Blood pressure and glomerular filtration rate in youth with tuberous sclerosis complex

Renal involvement is very common in tuberous sclerosis complex (TSC) and is characterized by the development of angiomyolipoma and cysts. The aims of the present study were to assess kidney function and clinical features of renal involvement in TSC, including kidney function and blood pressure (BP) levels in children, adolescents and young adults. Non-selected patients with a definite diagnosis of TSC attending the paediatric neurology outpatient department of a tertiary hospital were included in a cross-sectional study. All participants had a renal imaging study within 6 months of ambulatory blood pressure (BP) and glomerular filtration rate (GFR) assessment. Data on demographics, history, genotype, kidney function at diagnosis and last imaging were collected. Twenty patients were enrolled in this study with a median age of 15 years (IQR range 9 to 18). About 23.5% of the participants had ambulatory hypertension. Systolic BP levels correlated significantly with GFRDTPA values despite the absence of hyperfiltration. Patients that developed hypertension and possibly those with angiomyolipoma or cysts had higher GFR levels in childhood and adolescence. All the patients with ambulatory hypertension had angiomyolipoma or cysts on renal imaging studies. Conclusions: Hypertension may present with increased frequency in young patients with kidney disease associated with TSC. Routine ambulatory BP measurement should be part of the annual clinical assessment in patients with TSC. What is Known: • Nearly half of the patients with TSC have a premature decline in their renal function in their fifth decade of life. • Hypertension and hyperfiltration have been proposed as modifiable factors of progression of renal decline in patients with TSC-related renal disease. What is New: • Hypertension is prevalent in youth with tuberous sclerosis complex. • SBP levels have a positive relation with GFR levels within the normal range of GFRDTPA values. What is Known: • Nearly half of the patients with TSC have a premature decline in their renal function in their fifth decade of life. • Hypertension and hyperfiltration have been proposed as modifiable factors of progression of renal decline in patients with TSC-related renal disease. What is New: • Hypertension is prevalent in youth with tuberous sclerosis complex. • SBP levels have a positive relation with GFR levels within the normal range of GFRDTPA values.


Introduction
Tuberous sclerosis complex (TSC) is a rare multisystemic genetic disorder inherited in an autosomal dominant manner. Its incidence has been estimated to be 1 per 5800 live births [1]. TSC is caused by mutations in one of the two tumour suppressor genes, TSC1 (9p34.13) and TSC2 (16p13. 3) [2 -4]. Pathogenic mutations to these two genes disrupt the mechanistic target of rapamycin (mTOR) pathway leading to the formation of benign tumours in various organs [5].
Renal involvement is very common in patients with TSC, and its importance is highlighted by studies demonstrating that renal disease is a major cause of mortality in the TSC population [6][7][8]. Angiomyolipoma (AML) is the most prevalent renal lesion in TSC followed by cystic disease [9]. Renal cell carcinoma (RCC) has also been described in young patients with TSC, and its incidence is estimated around 1-2% [10,11]. Besides AMLs and renal cysts, hypertension, proteinuria and hyperfiltration have been recently proposed as modifiable factors of progression of renal decline and, subsequently, preventive treatment targets [12]. However, to our knowledge, the prevalence of hypertension by ambulatory blood pressure (BP) monitoring and assessment of BP associations with measured GFR or imaging findings, that could guide early management, has not been investigated in young patients with TSC. The aims of the present study were to assess kidney function and clinical features of renal involvement in TSC, including GFR, albuminuria and BP levels in children, adolescents and young adults.

Population
In this study, we included children, adolescents and young adults with a confirmed TSC diagnosis who were followed up at the neurology outpatient clinic of the 1st Department of Paediatrics. Inclusion criteria in the study were as follows: (a) definite clinical or genetic diagnosis of TSC, (b) age > 5 years and age < 24 years, (c) kidney imaging study within 6 months of GFR and BP assessment and (d).consent to participate in the study. From the prevalent cohort, after excluding older patients transferred to adult care and 3 patients that did not attend their annual follow-up visit, 20 patients fulfilled the study criteria.
Informed consent to participate in the study was obtained by the children's parents or the patients themselves if older than 12 years. The human research protocol was conducted according to the Helsinki declaration for human clinical studies and approved by the institutional review board.

Demographic, clinical and imaging data collection
Demographics, family history, genotype, presenting features, organ involvement, serum creatinine at diagnosi, and last imaging findings were recorded from the patient files. The presence of AMLs and cysts, with the number, size and location were recorded from the last imaging study either magnetic resonance imaging (MRI), or ultrasound (US) or computed tomography (CT). The diameter of the largest AML and cyst was also documented for each kidney.
Microscopic urine examination and albuminuria were assessed on a morning urine spot. Albuminuria was also measured on a 24 h collection and was defined as albumin excretion rate > 30 mg/24 h or albumin to creatine ratio > 30 mg/g [13].

Glomerular filtration rate
GFR was measured by 99m Tc-DTPA scan (GFR DTPA ) in all the patients for the study protocol [14,15]. DTPA-GFR was performed once for the purposes of the current study protocol and is not routinely monitored in our TSC cohort. A single intravenous dose of 99m Tc-DTPA (diethylenetriaminepentaacetic acid) was administered to each patient according to the age and weight of the patient. Radioactivity was recorded by a multi-function well counter after selection of the adequate energy peak and window, each plasma sample and standard vial being counted twice. Background activity was also measured in the beginning and at the end of the counting. All the measurements were corrected for 99m Tc decay, taking into account the delay of time between the successive measurements in the well counter. The decrease in 99m Tc-DTPA in blood plasma with time is given by the injected dose divided by the area under the curve; the latter being best described by a biexponential function. Haycock method has been used for body surface estimation from height and weight. GFR was normalized to body surface area, and its values were expressed in ml/min/1.73 m 2 .
GFR was also estimated by using the original Schwartz formula (GFR Schwartz ) [16]. Serum creatinine was measured using the Jaffe-based creatinine-picrate forming in an alkaline medium [17]. The measurement was performed using the Architect c16000 automated analyser (Abbott Diagnostics Inc., Park City, IL, USA). DeltaGFR Schwartz (GFR Schwartz change) was also defined as the difference between GFR Schwartz at last visit and initial GFR Schwartz . Chronic kidney disease (CKD) stage was classified according to the KDIGO guidelines [13]. Hyperfiltration was defined as GFR ≥ 140 ml/min/1.73 m 2 [18].

Blood pressure measurements
Office BP measurement was performed by a trained physician using the Mobil-O-Graph 24 h PWA Monitor (IEM GmbH, Stolberg, Germany). Appropriate cuff size was used, and BP was measured on the right arm to the nearest 1 mmHg with the child or adolescent quiet, seated with the back supported and feet uncrossed on the floor after a 5-min rest. The mean of three measurements was used for the analysis. Office hypertension was defined in children and adolescents younger than 16 years old as systolic BP (SBP) and/or diastolic BP (DBP) ≥ 95th percentile for sex, age and height. For adolescents ≥ 16 years old, the cut-off for office hypertension was SBP/DBP ≥ 140/90 mmHg [19].
The patients underwent 24 h ambulatory blood pressure monitoring (ABPM) using the Mobil-O-Graph 24 h PWA Monitor (IEM GmbH, Stolberg, Germany) for the study protocol. ABPM was performed in our TSC patients once for the purposes of the current study protocol and is not part of their routine care. They were instructed to rest or sleep between midnight and 06:00 (night-time) and to maintain their usual activities between 08:00 and 22:00 (daytime). BP readings were obtained every 15 min during daytime and every 20 min during night-time. Fitting the monitor, reading and analysing the ambulatory BP data has been previously described [20][21][22]. ABPM sessions with 75% valid BP measurements were included in the study [23]. Three patients with cognitive impairment and autism did not tolerate ABPM. Ambulatory hypertension was defined as 24 h and/or daytime and/or night-time BP greater or equal to the 95th ambulatory BP percentile according to sex and height in study participants < 16 years, or the adult ambulatory BP limits in adolescents ≥ 16 years [19].

Pulse wave analysis
Central SBP (cSBP) and pulse wave velocity (PWV) were obtained by Mobil-O-Graph 24 h PWA monitor using the ARCSolver algorithm, which reconstructs the central pulse wave by applying a transfer function [24]. In children and adolescent's estimation of cSBP, using a Mobil-O-Graph device showed good accuracy compared with simultaneous invasive recordings in children and adolescents (age range 1-18 years) who underwent a cardiac catheterization [25]. Furthermore, reference values for cSBP and PWV are available for the children and adolescents using the Mobil-O-Graph device [26].

Statistical analysis
The IBM SPSS 24.0 (SPSS Inc., Chicago, Illinois, USA) statistical package was used to analyse the data. Data were tested for normality with Shapiro-Wilk test. Standard descriptive statistics, t-test or non-parametric methods (chi-Square, Mann-Whitney tests) were used as appropriate for the comparison between the groups. Bland-Altman analysis was used for the comparison of the GFR Schwartz equation with GFR DTPA . Regression analysis was also used to examine associations between GFR and BP values. Estimated marginal means (EMMs) after Bonferroni adjustment for multiple comparisons were used to assess for differences on final GFR and deltaGFR levels between patients' groups by BP levels or presence of AML or cysts. A p value < 0.05 was considered statistically significant.

General characteristics of the cohort
Median age at presentation was 36 (IQR 7-66) months. At the last visit, the median age was 15 (IQR 9-18) years. The mean duration of follow-up was 10.1 ± 5.58 years. The demographic and clinical characteristics of the cohort are presented in Table 1.

Kidney function
The mean GFR Schwartz  There were no patients with GFR < 60 ml/min/1.73 m 2 or albuminuria. However, four patients presented with stage 2 CKD at the end of follow-up. No patient presented with hyperfiltration by GFR DTPA (maximum value 122 ml/ min/1.73 m 2 ) . Five patients presented hyperfiltration by GFR Schwartz (maximum value 169.68 ml/min/1.73 m 2 ). However, GFR Schwartz was found to overestimate GFR values compared to GFR DTPA with a mean difference between the two methods of − 27.04 ml/min/1.73 m 2 and estimated intersubject variability (SD) of 23.44 (p < 0.05). Bland-Altman analysis showed poor agreement with wide limits of agreement (− 91.88 to 18.9). Still, there was statistically significant association between GFR DTPA and GFR Schwartz change after adjustment for age and sex (R 2 = 0.30, p = 0.014).
GFR levels did not differ between patients by the presence of AML, cysts or hypertension. GFR Schwartz change was higher in those with ambulatory hypertension, but the difference did not reach significance. However, GFR Schwartz

Blood pressure
BP z score levels did not differ significantly between patients with AML or cysts and those with no findings on kidney imaging (Table 2). Five patients (26.5%) had office hypertension.
AML or cysts was found in 3 out of 5 children (60%) with office hypertension (Fig. 1a). Amongst 17 patients who underwent ABPM, 4 (23.5%) patients had hypertension. All the patients with ambulatory hypertension had AML or cysts (Fig. 1b). They were all male; half of them had a positive family history of tuberous sclerosis and 3 had multiple bilateral AMLs with maximum size ≥ 10 mm ( Table 3). None of the hypertensive patients identified was receiving antihypertensive therapy, while 3 were receiving treatments with everolimus (Table 3). Finally, there was no difference in height adjusted z scores of PWV and cSBP among those with AML or cysts and those with no findings on kidney imaging.

Discussion
In the present study, we found a 23.5% prevalence of ambulatory hypertension and significant positive associations of GFR DTPA with SBP values. GFR DTPA was used to accurately assess kidney function in this cohort. GFR Schwartz overestimated kidney function in the cohort, and only GFR DTPA presented associations with BP parameters. The present study is to our knowledge the first one that investigated the association of imaging findings and GFR with BP levels using ambulatory BP monitoring. All children with ambulatory hypertension had AML or cysts on kidney imaging. RCC was not found in our study population. Greater increase in GFR Schwartz from initial diagnosis to the study visit resulted in higher GFR levels in childhood and adolescence in those that developed hypertension.
The patients with TSC develop renal impairment more frequently and at an earlier age compared to the general population [27]. In the present cohort, none of the patients had GFR < 60 ml/min/1.73 m 2 . Only GFR DTPA values were found to correlate significantly with ambulatory BP parameters. Hypertension is a known risk factor for progressive decline of renal function. Its incidence in the TSC population with renal disease is higher than in the general population; it increases with age and peaks in the fifth decade of life [9,28]. The prevalence of hypertension in youth with TSC is poorly investigated. Published results from the TOSCA (TuberOus SClerosis registry to increase disease Awareness) study reported a 5.7% rate of elevated blood pressure in patients with TSC, which was much lower compared to other published studies. When authors stratified hypertension by age, there was a clear association of raised blood pressure with age. There was no reported case of HTN in those under 9 years of age, whereas there were 10 cases in the age group of 9-18 years old [29]. A retrospective study in 35 children and young adults with TSC revealed a 25.7% prevalence of hypertension [30]. A recent multicentre study from Belgium including both paediatric and adult patients reported that 23% (8/35) of patients with TSC < 18 years old had hypertension [12]. A few other reports have highlighted the correlation between TSC-associated renal disease and hypertension in the paediatric age group; the majority of which are TSC2-PKD1 contiguous gene syndrome cases [31][32][33][34][35]. In those patients, hypertension is often discovered in infancy, and the renal function decline is rapid.
The primary cause of hypertension in TSC is the formation of cysts or AML occupying the renal parenchyma. Interestingly, there are a few reports of patients who initially presented with hypertension associated with abdominal pain or distention and eventually diagnosed with TSC [36][37][38]. Of note, TSC can rarely be associated with vascular malformations leading to renovascular occlusion and refractory hypertension [39]. Regarding cardiac involvement in our study population, four patients had cardiac rhabdomyomas and in one of them were identified in antenatal scans. None of them was symptomatic. Finally, a significant number of patients with TSC may be under antiepileptic drug treatment, but there is no solid evidence of any effect of these drugs on BP [40].
It is of interest that a statistically significant increase from initial GFR was noted to those with ambulatory hypertension. Despite the greater increase in GFR Schwartz , none of the patients presented hyperfiltration with GFR DTPA . Five patients presented hyperfiltration with GFR Schwartz , but there was no difference in the prevalence of hyperfiltration in hypertensives compared to normotensive patients. Moreover, GFR Schwartz seems to overestimate the patients' GFR compared to the gold standard measurement. Still, the magnitude of GFR change since  [18]. There are no current studies investigating the underlying mechanisms of hyperfiltration in the TSC population. However, the encroachment of normal renal parenchyma by growing AML or cysts and the subsequent surgical interventions (i.e., nephrectomy or embolization), which lead to a reduced number of functional nephrons, could result in compensatory hemodynamic changes in response to nephron loss. As a consequence, there will be increased renal blood flow to the remaining normal renal parenchyma potentially leading to intraglomerular hypertension. Another hypothesis is that mTOR pathway overactivity caused by TSC1/TSC2 haploinsufficiency may lead to glomerular hypertrophy and hyperfiltration [41]. The absence of hyperfiltration by GFR DTPA in the present cohort could be explained by the low prevalence of renal lesions (45% and 20% of the study group had AMLs and cysts, respectively) compared to other paediatric cohort studies [12,42]. Additionally, none of the patients underwent surgical intervention. Growth of TSC-related renal lesions could impair intraglomerular pressure and overall renal function thus leading to hypertension. However, the mechanisms that link  The small population size is the main limitation of the present study. Moreover, the cross-sectional design does allow to establish causal relationships. The non-standardized imaging protocol could have impacted the comparison between the study participants. Nonetheless, all the patients that underwent a renal ultrasound had no kidney involvement. The lack of genetic testing was another study limitation as the impact of TSC2 or TSC1 mutation on the development of HTN could not be assessed. The study design also did not allow to study the effect of everolimus treatment on GFR and BP which has been previously associated in adults with reduction in the volume of AMLs [43]. On the other hand, objective methods, 99m Tc-DTPA and ABPM were used to accurately assess kidney function and BP status in the cohort. The Mobil-O-Graph 24 h PWA monitor has not been validated for ABPM, cSBP and PWV measurement in children. However, for central BP a recent study showed that values obtained with the methodology used in the present study have good accuracy compared with invasive measurements [30]. In the current study, we performed out of office 24 h BP and 24 h PWV measurements, which are known to have superior predictive value.
In conclusion, the present study provides preliminary evidence that hypertension is prevalent in youth with TSC and that SBP levels have a positive relation with GFR levels within the normal range of GFR DTPA values. The rate of GFR change since presentation may reflect a mechanistic insight in the development of hypertension at the early stages and younger ages. Children with AML or cysts would benefit from routine evaluation of their ambulatory BP levels. Still, future prospective studies in larger cohorts are needed to evaluate hypertension progression in youth with TSC.