Ambulatory blood pressure in 570 Danes aged years: the Odense Schoolchild Study

Transcription

1 Journal of Human Hypertension (1998) 12, Stockton Press. All rights reserved /98 $ ORIGINAL ARTICLE Ambulatory blood pressure in 570 Danes aged years: the Odense Schoolchild Study J Lambrechtsen 1, F Rasmussen 1, HS Hansen 2 and IA Jacobsen 1 Departments of 1 Internal Medicine and 2 Cardiology, Odense University Hospital, Odense, Denmark Ambulatory blood pressure (ABP) measurements were performed in a Danish population of 295 males and 275 females aged years. Individualised day and night periods were defined from the subjects own recording of bedtime and rising on the day of their ABP measurements. During these individualised periods the ABP values for daytime, night-time and for the whole 24-h period were measured. The mean s.d. values for systolic/diastolic ABP for the whole population were /70 7 mm Hg in the daytime, /60 9 mm Hg in the night-time, and /68 7 mmhgin the whole 24-h period. Males had a mean systolic ABP of 9 mm Hg and mean diastolic ABP of 5 mm Hg higher than females. In males mean s.d. systolic/diastolic ABP values in the daytime were /73 7 mm Hg, in the night-time /63 8 mm Hg, and in the whole 24-h period /71 7 mm Hg. The corresponding values in females were /68 6 mm Hg, /57 8 mm Hg, and /66 6 mm Hg, respectively. In conclusion this study provides sex-specific normal values for ABP in a 19 to 21-yearold age group based on individualised daytime and night-time periods. Keywords: ambulatory blood pressure; reference values; young adults Introduction Normal ambulatory blood pressure (ABP) values have been published from a few large true population studies, 1 5 and from studies including selected populations of bank employees 6 or employees from different factories. 7 Results from these studies give good reference frames for children 1 and for large age groups of adults (decades, years, older than 50 years). However, they contain only a few subjects in each decade particularly when sexspecific values are given. It has been shown that ABP both rises with age and that males have higher ABP values than females. 8 Thus, sex-specific reference values for ABP are needed for narrow age groups. Normal values estimated for night-time and daytime periods have to be individualised so that normality is calculated from information about true bedtime and rising in contrast to arbitrarily defined periods. Most published results of daytime and night-time ABP have been from arbitrarily defined periods. Only one large-scale study has used true bedtime information in presenting normal values. 9 Thus, the aims of the present study were to report sex-specific values for ABP in a 19 to 21-year-old age group based on individualised daytime and Correspondence: Dr J Lambrechtsen, Department of Internal Medicine, Odense University Hospital, Odense, Denmark Received 3 April 1998; revised and accepted 27 July 1998 night-time periods and to report the upper limits for normality. Subjects and methods The present results are based on a prospective epidemiological 11-year follow-up study in a population of 1369 children (696 males and 673 females) from municipal schools in Odense, Denmark. The population was first examined in at the age of 8 10 years 10 and the present examination was performed in when the subjects were years old. All measurements were performed under standardised conditions and in the same sequence for every subject. Height to the nearest 1 cm, and weight to the nearest 0.1 kg were measured. A body mass index (BMI) was calculated as weight/height 2. Arm circumference was measured at the midpoint between the acromion and olecranon on the left arm hanging relaxed and straight. Subjects who took medicine or had diseases that could influence blood pressure (BP) were excluded from the BP measurements. Resting clinic blood pressure Resting clinic BP was measured, using a random zero mercury sphygmomanometer (Hawksley and Sons, UK) to reduce observer bias. BP was measured after 10 min of rest in the sitting position, using an appropriate cuff size. A bladder size of cm was used in the cuff if the arm circumference was

2 756 Ambulatory blood pressure in young Danes 32 cm. For circumferences 32 cm, a cuff containing a bladder of cm was used. All BP measurements were performed by one of two physicians. The cuff was deflated at 2 3 mm Hg per second, and the first, fourth, and fifth Korotkoff sounds were recorded. The fifth Korotkoff phase was used as the estimate of the diastolic BP. Two measurements were performed on the left arm followed by one measurement on the right. The mean BP measured on the left arm was used as the resting BP, except for cases in which the difference between mean BP on the left and BP on the right arm was greater than 5 mm Hg. In these cases BP measured on the right arm was used. This was the case in only one subject. Ambulatory blood pressure A 4-week run-in period for familiarisation with ABP measurement was included. QuietTrak from Welch Allyn, USA was used to measure ABP. This equipment meets the validation criteria of the Association for the Advancement of Medical Instrumentation 11 and British Hypertension Society. 12 The QuietTrak monitor works by the auscultatory method with a deflation rate of 2 3 mm Hg per second. With an arm circumference 32 cm a cuff size of cm was used and with a circumference 32 cm a cuff size of cm was used. Calibration of the monitors was checked every 3 months. The cuff was placed around the left arm and the monitor was set to measure BP with 15 min intervals between and and with 30 min intervals between and Either a systolic BP 250 mm Hg or 60 mm Hg or a diastolic BP reading 160 mm Hg or 30 mm Hg were automatically discarded and a new measurement was performed 4 min later. The inflation was dynamic and the cuff was inflated to 30 mm Hg over the preceding systolic BP recording. The accuracy of the equipment was controlled in every subject by a simultaneous measurement of three clinic BP with a sphygmomanometer on the same arm, using a t-connector. The subjects were asked to keep the arm quiet during measurements and to avoid extreme physical stress. All subjects were asked to record bedtime, rising and activity during the day. The night-time and daytime periods were individualised from information about bedtime and rising. The criteria for acceptance of the whole 24-h period were at least 40 accepted BP measurements distributed at random throughout the day and night. To accept the night-time measurements at least five BP measurements, and in the daytime 16 measurements were needed. Ethics The participants gave written consent and the study was approved by the regional scientific ethical committee and the Danish data surveillance authority. Statistics Statistical calculations were made using the statistical packet for Social Science (SPSS). Data are reported as the mean ± s.d. Simple linear regression analyses were used to calculate the regression equation between clinic BP and ABP. Statistical significance was considered at the 5% level. Confidence interval (CI) is given at the 95% level. An independent t-test were used comparing non-participants with participants. Results A total of 900 subjects were included in the present follow-up, and of these 570 (295 females and 275 males) participated in the ABP measurements. Eleven subjects had less than 16 daytime measurements, 50 subjects had less than five night-time measurements. The results of valid measurements at daytime, night-time and 24-h were therefore based on 559, 520 and 550 subjects, respectively. The physical characteristics of the subjects are presented in Table 1. Weight, height, BMI, resting clinic BP, birth weight and disposition to hypertension were compared between participants and nonparticipants. No significant differences were found in females. In males participants in the study had significantly lower weight (75 kg vs 78 kg; P 0.001) and BMI (23 kg/m 2 vs 24 kg/m 2 ; P 0.001). No other significant differences were found when compared with the non-participants. There were usable BPs and measurement errors, which is an 11% error rate. The individual number of ABP measurements for the whole 24 h was 76 ± 12, for the day 59 ± 10, and for the night 19 ± 6. The distribution of the systolic and diastolic ABP had a skewness of 0.52 and 0.56 respectively. Information about true bedtime and rising was missing for 24 subjects. They were included in further analysis given arbitrarily bedtime (24.00) and rising (06.00). A comparison between the ABP values with and without the 24 subjects were performed and no significant differences were observed in either of the reported variables. The true bedtime was calculated to 23h 40 ± 80 min and the true rising to 08h 12 ± 100 min. Table 1 Physical characteristics of the participants (mean (range)) Males Females Age (years) 20.2 (19 21) 20.1 (19 21) Height (cm) 181 ( ) 167 ( ) Weight (kg) 75 (54 145) 64 (44 107) BMI (kg/m 2 ) 23 (17 38) 23 (15 38) Resting systolic BP (mm Hg) 123 (97 158) 115 (81 150) Resting diastolic BP (mm Hg) 71 (41 110) 72 (43 98) Bedtime (hour) ( ) ( ) Time of rising (hour) ( ) ( )

3 Ambulatory blood pressure in young Danes Table 2 Daytime, night-time, and 24-h ABP values for males (M) and females (F). Number of subject in parenthesis (mean ± s.d.) h Daytime Night-time M (264) F (286) M (269) F (290) M (250) F (270) Systolic (mm Hg) 125 ± ± ± ± ± ± 11 Diastolic (mm Hg) 71 ± ± ± ± ± ± 08 HR (beats/min) 74 ± 9 81 ± 9 78 ± 9 85 ± 9 60 ± ± 10 Ambulatory blood pressure When using true bedtime information the mean systolic/diastolic ABP values for the population were 120/68 mm Hg in the 24-h period, 124/70 mm Hg in the daytime period, and 106/60 mm Hg in the night-time period. When daytime and nighttime were defined arbitrarily with night-time defined from midnight to 06.00, and the daytime defined from to only small changes in the mean values were seen when compared with the individualised periods. The arbitrarily mean values were 124/71 mm Hg for daytime and the mean values were 107/60 mm Hg for night-time. The BP was significantly higher and heart rate significantly lower in males than in females during all periods. After adjustment for differences in height and weight, males still had significantly higher BP than females. The sex-specific values of mean, median, and s.d. for heart rates, daytime, night-time and 24-h systolic and diastolic ABPs are presented in Table 2. As shown in Figure 1 for systolic ABP and in Figure 2 for diastolic ABP the hourly mean values for males and females had a diurnal variation. The mean night/daytime ratio for systolic, diastolic and heart rates in males were calculated to 0.86, 0.87 and Figure 2 The hourly mean ( ) and 95% CI (I) for diastolic ABP. The top line has the values for males and bottom line for females The ratios were 0.87, 0.84 and 0.81 respectively for females. Relation between clinic BP and ambulatory BP The correlation between clinic BP and daytime systolic and daytime diastolic ABP is shown in Figure 3 for males and in Figure 4 for females. The mean differences (95% CI) in mm Hg between daytime ABP and clinic BP for the whole population were 4.7 (3.7 to 5.7) for systolic and 1.3 ( 2.2 to 0.4) for diastolic BP respectively. In females differences were 3.0 (1.7 to 4.4) mm Hg in systolic BP and 4.1 ( 5.2 to 2.9) mm Hg in diastolic BP and in males the differences were 6.5 (5.1 to 8.0) mm Hg in systolic BP and 1.6 (0.3 to 2.9) mm Hg in diastolic BP. Figure 1 The hourly mean ( ) and 95% CI (I) for systolic ABP. The top line has the values for males and bottom line for females. Discussion The response rate in a study of ABP measurements of 63% is equal to other studies of normal populations 2,4 and as pointed out by Mancia and co-workers 2 it is a satisfactory result when studying healthy subjects without any benefit of the inconvenient and time-consuming measurement. A study 3 of healthy subjects from different age groups has found that young subjects with a similar distribution as in our

4 Ambulatory blood pressure in young Danes 758 Figure 3 Scatter-plot of relation between ABP and clinic BP for systolic (A) and diastolic (B) in males. The fitted regression line is presented as the middle line. CI 95% for the mean (smaller interval) and the prediction interval (larger interval) are also shown. CI 95% for the coefficient in the regression line is listed in the figure. study, had a higher refusal rate when compared with other age groups. The main reason for not participating in the present study was that subjects did not want to spend time on the ABP measurement. In contrast to other population studies of ABP values, we had the opportunity to investigate differences between participants and non-participants. A selection bias might exist for males since participants had significantly lower weight and BMI than non-participants. We found no significant differences in either clinic systolic BP (122.6 vs mm Hg) or clinic diastolic BP (71.0 vs 72.1 mm Hg) in participating males compared to nonparticipants. No differences were found in the Figure 4 Scatter-plot of relation between ABP and clinic BP for systolic (A) and diastolic (B) in females. The fitted regression line is presented as the middle line. CI 95% for the mean (smaller interval) and the prediction interval (larger interval) are also shown. CI 95% for the coefficient in the regression line is listed in the figure. disposition to hypertension, and we believe that our results can be extrapolated to the background population. Day/night-time values from true or arbitrarily defined periods As also shown earlier 9 we found high agreement between day/night-time values based on arbitrarily defined periods and on individually defined periods for the whole population. It might be reasonable to use arbitrary defined daytime and night-time in a large population if information about true bedtime

5 is not available. It is, however, important to use individually recorded periods when values of ABP from small groups or individuals are concerned. 13 This is due to the large differences between ABP measured during sleep and awake. The majority of population studies comparable to the present have used arbitrarily defined day/night-time periods. 3 6 Ambulatory BP reference values in a population When comparing our results with others, sex-distributions in the populations have to be similar because ABP values in males are higher than in females. 14 In Table 3 the present results are compared with three population studies that used the same inclusion criteria and had nearly the same sex-distribution. The Allied Bank study by O Brien and coworkers 6 included 815 bank employees, a selection that makes it difficult to apply the results to the general population. However, it was one of the first largescale studies of normal ABP values. The Belgian population study done by Staessen et al 4 included randomly selected subjects from a geographically defined area. The Swedish population study by Nystrøm et al 9 randomly selected 100 males and 100 females from the national registry of a community. They used true bedtime information for definition of daytime and night-time. In general there was good agreement among the four studies. The diastolic ABP was in general lower in our study. It is well known that both clinic BP and ABP rise with age 5,6 and the lower BP found could be explained by the fact that our study-population was much younger than in the other studies. Both the auscultatory (QuitTrak) and the oscillometric (SpaceLabs 90202, WA, USA) methods use an algorithm to estimate the BP and differences in the algorithm chosen would give differences in measured BP. If that were the explanation for the differences, a similar trend should be found for nighttime, daytime and the whole 24-h period. The differ- Ambulatory blood pressure in young Danes ences were 0 2 mm Hg for night-time, 0 8 for daytime, and 0 5 mm Hg for 24-h. We therefore believe that the minor disagreements between the studies were due to age differences and true differences between the populations. Gender-specific ambulatory blood pressure reference values The sex-specific difference with higher ABP observed in males is in agreement with other studies. 3,7 The differences in mean ABP were higher for systolic BP (9 mm Hg) than for diastolic BP (5 mm Hg) and they were within the limits of 3 11 mm Hg for SBP and mm Hg for DBP found in a meta-analysis of 23 studies. 8 The study by Wiinberg et al 3 had 30 females and 31 males in the age group years and the study of O Brien et al 6 had 107 females and 174 males in the age group years. When comparing our results separately for both gender with the selected age group from the two studies no major disagreements were found. The differences in systolic BP between the studies for both gender were 0 3 mm Hg at daytime and 24-h and 1 6 mm Hg at night-time. The differences for diastolic were 0 5 at daytime and 24-h and 2 4 mm Hg at night-time Upper limits for normal values There is an ongoing debate of presenting upper limits for ABP either as the 95th percentile or as mean +2 s.d. depending on the distribution of ABP. 15 However in the present study we found that the ABP did not deviate much from the normality distribution with a skewness close to zero. If the mean +2 s.d. is chosen as the upper limit for normal values, the normal limits of the 24-h ABP were 142/83 mm Hg, the daytime 146/84 mm Hg and the night-time 130/78 mm Hg for the overall population. A meta-analysis 8 has proposed the following values for hypertension based on the mean 759 Table 3 Characteristics of our study, the Allied Bank Study 6 and the Belgian 4 and Swedish 9 population study Belgian population Allied Bank study** Swedish population Present study*** study* study*** No Men (%) Age (years) Method oscillometric oscillometric oscillometric auscultatory Systolic BP (mm Hg) Clinic 126 ± ± ± ± h 119 ± ± ± ± 11 Daytime 125 ± ± ± ± 11 Night-time 108 ± ± ± ± 12 Diastolic BP (mm Hg) Clinic 76 ± ± ± 8 72 ± h 71 ± 7 72 ± 8 73 ± 7 68 ± 7 Daytime 76 ± 8 78 ± 8 78 ± 7 70 ± 7 Night-time 62 ± 8 61 ± 8 62 ± 8 60 ± 9 Values are mean ± s.d. (mm Hg) *Daytime defined from to and night-time from midnight to **Daytime defined from to and night-time from to ***Both daytime and night-time were individualised.

6 760 Ambulatory blood pressure in young Danes Table 4 The 95th percentile for systolic and diastolic ABP in males and females (results are in mm Hg) Male Female Day Night 24-h Day Night 24-h Wiinberg et 145/91 120/76 142/89 137/87 109/73 133/85 al 3 O Brien et al 6 144/88 125/70 136/81 131/83 117/68 125/78 Present study 149/85 137/81 145/84 137/80 123/71 132/77 +2 s.d., 139/87 mm Hg for 24-h, 146/91 mm Hg for daytime and 127/79 mm Hg for night-time pressures. Using 95th percentile as the upper limit for normal values, we found normal limits for 24-h to be 141/81 mm Hg, for daytime 145/83 mm Hg and for night-time 131/76 mm Hg. In the Allied Irish Bank Study the 95 percentiles were found to 134/84 mm Hg, 143/91 mm Hg, and 123/75 mm Hg, respectively. Table 4 presents sex-specific upper normal limits as the 95th percentile for the present and two comparable studies, 3,6 from which comparable age groups were selected (20 29 years old from Wiinberg 3 and years old from O Brien 6 ). In general, the present normal values and upper normal limits for both sex-specific and the whole population show higher or equal systolic limits and lower diastolic limits than the studies we compared to. The reason for this is likely to be differences in age and sex-distribution with younger subjects in our population. The ABP populations studies done have shown that gender distribution is more decisive than age distribution for the results of ABP. The review by Nielsen PE et al 14 showed that systolic daytime ABP only increased approximately 2 mm Hg from 20 to 59 years and diastolic daytime ABP approximately 7 mm Hg from years. The PAMELA study 2 measured ABP in 1438 subjects older than 25 and found almost the same systolic limits as we did but higher diastolic limits. The arbitrarily defined daytime and night-time periods used in other studies compared with the individualised periods used in our study could explain some of the differences in daytime and night-time values. In addition young individuals may be more physically active and deviate more in sleeping habits, than older ones. Systolic BP is influenced by physical activity and this may be a reason for the higher limits for systolic BP in the present study. In conclusion, our study provides information about true ABP values in a large population of young adults in a narrow age group. The values found and the normal limits calculated were in general in good agreement with the present literature. We found in average mean systolic ABP 9 mm Hg higher and mean diastolic ABP 5 mm Hg higher in males than in females. Thus before evaluating normal ABP in a population it is necessary to evaluate the results according to male female ratios, because of the large differences in ABP between the sexes. Acknowledgements We owe a debt of gratitude to Niels Hyldebrandt who encouraged us to start this study. The study was supported by The Danish Heart Foundation and The Danish Medical Research Council to whom we express gratitude. References 1 Soergel M et al. Oscillometric twenty-four-hour ambulatory blood pressure values in healthy children and adolescents: a multicenter trial including 1141 subjects [see comments]. J Pediatr 1997; 130: Mancia G et al. Ambulatory blood pressure normality: results from the PAMELA study. J Hypertens 1995; 13: Wiinberg N et al. 24-h ambulatory blood pressure in 352 normal Danish subjects, related to age and gender. Am J Hypertens 1995; 8: Staessen JA et al. Ambulatory blood pressure and blood pressure measured at home: progress report on a population study. J Cardiovasc Pharmacol 1994; 23 (Suppl 5): Imai Y et al. Ambulatory blood pressure of adults in Ohasama, Japan. Hypertension 1993; 22: O Brien E et al. Twenty-four-hour ambulatory blood pressure in men and women aged 17 to 80 years: the Allied Irish Bank Study. J Hypertens 1991; 9: Battistella P et al. [Reference values of ambulatory arterial pressure in activity and during the night. Multicenter study of 394 normotensive subjects at rest]. Valeurs de reference de la tension arterielle ambulatoire d activite et de nuit. Etude multicentrique de 394 sujets normotendus au repos. Arch Mal Coeur Vaiss 1989; 82: Staessen JA et al. Mean and range of the ambulatory pressure in normotensive subjects from a meta-analysis of 23 studies. Am J Cardiol 1991; 67: Nystrom F, Malmstrom O, Karlberg BE, Ohman KP. Twenty-four hour ambulatory blood pressure in the population. J Intern Med 1996; 240: Hansen HS, Hyldebrandt N, Nielsen JR, Froberg K. Blood pressure distribution in a school-age population aged 8 10 years: the Odense Schoolchild Study. J Hypertens 1990; 8: White WB et al. Multicenter assessment of the Quiet- Trak ambulatory blood pressure recorder according to the 1992 AAMI guidelines. Am J Hypertens 1994; 7: O Brien E et al. The British Hypertension Society protocol for the evaluation of automated and semi-automated blood pressure measuring devices with special reference to ambulatory systems [see comments]. J Hypertens 1990; 8: Gosse P, Ansoborlo P, Lemetayer P, Clementy J. Daytime and nighttime ambulatory blood pressures should be calculated over the true sleep/waking cycle and not over arbitrary periods. Am J Hypertens 1996; 9: Nielsen PE, Christensen HR, Hilden T. Ambulatory daytime blood pressure in relation to age and gender compared to previous population studies based on office blood pressure. The Danish Study Group on Ambulatory Blood Pressure. J Hum Hypertens 1996; 10 (Suppl 3): Staessen JA et al. A consistent reference frame for ambulatory blood pressure monitoring is found in different populations. J Hum Hypertens 1994; 8: