P = NS, respectively). As a consequence, average 24-h SBPs and DBPs were marginally higher in winter months (1.0 mmHg, P = NS; 1.4 mmHg, P = 0.01, respectively). When patients were categorized as being characterized by a nondipping pattern, defined as a nighttime reduction in SBP lower than 10% compared to daytime values, and by isolated nocturnal hypertension (nighttime BP >120 mmHg SBP or >70 mmHg DBP), these BP phenotypes were more frequently found in the summer than in the winter period (62 versus 42%, and 15 versus 10%, respectively, P = 0.001 for both). Analyses of subgroups, according to sex, age and hypertension control confirmed the general trend of higher summer nighttime BP values, although statistical significance was lost in men, in younger people and in controlled hypertensive individuals. It is worth of note that BMI and smoking were not taken into account in these subanalyses, despite previous reports showing that seasonal changes in ABP are consistently related to such variables. Kristal-Boneh et al.[27] demonstrated that the winter increase in office and ambulatory SBP was highest among individuals in the lowest BMI quartile and lowest among those in the highest BMI quartile. Furthermore, it has been suggested that smoking has similar effects on BP, as cold-exposed smokers may have higher BP levels than nonsmokers. A study performed in 97 healthy men (73 nonsmokers and 24 smokers) showed that both smokers and nonsmokers had higher systolic and diastolic daytime BP in the winter than in the summer period; the BP increase during winter in smokers, however, was significantly higher for mean daytime SBP (7.3 versus 2.7 mmHg, P < 0.01) and for mean daytime DBP (4.4 versus 3.1 mmHg, P = 0.05) [28] Some further aspects and limitations of the study by Fedecostante et al.[35] deserve to be commented. First, summer–winter variations in average 24-h SBP were lower than those reported in two previous large-scale Italian surveys mentioned above, that is the studies by Sega et al. (4 mmHg) [12] and Modesti et al. (3 mmHg) [18]. This discrepancy could be ascribed to differences in winter–summer temperature excursion as well as in demographic and clinical characteristics of the subjects examined in the three studies. For instance, BMI in the Fedecostante et al.[35] study was higher than in the above mentioned studies; on the contrary, mean age and the prevalence of elderly were higher in the study by Modesti et al.[18]. A body of evidence indeed supports the view that seasonal variations in BP are related to aging. In agreement with the study by Lewington et al. [29], also published in this issue of the journal ( see above), the Three-City study, a population-based longitudinal study including 8801 individuals aged 65 years or older showed that BP variations with temperature were substantially greater in older persons (80 years and older) than in younger individuals [19]. Second, the study design was cross-sectional with different individuals examined in different months over a 10-year period (from 2002 to 2011), raising the possibility that observed BP variations were, at least in part, the result of chance. Third, factors and mechanisms implicated in the increase of nighttime BP and in the higher frequency of nondipping pattern during the summer months remain unclear. These findings are in keeping with observations made by Modesti et al.[18] in the elderly fraction of their population but are in contrast with the results in the PAMELA (Pressione Arteriose Monitorate E Loro Associazioni) population [12]. Extent and intensity of daytime physical exercise, duration and quality of sleep as well as the absolute indoor temperature and the gradient between outdoor an indoor temperature may all play a role in determining the magnitude of day–night BP variations. Unfortunately, the retrospective nature of the study by Fedecostante et al.[35] prevented a satisfactory assessment of these relevant aspects. This is the case also regarding the possible role of seasonal variations in antihypertensive therapy, characterized by a trend toward a less intensive treatment in summer, which in the study by Modesti et al.[18], was associated with higher nocturnal ABP values observed in warmer days. Studies on the clinical relevance of BP variability have so far mostly focused either on short-term fluctuations (i.e. BP fluctuations within the 24-h period) or on longer term changes between days or visits that may have prognostic implications [36–38]. Also, seasonal changes in BP might have prognostic value, however. Indeed, weather-related BP increments in winter might be a plausible mechanism contributing to the cardiovascular morbidity and mortality excess observed during the cold season as highlighted in the fist part of this editorial [39,40]. Moreover, the article by Fedecostante et al.[35] has clearly shown that higher outdoor temperature is associated not only with lower daytime BP values but also with reduced nocturnal BP decline, resulting in higher prevalence of BP phenotypes such as isolated nocturnal hypertension and nondipping pattern, both known to have adverse prognostic implications (Fig. 1). However, no prospective studies have been conducted specifically addressing these issues so far. In fact, rather than prospectively assessing the BP changes between seasons in the same individuals, different individuals throughout different months of the year have been considered for the available mostly cross-sectional or retrospective analyses, thus preventing a proper assessment of the prognostic relevance of seasonal BP variations, an issue that still needs to be adequately assessed. It has been suggested by several studies that seasonal BP changes should be taken into account also during the diagnostic and therapeutic approach to hypertension by several reasons. Indeed, the season of the year when BP levels are measured may affect the identification of the presence of hypertension, a lower frequency of hypertension diagnosis being likely to characterize summer than winter months. Seasonal factors might also influence the decision about starting antihypertensive treatment, particularly in people with high-normal BP levels, and, in treated hypertensive individuals, they might influence the assessment of the degree of BP control achieved over 24 h. As in treated hypertensive individuals, higher doses or higher numbers of drugs may be required in winter to achieve the same BP control as at other times of the year, seasonal BP variations should be considered for a proper titration of antihypertensive medication, according to the so-called ‘season-tailored chronotherapeutic approach’ aimed at obtaining adequate BP control during the daytime as well as during the night hours in all seasons. On the basis of these considerations, properly designed prospective studies would need to be performed to adequately explore the prognostic relevance of seasonal BP variations, as well as to understand to which extent they might influence the diagnostic and therapeutic management of hypertension. ACKNOWLEDGEMENTS Conflicts of interest The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the article.