Low prevalence of hypertension in children with renal cysts and diabetes syndrome is the hallmark of the disease

Cystic kidney diseases such as autosomal recessive or dominant polycystic kidney disease (ARPKD and ADPKD) are associated with high prevalence of arterial hypertension. On the contrary, studies on hypertension in children with renal cysts and diabetes (RCAD) syndrome caused by abnormalities in the HNF1B gene are rare. Therefore, the aim of our study was to investigate the prevalence of hypertension in children with RCAD syndrome due to HNF1B gene abnormalities and to search for possible risk factors for development of hypertension. Data on all children with genetically proven RCAD syndrome from three pediatric nephrology tertiary centers were retrospectively reviewed (oce blood pressure (BP), ambulatory blood pressure monitoring (ABPM), creatinine clearance, renal ultrasound, echocardiography, albuminuria/proteinuria). Hypertension was dened using the current ESH 2016 criteria and/or by the use of antihypertensive drugs. Thirty-two children with RCAD syndrome were investigated. Three children received ACE-inhibitors for hypertension and/or proteinuria. Hypertension was diagnosed using oce BP in 22% of the children (n = 7) In the 7 performed ABPM 1 child (14%) was diagnosed with hypertension. Creatinine clearance, proteinuria, albuminuria, body mass index, enlargement or hypodysplasia of the kidneys and prevalence of HNF1B-gene deletion or mutation were not signicantly different between hypertensive and normotensive children. Conclusion: The low


Introduction
What is known: Arterial hypertension is a common complication in children with polycystic kidney diseases.

What is new
Hypertension is a rare clinical manifestation of children with renal cyst and diabetes (RCAD) syndrome.
The low prevalence of hypertension seems to be the hallmark of children with RCAD syndrome.
Cystic kidney diseases belong to the most common structural kidney abnormalities with autosomal dominant polycystic kidney disease (ADPKD) being the most common one and autosomal recessive polycystic kidney disease (ARPKD) being the most severe with high morbidity and mortality [1,2]. Both polycystic kidney diseases are associated with high prevalence of arterial hypertension in childhood.
Elevated blood pressure is documented in 35% of children with ADPKD [3,4] and in 75% of children with ARPKD [2]. By contrast, studies on hypertension in children with renal cysts and diabetes (RCAD) syndrome due to HNF1B gene abnormalities are lacking. Adult patients with RCAD syndrome show a variable prevalence of hypertension ranging from 7% to 80% [5 -7]. Only one pediatric study investigating patients with HNF1B mutations showed normal mean blood pressure (BP) in comparison to healthy children. Other pediatric studies on patients with RCAD syndrome did not give any data on BP or arterial hypertension.
Therefore, the aim of our study was to investigate the prevalence of hypertension in children with RCAD syndrome due to HNF1B gene abnormalities and to search for possible risk factors for its development.

Subjects
A total of 32 children (20 boys and 12 girls) with RCAD syndrome were retrospectively analyzed using the medical records of three tertiary pediatric nephrology centres in Prague, Jena, and Munich. The inclusion criteria were genetically proven RCAD syndrome (whole HNF1B-gene deletion or pathogenic variant of the HNF1B-gene in combination with presence of any renal anomaly) and o ce BP data. Children with CKD stage 5 were excluded from the study. The median age was 3.0 years (range 0.3-18.5 years). The HNF1Bgene deletion was found in 72 % of patients and a pathogenic variant of the HNF1B-gene in 28 % of patients. Renal ultrasound was pathological prenatally in 48 % of patients and postnatally in 52 % of children. The median age at diagnosis was 3.0 years (range 1 day -18 years).

Antihypertensive medication
Three (9%) children received ACE-inhibitors for hypertension and/or proteinuria. No child received any other antihypertensive medication at the time of the study.
O ce blood pressure measurement O ce BP was measured using an oscillometric device Omron and hypertension (HT) was de ned as systolic and/or diastolic BP ≥ 95th percentile according to the current ESH guidelines [8] and/or the use of antihypertensive medication.
Ambulatory blood pressure monitoring (ABPM) ABPM was performed in 7 patients because of elevated OBP (n=2) or because of CKD despite of normal OBP (n=5). In the remaining 25 patients ABPM was not performed mainly because of normal o ce BP.
ABPM studies were carried out using oscillometric SpaceLabs 90207 or 90217 monitors (SpaceLabs Medical, Redmont, WA). An appropriate cuff was placed on the non-dominant arm by a physician who also informed the child and parents in detail how to operate the monitoring system. Monitors were programmed to measure BP automatically every 20 min during the day and every 30 min at night. The criteria for omitting BP outliers from ABP recordings were systolic BP >200 and <70 mmHg, diastolic BP >150 and <40 mmHg, and mean arterial pressure (MAP) >200 and <40 mmHg. According to the reference values by Wuhl et al. [9], data were analyzed by using an individualized daytime and nighttime period according to the individual diaries. Mean systolic and diastolic BP at daytime and nighttime were calculated. The BP index was calculated as mean BP, divided by the 95th percentile that was determined according to the body height and sex of the patient [9]. For children with body height <120 cm the 95 th percentile was used for children with body height 120 cm. Ambulatory HT was de ned as systolic and/or diastolic mean BP at daytime and/or nighttime ≥95 th percentile.

Urinary excretion of albumin and protein
Freshly voided urine ( rst morning urine) was obtained for quantitative measurement of total protein (Biuret method), albumin (turbidimetry) and creatinine (enzymatic). Pathological albuminuria was de ned as albumin/creatinine ratio >3 mg/mmol creatinine and proteinuria as protein/creatinine ratio >22 mg/mmol creatinine [10].

Renal function
Chronic kidney disease (CKD) stages 1 -4 according to the K-DOQI guidelines were detected [K-DOQI] using estimated glomerular ltration rate according to the Schwartz formula using serum creatinine (enzymatic method) and body height [11]. Children with CKD stage 5 were excluded from the study.

Renal ultrasonography
Data on renal ultrasound ndings were collected (cysts, renal length, dysplasia, dilation of renal pelvis). Kidney lengths were analysed, compared with normal standards and expressed as SDS [12].

Echocardiography
Echocardiogram (standard two-dimensional echocardiogram (GE/Wingmed system 5, Vivid 7, Horten, Norway) was performed in 8 children on the same day as the o ce BP measurement (according to the recommendations of the American Society of Echocardiography) [13]. Left ventricular mass (LVM), was calculated according to the formula of Devereux from the left ventricular internal dimension at end diastole, interventricular septal thickness and left ventricular posterior wall thickness [14]. Left ventricular mass was indexed to height2.7 (left ventricular mass index LVMI) to account for body size [15]. Left ventricular hypertrophy (LVH) was de ned as LVMI ≥95 th percentile for normative pediatric LVMI data [16].

Extra-renal ndings
The extra-renal ndings (maturity onset diabetes of the young -MODY, hypomagnesemia or liver abnormalities) and medical history data such as birth weight or oligohydramnios were collected retrospectively from the medical charts.

Statistical analysis
The data were analysed by using the STATA software package. SDS values of hypertensive and normotensive patients were investigated by using the Mann-Whitney U test. Values with p<0.05 were considered statistically signi cant.

Results
O ce blood pressure measurement Seven out of 32 patients (22 %) were regarded as hypertensive on the basis of the o ce BP measurements or by the use of antihypertensive drugs. All hypertensive children had stage 1 hypertension. Three patients had isolated diastolic hypertension, three patients combined systolic and diastolic hypertension and no child had isolated systolic hypertension. One patient on ACEI therapy had normal o ce BP values. Among children with CKD stage 2 -4 (n=11) 36% were hypertensive on contrary to 14% in children with CKD stage 1 and normal eGFR (p=0.13).

Ambulatory blood pressure monitoring
Only one patient from seven measured children (14%) was found to have hypertension de ned by ABPM.
He had borderline isolated night-time diastolic hypertension with a diastolic BP of 65 mmHg (95 th percentile 65 mmHg). He had also hypertension de ned by o ce BP. One patient had white-coat hypertension, no patient showed masked hypertension and ve children were normotensive both by ABPM and o ce BP.

Risk factors for hypertension
The data on risk factors for development of hypertension are given in Table 1. In summary, no one from the investigated risk factors for hypertension were signi cantly different in hypertensive in comparison to normotensive children with RCAD syndrome Echocardiography No child from eight investigated children (three hypertensive and ve normotensive) had left ventricular hypertrophy.

Discussion
Arterial hypertension is present in more than 90% of perinatal diagnosed ARPKD patients [2] and in 35% of children with ADPKD [3,4]. In contrast to this high prevalence of hypertension in children with polycystic kidney diseases, the frequency of hypertension in children with RCAD syndrome is not known.
Our study is the rst systematic investigation of blood pressure and hypertension in children with RCAD syndrome. We could demonstrate that hypertension is rare in children with RCAD syndrome. Only 14-22% of children with HNF1ß mutation presented with arterial hypertension, detected by o ce BP or by ABPM, mainly those with decreased eGFR, i.e. chronic kidney disease stages 2-4.
In adults with RCAD syndrome the data on prevalence of hypertension are very heterogeneous. Several studies demonstrated prevalence ranging from 0% to 80% [5 -7, 17]. A French study on 27 adult patients showed a very low prevalence of hypertension of only 7% [5]. These authors stated, that this low prevalence of hypertension -together with the slowly progressive kidney function is the hallmark of the RCAD disease [5]. By contrast, an even smaller Brazilian study showed a prevalence of hypertension of 80% adults with RCAD syndrome [7]. Dubois-Laforgue et al. reported in one of the largest cohort studies on more than 200 adults a prevalence of 58% of hypertension in patients with RCAD syndrome. The CKD stages 3-5 were the most important risk factors associated with elevated BP. The analysis showed a trend for more hypertension in patients with HNF1ß-mutations in comparison to patients with HNF1B-deletion [6]. In our study, a numerically higher prevalence of hypertension was noted in children with CKD stages 2-4 (36%), compared to those patients with CKD stage 1 and normal renal function (14%). Thus it seems that the prevalence of hypertension in patients with RCAD syndrome depends mainly on the stage of CKD, eGFR and somewhat age of the patients (higher prevalence in adults than in children).
The prevalence of hypertension 14-22% in our children with RCAD is considerably lower than in children with ARPKD (75%) [2] or ADPKD (35%) [3,4]. The reason for this large discrepancy could be the amount of cystic degeneration of the kidneys being the biggest in patients with ARPKD. Those patients demonstrate the highest prevalence of hypertension and decreased GFR. By contrast, patients with RCAD syndrome have usually normal sized or small kidneys with only small cysts and show a low prevalence of arterial hypertension. This low prevalence of hypertension seems to be the hallmark of this disease in children as it has been stated in adults [5]. Therefore, RCAD syndrome resembles more congenital anomalies of the kidney and urinary tract (CAKUT), where the prevalence of hypertension is also very low [18].
So far, there are no studies dealing primarily with BP or hypertension in children with RCAD syndrome.
Thomas et al. investigated patients with HNF1B (n=4) and PAX2 gene mutations from the CKiD (Chronic Kidney Disease in Children) study and stated that children with HNF1B and PAX2 gene mutations have normal blood pressure. [19]. Unfortunately, they did not indicate the prevalence of hypertension in this cohort. Only the BP percentiles were mentioned (62 nd and 69 th for systolic and diastolic BP), which were normal. Other larger studies did not show data on the prevalence of HT or on BP levels [20 -27]. The relatively low prevalence of hypertension in our cohort and the fairly normal BP in the very small cohort from the CKiD study underline that there is a low prevalence of hypertension among children with RCAD syndrome. This low prevalence seems to be the hallmark of this genetic multisystem disease [5] and is similar to autosomal dominant tubulointerstitial kidney disease [28]. Indeed, one subtype of ADTKD is caused by deletion/mutation of the HNF1B gene (named HNF1B-ADTKD). Moreover, both diseases are primarily tubular diseases which show in general a low prevalence of arterial hypertension.
In our analysis, it was very di cult to discover risk factors for hypertension in the few hypertensive children with RCAD syndrome due to the very low number of hypertensive patients (n=7). We did not nd any statistically signi cant marker associated with arterial hypertension. The only trend for higher prevalence of hypertension was CKD stages 2-4. This nding is in line with published data from the largest adult cohort study demonstrating CKD stages 3-5 being the only risk factor for the development of hypertension [6]. In adult and pediatric patients with ADPKD the kidney size is positively associated with BP and patients with larger kidney size have higher risk of hypertension [4,29,30]. In our cohort of children with RCAD syndrome, only 10% of children had enlarged kidneys, a percentage similarly low to the prevalence of hypertension. Furthermore, more children (40%) had small kidneys or unilateral renal agenesis that do not cause hypertension.
In conclusion, we have demonstrated that children with renal cyst and diabetes syndrome have a low prevalence of arterial hypertension.This seems to be the hallmark of the disease in children.