Blood Pressure in Children with Congenital Adrenal Hyperplasia due to 21-Hydroxylase Deficiency Free

Congenital adrenal hyperplasia (CAH) due to 21-hydroxylase deficiency (21OHD) is the commonest inherited disorder of steroid hormone biosynthesis which usually presents with adrenal dysfunction with or without associated atypical genitalia due to androgen excess [1]. The phenotype and the clinical features vary widely depending on the underlying genotype and the severity of the enzyme defect [1, 2]. There is a clinical spectrum of severity ranging from classic CAH in approximately 1:16,000 live births, to the milder non-classic form with an incidence of about 1:1,000. The non-classic form commonly presents with androgens excess but sufficient mineralocorticoid (MC) production to avoid salt loss and is usually diagnosed after infancy [3].

The management requires adequate glucocorticoid (GC) replacement in a dose sufficient to reduce the ACTH drive leading to excess adrenal androgens, while avoiding overtreatment with GCs and associated side effects. MC replacement is required to treat clinically relevant salt loss. Management is challenging due to the difficulty in mimicking the diurnal variation of hormones, the variability of receptor maturation with age, variable GC absorption and clearance, as well as the challenges around patient adherence, which is hard to monitor but can markedly effect the success of treatment. To meet therapy targets, most patients require supraphysiological doses of GCs [4].

The prevalence of hypertension in otherwise healthy children is 2–5% (95% CI: 3.3–4.8%) with the peak at the age of 14 years. In adults, hypertension contributes to 7.6 million deaths annually with global prevalence rising to 75–79% in the last decade [5‒8]. Childhood hypertension has been identified as a well-known risk factor for poor cardiovascular health in later life [9]. In CAH, the inadequate replication of the physiological circadian cortisol secretion results in transient episodes of hyper- and hypocortisolaemia, which may lead to metabolic problems such as insulin resistance, obesity, hypertension, and increased cardiovascular risk, usually diagnosed when the patient reaches adulthood. Increased blood pressure (BP) does increase the cardiovascular risk independently contributing to the CAH-related mortality and morbidity [10]. To date, there is not a comprehensive summary of the literature on the incidence and key aspects of increased BP in children with CAH [6]. Here, we review the data published on the variation in BP in children with CAH due to 21OHD (Table 1), and its correlation with disease and treatment factors.

Hypertension in childhood carries a significant cardiovascular risk in later life [29‒31]. The estimated prevalence of hypertension in 21OHD is higher than the 2–5% reported in the general paediatric population [6, 7]. However, the exact prevalence in children with CAH varies among studies, from as high as 20–66% [14, 18, 27, 28, 32, 33] to some reporting normal BP [25, 34, 35]. GC and MC treatment, salt supplementation, androgen excess, and the disease severity are possible predisposing factors for hypertension and cardiovascular morbidity [36].

BP is influenced by various mechanisms and lifestyle factors including anxiety, activity level, and emotional factors. Consequently, serial ambulatory BP measurements are a more reliable indicator of cardiometabolic risk compared to single-point measurements. Different practices have been employed for monitoring BP in children with CAH throughout the literature with 24-h BP profiles [10, 13, 16, 25‒28, 34] being most widely used. However, some studies defined hypertension by point measurements above the 95th centile for age and sex [11, 15, 37], which may partly account for the wide range in the reported prevalence.

Cardiovascular risk in CAH is not precisely understood. Ambulatory BP and heart rate monitoring showed that 20% of adults with CAH had a higher heart rate in relation to the reference population [38]. Left ventricular dysfunction (LVD), endothelial dysfunction, and intimal thickening are known cardiovascular changes associated with CAH [20, 23, 35, 39, 40]. Studies conducted on adult cohorts suggested that such vascular changes in patients with CAH are driven by the early elevation of systolic blood pressures (SBPs) [41]. High carotid intima media thickness (CIMT) and increased high-sensitivity C-reactive protein have been shown in research into atherosclerosis to serve as early predictive factors for cardiovascular events [20, 42]. However, their usefulness in routine clinical practice is questionable.

Studies have described significantly higher CIMT in hypertensive patients with CAH compared to a normotensive reference group (p = 0.013), and in particular, in the case of patients who had nocturnal hypertension (p = 0.02). CIMT was inversely associated with the nocturnal dip in BP (p = 0.037, r = −0.632) suggesting it may play a role in the mechanism behind the unfavourable cardiovascular profile in CAH [16, 17, 22]. There was no difference in CIMT between salt wasting and simple virilizing subtypes (p > 0.05) [20, 22]. There is no evidence to support the relationship between CIMT and biochemical markers of disease control [10]. However, one study showed abdominal aorta intima media thickness to be negatively correlated with androstenedione (A4) concentrations (r = −0.6, p = 0.008) [22].

Incomplete right bundle branch block on electrocardiograms, significantly thinner left ventricular posterior wall thickness during diastole, and shorter isovolumetric relaxation time on echocardiogram have been reported in adults with CAH. It can be hypothesized that persistent long-term hypertension may lead to some of these changes, but they did not directly correlate with 24-h BP profiles [43]. Reduced global longitudinal strain associated with subclinical LVD was found in CAH patients with hypertension, which may be suggestive of a higher risk of LVD in later life [43]. It was also found in a small prospective case-control study that neonates with CAH (n = 9) can have transiently altered cardiac contractility, although this did not correlate with long-term cardiovascular morbidity or hypertension [44].

High BP is more commonly reported in younger children with CAH, with the prevalence being influenced by sex and CAH subtype, while in older children and adolescents, the incidence of hypertension appears to be more variable [11, 19, 32]. Studies have reported that the majority of children with CAH (57–91%) tend to develop hypertension before the age of five, coinciding with the time when GC and MC replacement doses are highest [11, 14, 19, 45]. However, a pilot study which included 189 BP measurements in 24 children with CAH during the first year of life found no difference in BP when compared to reference values [37]. In a follow-up study including 33 children diagnosed via newborn screening, SBP and diastolic BP (DBP) were high at the ages of 12, 18, and 24 months (p < 0.01), peaking at 18 months when more than half of patients were hypertensive (57% and 75% for SBP and DBP, respectively). BP levels decreased with increasing age reaching a nadir towards childhood (9% at the end of the study period of 4 years). However, nearly half of the patients remained hypertensive at 2.5 years of age (n = 42) [11]. Similarly, a chart review study including 180 patients showed that most of the children after the initial diagnosis of hypertension either fluctuated between normal and high BP values (65%) or became normotensive as they grew up (35%) [14]. In another retrospective chart review study, five of 91 children remained hypertensive despite attempts to optimize treatment factors and required antihypertensive medication for between 9 months and 7.1 years duration, of which angiotensin-converting enzyme inhibitors were the most commonly employed medication. Hydralazine, calcium channel blockers, and hydrochlorothiazide were the other antihypertensives used. Of note, angiotensin-converting enzyme inhibitors can lead to a reactive rise in renin, and thus, their use reduces the reliability of plasma renin activity (PRA) as a marker of MC therapy in these children [45]. Three out of 33 CAH children remained hypertensive beyond the age of four in a follow-up study and one required antihypertensive therapy [19]. Despite the discrepancies in the published evidence, the overall evidence shows that hypertension is common in early childhood in CAH and few patients remain hypertensive throughout life.

Only a few studies have examined the differences in BP between CAH subtypes. A large longitudinal cohort study including 180 children with classic CAH (120 SW and 60 SV) followed up between 1970 and 2013 showed that the age of the onset of hypertension was significantly related to the severity of CAH (SW vs. SV), where the most patients with SW-CAH tend to develop high BP before the age of five (p = 0.0004) [14]. Yet no difference in the incidence of hypertension between the SW and SV subgroups was found during adolescence (24 SW vs. 8 SV) [20].

Males may have a tendency of developing hypertension earlier during the first 5 years of life, while females have less marked deviation from normal throughout childhood up to the age of 18 years [14]. Direct comparison has shown no difference in BP between sexes in children with CAH (n = 38, 15 males) [28]. Absence of a gender dimorphism in hypertension and cardiovascular parameters might be related to the absence of a significant difference in androgen concentrations between males and females or other treatment-related factors such as exposure to GCs or MCs. The same can be considered as an additional risk factor for hypertension and cardiovascular morbidity in females in the long run. The protective effect of oestrogens secreted during minipuberty against metabolic adverse outcomes was considered as a possible explanation for the difference in the onset of hypertension between sexes. Studies in adults with CAH further support this hypothesis by demonstrating increased prevalence of hypertension in females with CAH compared to a reference population [32, 46].

Abnormalities in the GC receptor gene or its metabolism are thought to be the possible contributors for persistent hypertension [24, 45]. Carriers of the BclI GC receptor polymorphism have a higher risk of systolic hypertension and unfavourable metabolic syndrome indicators compared to the wild-type controls [24].

In 21OHD, GC replacement aims to correct cortisol deficiency, to reduce the ACTH drive to thereby normalising the production of excess adrenal androgens. The recommended GC replacement in children is hydrocortisone (HC) in a daily dose of 10–15 mg/m²/day. This is higher than the physiological cortisol secretion of 6–8 mg/m²/day for children and 7–9 mg/m²/day in neonates [47, 48]. The usual practice is to divide the daily dose into three or four doses per day according to a circadian regime (with the largest dose in the early morning). GC-induced hypertension is an extensively studied topic in the literature, and thought to be due to the dysregulation of arteriolar endothelial function through the overproduction of oxygen free radicals causing reduced nitric oxide and arteriolar dilatation (shown in Fig. 1) [49].

There is a divergence in the results of the studies exploring the correlation between GC dose and hypertension in CAH, with some indicating a positive correlation [11, 18, 27, 50] while most failing to reveal any association [15, 19, 28, 32, 37]. A study including 11 children with CAH where the mean HC dose was 12.8 mg/m²/day for a mean duration of 11.7 years (range: 1.5–27.2 years) showed significantly higher mean awake ambulatory SBP (p = 0.04) and DBP (p = 0.02) [34]. Furthermore, a large multicentre cohort (n = 331) which analysed the correlation between HC dose and the BP at the age of one, two, and 3 years found a positive correlation between SBP and dose of HC at the ages of one (r = 0.44, p < 0.001) but no association at the later ages. There was also a correlation reported between SBP with calculated cumulative GC action of HC, and with dose of fludrocortisone (FC) (r = 0.47, p = 0.004). However, all children in this study received a higher mean HC dose above the recommended (more than 15 mg/m²/day) in the first 4.5 months of life [11].

A reverse circadian pattern of administration of HC has been associated with significantly higher daytime SBPs and DBPs and fewer nocturnal dips compared to the circadian GC regime, despite being associated with satisfactory biochemical control [26]. Prescribing GC in a reverse circadian pattern has not shown any added benefit, and is thus not recommended due to the possibility of unknown long-term adverse effects [51]. Duration of GC treatment was found to correlate with both elevated DBP and SBP (r = 0.31, p < 0.01; r = 0.26, p < 0.05) [12]. Moreover, hypertension has been more frequently observed in tightly controlled CAH with persistently lower 17-hydroxyprogesterone (17OHP) concentrations below 400 ng/dL (12.1 nmol/L) in the absence of a significant association with the HC dose [14]. Conversely, most studies failed to reveal a significant association between the increased BP and HC dose or biochemical parameters of dose adequacy (17OHP, A4), with the largest study reporting on 716 children with CAH aged 3–18 years [15]. Initial reports on adult patients (n = 39) comparing newer modified-release HC to prednisolone have shown promising results in disease control measured by improving androgen profiles with concentrations of 17OHP <36 nmol/L and A4 within the reference range [52]. However, there is yet insufficient data to support any direct benefits of modified release preparations on BP or other markers of metabolic risk.

The renin-angiotensin system has a vital role in maintaining BP and electrolytes in the body. Angiotensin II mediates these functions by arteriolar constriction via inhibiting nitrogen synthase and increasing aldosterone synthesis and release from the zona glomerulosa (shown in Fig. 1) [53‒57]. CAH affects the biosynthesis of adrenal steroids including MC synthesis, particularly in the salt-losing form. Additionally, FC is used by clinicians even in SV CAH owing to its GC-sparing effects. The international recommendation is to use a standard dose of FC of 50–200 μg given once daily [47], while admitting that there are limited data on MC treatment strategies and dose optimisation.

PRA and plasma renin concentration are commonly used measures of the adequacy of MC replacement. However, PRA alone appears not to be a reliable marker for monitoring and adjusting FC treatment due to its complex relationship with the FC dose and the variable factors that influence the accuracy of measurement. Clinical parameters including electrolyte concentration and BP should also therefore be considered when adjusting the FC dose [58]. Aldosterone was shown in one study in adults to correlate with DBP (r = 0.45, p = 0.002), but this has not been seen in paediatric cohorts and thus is used less frequently [59].

CAH patients of either sex prescribed higher doses of FC have a tendency towards increased BP, and suppressed PRA levels [11, 14, 18, 19, 32, 60]. A study which evaluated the BP beyond infancy up to 4 years found the lowest PRA (1.8 ± 2.7 ng/mL/h) at the age of 24 months, inversely correlated with BP (r = −0.5, p = 0.003), and a positive correlation between the SBP and DBP and the FC dose (r = 0.3, p = 0.005) [19]. Similar findings were reported by another study regarding the relationship between the SBP at the age of one and the FC dose (r = 0.41, p = 0.002). The cumulative MC action of FC and HC was positively correlated with both DBP and SBP (r = 0.30, p = 0.04; and r = 0.50, p < 0.001) [11]. In response to decreasing PRA levels, the FC dose was adjusted from 3 months of age when it was highest (344 ± 149 μg/m²/day) until the age of 2 years when it was lowest (25 ± 15 μg/m²/day) [19]. Conversely, retrospective data with calculated mean arterial pressure on infants with CAH due to 21OHD diagnosed in the newborn screening (n = 24) suggested a positive association between the serum renin concentration and the mean peak SBP and DBP only in weeks 0–2 (r = 0.61, p = 0.048; r = 0.72, p = 0.018, respectively), while there was a negative correlation in weeks 25–32 (r = −0.554, p = 0.026). There was no correlation beyond these ages, with only a few incidental high values in patients with low renin values. The same study did not find an association between BP and the FC dose [37]. The prompt dose adjustments in response to altered biochemistry, the short follow-up interval of 24 months and the small sample size may have contributed to this lack of association.

Tendency of increasing BP beyond infancy up to 24 months of age and its response to FC dose reduction can be expected from the increasing MC receptor sensitivity with age [61]. Considering the risk of hypertension when on FC treatment, it is important to re-evaluate the need of prolonged MC replacement in non-SW patients and it is important to adjust treatment on an individual patient basis depending on BP, growth, and biochemical markers of aldosterone deficiency. It is recommended that all children should be re-evaluated for persistent MC requirement during transition to adult care [47].

Atrial natriuretic peptide (ANP) has an action opposing the actions of angiotensin II and aldosterone in the intravascular volume homoeostasis. It causes natriuresis, diuresis, vasodilatation, and chronic restriction of intravascular volume resulting in control of BP [62, 63]. ANP concentrations have been suggested as a good marker of adequacy of MC replacement in CAH. There is a significant negative correlation between ANP and PRA when PRA is >5 ng/mL/h (r = −0.46, p < 0.0001). Higher FC doses are associated with suppressed PRA and concomitant elevated ANP levels (p < 0.05) [64]. However, data exploring the correlation between ANP and BP in CAH are limited to one study (n = 23 children and 11 adults) which suggested measuring ANP in clinical practice was a poor marker to detect hypertension in children and adults with CAH [65].

Infants have a lower glomerular function compared to adults and also have immature renal tubular function. Furthermore, breast milk consists of lower sodium content compared to the average sodium content in complementary food. Thus, infants with SW-CAH have a higher risk of hyponatremia, and should be treated with additional salt supplements at a dose of approximately 2 g/day [47, 66]. Salt supplementation can be stopped around the age of 6–8 months once weaning to adult food has begun [67]. However, clinical practice is heterogeneous: multinational data revealed that nearly 60% of children with 21OHD receive salt supplements during the infancy and 15% remain on salt supplements even at 2 years of age [11], while some do not advocate routine salt supplementation [68].

High salt intake in adulthood is an independent risk factor for hypertension. Reduction of BP has been shown in hypertensive non-CAH children with salt restriction [69]. However, a comparative retrospective multicentre registry study on 327 children, 203 of whom received salt at a dose of 0.81–1.03 g/day, did not identify a significant difference in BP in patients receiving salt supplements versus those without [23]. There are no further robust clinical trials that have tested the association of salt supplementation with developing hypertension in CAH, and further research in this area would be beneficial.

Increased body mass index (BMI) represents an independent risk factor for early hypertension and metabolic imbalance in childhood [70]. Both children and adults with CAH have been shown to have increased BMI and waist-to-hip ratio, suggestive of visceral adiposity and insulin resistance [12, 13, 20, 42, 71‒73]. GC treatment, GC receptor gene polymorphism (NR3C1), and excess androgen secretion due to poor disease control in CAH are known contributors of metabolic pathology [13, 24, 59].

Reports on increasing BMI centiles and associated hypertension in children with CAH are inconsistent. The dramatic increase of obesity prevalence in children across the world, mainly due to lifestyle factors, has doubtlessly contributed to this [74, 75]. Whether increased BMI is associated with raised BP in CAH is contentious, despite the large prevalence of obesity in patients within the studies that have investigated any association [18, 19, 32, 45]. Six studies [14, 15, 24, 28, 42, 71], including a large cohort of 726 CAH children [15], confirmed a relationship between weight and BP, the strongest correlation with SBP reported as r = 0.63, p < 0.001 [42]. 24-h ambulatory BP profiles revealed a significant association with high BMI in children and adolescents with CAH (age range: 5.3–19 years) with reported absence of the expected nocturnal dip [13, 27, 28, 38]. A positive correlation between BMI standard deviation score and the left ventricular mass index (LVMI) was demonstrated in a paediatric CAH cohort [25]. This may be interpreted as an early warning for hypertension and cardiac overload in later life.

Children with CAH tend to have increased leptin concentrations as a consequence of stimulatory effects by GCs on the secretion by the white adipose tissue [76]. This in turn most likely leads to metabolic dysregulation, in particular insulin resistance and increased androgen secretion from ovarian theca cells. Such factors contribute to obesity and cardiovascular disease with altered vascular tone in children with CAH. However, the association between hypertension and insulin resistance in CAH is not clear. One study of 27 children demonstrated a negative association between daytime DBP standard deviation score and homoeostatic model assessment for insulin resistance, suggesting the relationship between insulin resistance and BP in CAH is complicated and poorly understood [13].

Hyperandrogenism is closely associated with insulin resistance [57]. A cross-sectional study which used final adult height as a surrogate marker of disease control found higher risk of increased BP among SV females diagnosed after the age of one, suggestive of prolonged androgen exposure leading to early BP programming. However, there are no conclusive data in human studies to suggest the direct relationship between hyperandrogenism and hypertension in CAH [21, 77]. This review emphasises the need of more evidence on the association between hypertension, adiposity, hyperandrogenism, and insulin resistance in patients with CAH. The role and benefit of lifestyle modifications in alleviating the metabolic risks in CAH also warrants further research, to support targeted interventions and improve the lives of those living with this disease.

There is significant diversity in research evidence regarding the incidence of hypertension and its associations and causative factors in children with CAH. The prevalence of hypertension is highest in younger patients, with most studies reporting a decreasing trend as children go through adolescence into adulthood. There is a lack of consensus about the optimal approach to hormonal replacement that might mitigate the risk of hypertension, and particularly a lack of studies that inform the optimum dose of replacement MCs to use in CAH.

Most studies indicated transient hypertension in early childhood associated with FC treatment. However, whether this leads to adverse cardiovascular and metabolic effects is not well understood. Fine tuning of the FC dose in parallel to the increasing MC receptor sensitivity, alongside periodical reassessment for adequacy of treatment is recommended. However, the benefits of a GC sparing effect should be considered in the long-term treatment strategies of simple virilizing CAH due to its accompanying risk of iatrogenic hypertension. There is insufficient evidence to support a relationship between hypertension and either GC dose or salt supplement in infancy in CAH.

Persistent hypertension is rare in childhood and more obvious in adolescent and adulthood. Cardiovascular morbidity is high in young adults with CAH and increased CIMT has been described as an early surrogate marker of the risk thought to be driven by persistently high BP levels. However, up-to-date guidelines do not suggest routine screening to assess cardiovascular risk in children and adolescents with CAH [47].

Overall, this review highlights the need for more robust evidence that informs about the long-term impacts of raised BP in children with CAH. Since CAH is a rare disease, large-scale longitudinal, multicentre studies are required to investigate causal relationships between different treatment strategies. Long-term follow-up and data linkage between child and adult services will further improve our understanding of the most important contributory factors and consequences of different approaches to the management of CAH.

Authors declared no conflict of interests.

This article had no external funding.

N.P.K. conceived the manuscript outline. C.B. and N.R.L. performed bibliographic search and interpretation of data. C.B., N.R.L., I.A.B., R.K., and N.P.K. drafted the article, contributed to the critical revision of previous versions of manuscript, and approved the final version. N.P.K. and R.K. supervised the study.