Unmasking hypertension in children and adolescents with sickle/beta‐thalassemia

Abstract

Sickle cell disease (SCD) is associated with increased risk of cardiovascular disease, although blood pressure (BP) levels have been reported to be lower in SCD patients compared to general population. Aims of the present study were to investigate the prevalence of BP phenotypes and levels of arterial stiffness in pediatric patients with SCD and to assess the differences with children at risk for hypertension. We included in the study 16 pediatric SCD (HbS/β‐thalassemia, S/β‐thal) patients and 16 consecutive children at risk for hypertension referred to our hypertension clinic that served as high‐risk controls. All patients underwent ambulatory BP monitoring and measurement of carotid‐femoral pulse wave velocity (PWV). S/β‐thal patients had lower office systolic BP than the high‐risk control group (115.43 ± 10.03 vs 123.37 ± 11.92, P = .05) but presented similar levels of day and night ambulatory BP. Office hypertension was found in 12.5% of the S/β‐thal patients and in 43.8% of the high‐risk controls (P = .06), while 18.8% of the S/β‐thal patients and 25% of the high‐risk controls presented hypertension by ambulatory BP levels (P = .21). All of the S/β‐thal patients with ambulatory hypertension had night hypertension (one combined night and day hypertension) with office normotension (masked hypertension). S/β‐thal patients and high‐risk controls presented equal prevalence of masked hypertension (18.8%). Children and adolescents with S/β‐thal present similar prevalence of BP phenotypes and levels of PWV with children at risk for hypertension. A significant number of children and adolescents with S/β‐thal may have masked nighttime hypertension despite normal office BP levels.

1. INTRODUCTION

Sickle cell disease (SCD) has been associated with increased risk of cardiovascular events and renal impairment. ¹ , ² , ³ , ⁴ Sickle cell disease trait was reported to have increased risk for kidney disease, but a null association with cardiovascular outcomes including stroke in two recent meta‐analyses in adults. ⁵ , ⁶ Similarly, beta‐thalassemia minor has been reported to have a protective effect against the development of cardiovascular disease. ⁷ However, SCD resulting from the combined heterozygosity for sickle cell (β^S) and β‐thalassemia (β^thal) genes has been associated with increased cardiovascular risk. ⁸ , ⁹

Evidence on the prevalence of hypertension and subclinical vascular damage as assessed by arterial stiffness in SCD populations is relatively few and with inconsistent results. ⁹ , ¹⁰ , ¹¹ , ¹² Some studies showed low blood pressure (BP) levels in SCD patients and similar levels of pulse wave velocity (PWV) with healthy controls, ¹⁰ while others showed increased stiffness and an inverse association of PWV with BP. ⁹ , ¹¹ The interaction of BP and arterial stiffness on the pathogenesis of cardiovascular disease in SCD could improve the understanding of underlying mechanisms and classification of cardiovascular risk in these patients.

The hypothesis of the present study was that despite low office BP levels, ambulatory BP levels may be high in the SCD pediatric population and possibly at similar levels with high‐risk children referred for possible hypertension providing evidence for increased cardiovascular risk in the SCD children. Thus, the aims of the present study were to investigate the prevalence of abnormal BP phenotypes and levels of arterial stiffness in SCD pediatric patients with S/β‐thal genotype in comparison with a control group of children at risk for hypertension.

2. METHODS

2.1. Study population

We included in the study 16 pediatric SCD patients with S/β‐thal genotype and 16 consecutive patients at risk for hypertension. The SCD group included the pediatric cohort of SCD patients with S/β‐thal genotype followed up in the tertiary reference center for hemoglobinopathies in North Greece and were recruited by the caring physicians to participate in the study. All patients in the SCD cohort accepted to participate in the study. All SCD patients were free of disease symptoms or complications at the time of study participation. Patients at risk for hypertension, hereinafter referred to as high‐risk controls, presented with high office BP measurements in the primary care setting and were referred to our outpatient hypertension clinic for confirmation of high BP levels. Exclusion criteria for the participants at the high‐risk control group were secondary cause of hypertension or any chronic disease or use of antihypertensive drugs. All S/β‐thal patients were under treatment with hydroxyurea.

Informed consent to participate in the study was obtained by the children's parents. The human research protocol was conducted according to the Helsinki Declaration for human clinical studies and approved by the institutional review board.

2.2. Anthropometric and laboratory measurements

Body weight and height were measured at the closest 0.1 Kg and 0.1 cm, respectively, with the participant in light clothing without shoes. Body mass index (BMI) was calculated as weight (kg)/height (m²). BMI z score was calculated using Cole's LMS method. ¹³ Laboratory values of cardiovascular risk factors including serum creatinine, glucose, lipid profile, and uric acid were measured after overnight fast. A 24‐hour microalbumin sample was collected for the S/β‐thal patients. Microalbuminuria was defined as excretion of >30 mg and <300 mg a day of albumin in the urine. Estimated glomerular filtration rate (eGFR) was calculated according to the Schwartz formula. ¹⁴

2.3. Office blood pressure measurements

A mercury sphygmomanometer was used for office BP measurements performed at the first day of ambulatory blood pressure monitoring (ABPM). Appropriate cuff size was used, and BP was measured on the right arm to the nearest 1 mm Hg with the child or adolescent quiet, seated with the back supported, and feet‐uncrossed on the floor after a 5‐minutes rest. The mean of these three measurements was used for the analysis. BP groups were stratified according to the European Society of Hypertension (ESH) 2016 staging scheme for children and adolescents. ¹⁵ BP index was calculated by dividing the average office BP by the 95th BP percentile specific for each patient. The reference BP tables by age in 949 patients with SCD from the Silent Cerebral Infract Multicenter Clinical (SIT) trial were also used to assess office BP status. ¹⁶ These tables’ thresholds differ from those of the ESH 2016 as they are not based on healthy population, but on a SCD cohort including patients that may be hypertensive.

2.4. Ambulatory blood pressure monitoring

All participants underwent 24‐hours ABPM on a usual school day. The Spacelabs 90217 ambulatory BP monitor (Spacelabs Inc) was used. The appropriate size cuff was placed around the non‐dominant arm, and three BP determinations were made along with sphygmomanometric measurements to verify that the average of the two sets of values did not differ by more than 5 mm Hg. Reading and analyzing the ABP data have been previously described. ¹⁷ , ¹⁸ All participant were instructed to rest or sleep between midnight and 06:00 (night) and to maintain their usual activities between 08:00 and 22:00 (day). Participant were instructed to keep their arm still and relaxed at the side during measurements. All 24‐hours ABPM sessions had at least 72 valid BP measurements. Mean day and night BP levels were calculated and then transformed to age and height z scores based on previous published normative data. ¹⁹ Daytime BP index was calculated by dividing the average daytime BP by the 95th BP percentile specific for each patient.

Ambulatory BP percentiles according to sex and height were used for the stratification of ambulatory BP in participant <16 years, while adult ambulatory BP limits were used for older participant. ¹⁵ Using the combination of both office BP and ambulatory BP measurements, the participant were divided into four BP phenotypes. Participant who showed normotension or hypertension based on both office and ambulatory BP measurements were characterized as having sustained normotension or hypertension, respectively. White‐coat hypertension (WCH) was defined as office hypertension with ambulatory normotension and masked hyperextension (MH) as office normotension with ambulatory hypertension. ²⁰

2.5. cf‐PWV measurement

The Complior System^® (Colson) was used for the cf‐PWV assessment. The cf‐PWV measurements took place in a quiet, semi‐darkened, temperature‐controlled room (21°C). Before measurement, all participants rested in recumbent position for about 15 minutes. The participants were instructed to refrain from eating, smoking, and drinking beverages containing caffeine for 3 hours before measurement and from drinking alcohol 10 hours before measurement. Participant remained in the supine position and were advised neither to speak nor sleep during measurements. cf‐PWV was calculated according to the equation cf‐PWV = D (m)/t (s), where t denotes the transit time of the arterial pulse along the distance, and D the distance assimilated to the surface between the recording sites. D was measured directly using a centimeter tape, while the Complior device obtained automatically. The same doctor, who was unaware of the BP measurements, conducted all PWV measurements.

2.6. Statistical analysis

The IBM SPSS 24.0 (SPSS Inc) statistical package was used to analyze data. Continuous variables are reported as mean ± SD and categorical variables as counts and percentages. Standard descriptive statistics, t test, or non‐parametric methods were used for the comparison between the groups as appropriate. Simple linear regression models were used to assess associations between BP parameters. A P‐value <.05 was considered as statistically significant.

3. RESULTS

Characteristics of the population are described in Table 1. Despite similar ambulatory BP levels, S/β‐thal patients had lower office systolic BP (115.43 ± 10.03 vs 123.37 ± 11.92, P = .05) than the high‐risk control group. S/β‐thal patients also presented lower total cholesterol and glucose levels. S/β‐thal patients presented similar levels of PWV to high‐risk controls (5.81 ± 1.16 vs 5.67 ± 1.01 m/sec, P = .72). Moreover, 18.8% of the primary hypertensives and 25% of the S/β‐thal patients presented cf‐PWV levels above the 95th percentile for age and sex (P = .66). None of the S/β‐thal patients had microalbuminuria, probably due to the protective effect of hydroxyurea.

TABLE 1.

Demographic, anthropometric, and clinical characteristics of the population

	High‐risk control group (n = 16)	S/β‐thal patients (n = 16)	P
Age (years)	13.41 ± 4.65	13.19 ± 4.76	.896
Male (n/%)	6 (37.5)	5 (31.3)	.710
Height (cm)	159.50 ± 0.16	152.06 ± 0.16	.210
Height z score	1.09 ± 0.93	0.17 ± 1.26	.054
BMI (kg/m2)	25.27 ± 4.44	19.49 ± 3.26	<.001
BMI z score	1.39 ± 1.21	−0.02 ± 1.41	<.05
Office SBP (mm Hg)	123.37 ± 11.92	115.43 ± 10.03	.051
Office SBP z score	1.30 ± 1.23	0.78 ± 1.08	.140
Office SBP index	0.96 ± 0.11	0.90 ± 0.11	.212
Office DBP (mm Hg)	77.12 ± 6.43	71.37 ± 1.59	.077
Office DBP z score	0.54 ± 1.36	0.09 ± 1.101	.419
Office DBP index	0.93 ± 0.08	0.87 ± 0.14	.223
Office HTN (n/%)	7 (43.8)	2 (12.5)	.06
Day SBP (mm Hg)	115.45 ± 6.22	113.70 ± 7.88	.491
Day SBP z score	−0.44 ± 1.16	−0.39 ± 0.83	.892
Day SBP index	0.88 ± 0.06	0.88 ± 0.05	.848
Day DBP (mm Hg)	68.74 ± 5.05	67.69 ± 2.46	.459
Day DBP z score	−0.68 ± 0.95	−0.77 ± 0.36	.739
Day DBP index	0.83 ± 0.06	0.82 ± 0.02	.541
Day PP (mm Hg)	47.70 ± 4.93	46.01 ± 6.34	.731
Night SBP (mm Hg)	105.36 ± 66.99	103.30 ± 9.86	.499
Night SBP z score	0.33 ± 0.86	0.30 ± 1.19	.932
Night SBP index	0.91 ± 0.06	0.90 ± 0.08	.773
Night DBP (mm Hg)	60.05 ± 6.00	55.85 ± 5.67	.051
Night DBP z score	0.68 ± 1.00	0.02 ± 0.79	<.05
Night DBP index	0.91 ± 0.09	0.84 ± 0.07	<.05
Night PP (mm Hg)	45.37 ± 4.84	47.44 ± 6.87	.331
Ambulatory BP HTN (n/%)	4 (25)	3 (18.8)	.210
PWV (m/sec)	5.81 ± 1.16	5.67 ± 1.01	.720
PWV z score	0.90 ± 1.72	0.95 ± 1.38	.922
Cholesterol (mg/dL)	158.00 ± 48.48	120.40 ± 23.03	.219
Triglycerides (mg/dL)	79.00 ± 54.48	68.79 ± 15.90	.700
HDL (mg/dL)	42.20 ± 9.75	38.92 ± 6.07	.513
Glucose (mg/dL)	86.13 ± 9.34	77.00 ± 7.98	.438
Uric acid (mg/dL)	4.65 ± 1.38	3.89 ± 1.065	<.05
eGFR(mL/min/1.73 m2)	134.33 ± 20.26	128.31 ± 18.73	.441
24‐h urine microalbumin (mcg/min)	4.93 ± 4.87	NA

Values represent mean ± SD or n (%).

Abbreviations: BMI, body mass index; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate; HTN, hypertension; NA, non‐available; PP, pulse pressure; PWV, pulse wave velocity; SBP, systolic blood pressure; SCD, sickle cell disease.

Based on office BP measurement at the hypertension clinic, 12.5% of the S/β‐thal patients and 43.8% of the high‐risk control group presented office hypertension (P = .06), while 18.8% of the S/β‐thal patients and 25% of the high‐risk control group presented hypertension by ambulatory BP levels (P = .21). All S/β‐thal patients with office hypertension presented normal ambulatory BP values (white‐coat hypertension). All of the S/β‐thal patients with ambulatory hypertension had night hypertension (one combined night and day hypertension) with office normotension (masked hypertension). S/β‐thal patients and high‐risk controls presented equal prevalence of masked hypertension (18.8%) (Figure 1). The use of the proposed BP tables for SCD patients resulted in identical prevalence of ambulatory hypertension (18.8%) in S/β‐thal patients, all having masked hypertension.

FIGURE 1.

Prevalence of BP phenotypes in the S/β‐thal patients and children at risk for primary hypertension

4. DISCUSSION

The results of the present study show that children and adolescents with S/β‐thal present similar ambulatory BP levels with youth at risk for primary hypertension referred to a hypertension clinic. Moreover, S/β‐thal patients may present with undiagnosed masked hypertension. These findings suggest that S/β‐thal patients may have similar risk for future cardiovascular complications as primary hypertensive children. Finally, office BP in S/β‐thal patients is usually normal, lower than that of youth at risk for primary hypertension, and therefore may not represent their actual cardiovascular risk.

Early reports on BP in SCD populations came at the 1970s showing that SCD patients have lower BP compared to healthy controls. ²¹ , ²² However, further studies showed that BP levels in the upper tertiles of the normal range were associated with adverse outcomes. ⁸ , ²³ In a multicenter clinical trial including 814 children with SCD, aged 5‐15 years, office SBP was linearly associated with the risk of silent cerebral infarct on magnetic resonance imaging (MRI) despite normal office BP levels. ⁸ Of note, the latter study used mean office BP values, not adjusted for age, sex, and height, and the age population‐adjusted levels were not evaluated. The present study demonstrated that children with SCD and normal office BP may have MH. Three previous pediatric studies have shown increased prevalence of MH, ranging from 25% up to 35%, in children with SCD. ²⁴ , ²⁵ , ²⁶ The above finding may explain the higher risk for adverse cerebrovascular ⁸ or renal ²⁷ , ²⁸ outcomes in SCD children with normal office BP level.

In the present study, nighttime hypertension was present in all SCD children with ambulatory hypertension. Our data support previous findings that night BP abnormalities are frequent in SCD pediatric patients and do not associate with office BP. ³ , ²⁴ The above finding also explains the observation of the present study that using office BP reference tables proposed for SCD children did not improve identification of ambulatory hypertension. Another interesting finding of the present study is that all SCD patients with office hypertension presented white‐coat hypertension supporting the inability of office BP levels to appropriately classify BP status in pediatric SCD patients.

Studies evaluating arterial stiffness in pediatric SCD showed similar arterial stiffness indices with healthy children. ²⁹ , ³⁰ The authors suggested that changes in arterial stiffness may became obvious later life, even within low BP levels. ¹¹ The present study does not oppose previous findings but provides evidence that arterial stiffness in SCD pediatric patients is similar with healthy children at risk for hypertension and that some children may have high levels of PWV according to height and sex‐specific normative values.

The present study has some limitations. The study population is small and did not allow assessment of associations between BP parameters and eGFR, urine albumin, and PWV. The cross‐sectional design does not provide evidence of any mechanistic cause. In addition, SCD group refers to patients on hydroxyurea, and its protective effect on multiple complications of the disease is well known. On the other hand, the strong points of the study are the homogeneity of the SCD population (S/β‐thal genotype) and, the firstly, to our knowledge, reported comparison between SCD children and children at increased risk for hypertension. Future studies with larger population samples and longitudinal design need to confirm our findings.

In conclusion, children and adolescents with S/β‐thal present similar prevalence of BP phenotypes and levels of cf‐PWV with children at risk for hypertension. A significant number of children and adolescents with S/β‐thal may have masked nighttime hypertension despite normal office BP levels. ABPM may play an important role in the risk assessment of SCD children and could reclassify patients from a previous office normotensive status.

CONFLICT OF INTEREST

The authors declare no conflict of interest.

AUTHOR CONTRIBUTIONS

SS and ME designed the present work; EP and AT performed the data acquisition, and CA and VK performed ABPM and PWV measurements. SS and CA drafted the manuscript. All authors revised the manuscript critically for important intellectual content.

Stabouli S, Antza C, Papadopoulou E, Teli A, Kotsis V, Economou M. Unmasking hypertension in children and adolescents with sickle/beta‐thalassemia. J Clin Hypertens. 2020;22:1444–1449. 10.1111/jch.13957