Introduction

Neurofibromatosis type 1 (NF1) is one the most common autosomal-dominant diseases (estimated prevalence 1/2,500) [1]. The protein encoded by the NF1 gene, neurofibromin, regulates the Ras proto-oncogene that plays important roles in cell growth and differentiation [2, 3]. The principal features of NF1 are café-au-lait spots, skinfold (axillary and inguinal) freckling, cutaneous neurofibromas, and pigmented iridal hamartomas (Lisch nodules) [3]. Vasculopathy is relatively common and may affect all blood vessels, especially the renal arteries. Renovascular hypertension caused by renal artery stenosis and renal artery aneurysms are the most common vasculopathic presentation in NF1 patients [4, 5]. The etiology of NF1 vasculopathy remains poorly understood. However, neurofibromin is expressed in the vascular endothelium, and may thus possibly play a role in the etiology of vasculopathy [6]. Neurofibromas are prominent in NF1 patients. These benign nerve tumors may present as focal growths or extend longitudinally along nerves, thus involving multiple fascicles. Plexiform neurofibromas can affect the genitourinary system, obstructing the urinary tract and triggering hydronephrosis [7,8,9]. Hypertension may develop even in childhood, and affects many adults with NF1. Although primary (essential) hypertension is most common, NF1 increases the risk of secondary hypertension caused by renal artery stenosis, and pheochromocytoma [3, 10]. Glomerular diseases including membranous nephropathy [11,12,13,14], immunglobulin A nephropathy [15], idiopathic nephrotic syndrome [16], focal segmental glomerulosclerosis [17, 18], secondary glomerulonephritis [19], and acute [20] and chronic [21] kidney injuries have also been reported in NF1 patients.

Here, we evaluated renovascular hypertension, urinary abnormalities, voiding dysfunction, and renal tubular and glomerular function in children with NF1. To the best of our knowledge, this study is the first to evaluate comprehensively renal findings and glomerular and tubular function in such children.

Methods

Study population

NF1 patients aged 0–18 years followed up in our pediatric neurology department who fulfilled the National Institutes of Health Neurofibromatosis Type 1 criteria [22] were prospectively recruited. Same-aged children who lacked chronic diseases but visited our pediatric outpatient clinics for routine follow-up served as controls. Our ethics committee approved the study. Written informed consent was obtained from all parents. All subjects underwent physical examinations by the same pediatric neurologist and nephrologist (two professionals). Age, sex, weight, height, and blood pressure were recorded. The same practitioner performed blood pressure measurements at least three different times on separate visits. Blood pressure measurement and evaluation were performed according to the 2017 American Academy of Pediatrics Guidelines for Childhood Hypertension [23]. Information on voiding dysfunction, enuresis, and recurrent urinary tract infection were obtained from parents during visits to the pediatric nephrology and neurology outpatient clinics. The data of the two groups were compared.

Imaging and laboratory evaluation

All subjects underwent urinary ultrasound and renal Doppler ultrasonographic evaluation. Conventional or magnetic resonance (MR) angiography was scheduled for patients with suspected renal artery stenosis apparent on Doppler ultrasound evaluation. Patients with recurrent urinary tract infections were scheduled for dimercaptosuccinic acid (DMSA) scans; if renal scars were found, voiding cystourethrography was performed to explore vesico-ureteral reflux status. Urine and blood samples were obtained for renal, and glomerular and tubular function, tests. A complete blood count was performed, and the levels of blood urea nitrogen, as well as serum creatinine, sodium, potassium, calcium, phosphorus, magnesium, and uric acid, were measured. To calculate the tubular reabsorption of phosphorus (TRP), the urinary excretions of sodium, potassium, magnesium, calcium, phosphorus, uric acid, and creatinine were determined. The random urine protein-to-creatinine ratio was calculated to evaluate proteinuria status. The TRP was calculated using the formula [1-(plasma creatinine x urine phosphorus/urine creatinine x plasma phosphorus)]x100, and values below 85% were considered abnormal. The patient urine electrolyte-to-creatinine ratio was compared to that of the control group. The glomerular filtration rate was estimated using the formula of Schwartz et al. [24] with κ = 0.413. For hypertensive cases, the levels of urine catecholamines (normetanephrin, metanephrin, and vanillyl mandelic acid) were measured to exclude pheochromacytoma.

Statistical analysis

Statistical analyses were performed using IBM SPSS for Windows ver. 22.0. Categorical variables were compared using the χ2 test. Normality of numerical variable distributions was evaluated employing the Shapiro–Wilk test. If the data were normally distributed, the Student t-test (on means ± SDs) was performed. The Mann-Whitney U-test was used to compare data with skewed distributions (the median [min-max] values were compared). The Pearson correlation test was performed to assess relationships among variables with normal distributions. A p-value < 0.05 was considered significant for all analyses. The Spearman correlation test was performed to assess the relationships among variables with skewed distributions; again, a p-value < 0.05 was considered statistically significant.

Results

We included 46 (female/male: 20/26) NF1 patients aged 1.5–18 years and 33 (female/male: 15/18) healthy children aged 3–18 years. The mean ages of patients and controls were 10.1 ± 4.6 and 10.6 ± 4.3 years respectively. The mean systolic and diastolic blood pressures of the NF1 group were 110 ± 17 and 70 ± 11 mmHg respectively, and those of the control group 101 ± 11 and 64 ± 6.4 mmHg. The mean systolic and diastolic blood pressures were significantly higher in patients than controls (both p < 0.05) (Table 1).

Table 1 Clinical and laboratory data of the included patients
Full size table

Systolic blood pressure was between the 75th and 90th percentile in three (6.5%) subjects of the NF1 group, between the 90th and 95th percentile in four (8.7%), and above the 95th percentile in three (6.5%). In the control group systolic blood pressure was below the 75th percentile in all subjects. Diastolic blood pressure was between the 75th and 90th percentile in two (4.3%) subjects of the NF1 group, at the 90th percentile in eight (17.4%) patients, between the 90th and 95th percentile in three (6.5%), and above the 95th percentile in six (13%). In the control group, except two subjects who had diastolic blood pressures between the 75th and 90th percentiles, all the subjects had diastolic blood pressures below the 75th percentile according to the age, height, and sex of the subject.

One patient in NF1 group who had systolic and diastolic blood pressures above the 99th percentile was evidenced renal artery stenosis on Doppler ultrasound evaluation. This patient underwent conventional angiography and a stent was put in place. In another hypertensive NF1 patient, renal artery stenosis was suspected on renal Doppler ultrasound. However, renal MR angiography revealed normal renal arteries. Renal artery stenosis was apparent in one of six (17%) hypertensive patients. The urine catecholamine levels of hypertensive children were normal. Hypertension was not detected in any control subject. Urinary system abnormalities were found in 22% of the NF group, but only 9% of the control group (p < 0.05). The most common urinary anomaly was hydronephrosis, detected in six (13%) NF1 patients. The other urinary system disorders of NF1 patients were nephrolithiasis (one patient), recurrent urinary tract infections (two), a pelvic kidney (one), vesicoureteral reflux (one), and solitary functioning kidney (one). Only three control children exhibited hydronephrosis. No tumor was detected in any subject. The prevalence of primary enuresis was higher in the NF1 group [10 patients (22%) compared to five (15%) controls]. Voiding dysfunction was slightly more common in the NF group. Neither difference was statistically significant.

The mean eGFRs of the NF1 and the control groups were 113 ± 34 and 134 ± 49 mL/min/1.73 m2 respectively. The mean eGFR of the NF1 patients was lower than that of controls (p = 0.031). In six NF1 patients, the eGFR was < 90 mL/min/1.73 m2. The random urine protein/creatinine ratios were similar in both groups (0.13 ± 0.11 vs. 0.11 ± 0.03 mg/mg) (p = 0.23). The mean phosphorus tubular reabsorption of NF1 patients was 91%, but 94% in controls (p = 0.03). The calcium-to-creatinine ratio of the patient group was significantly higher than that of the control group (0.16 ± 0.18 vs. 0.10 ± 0.07 mg/mg) (p = 0.015). The spot uric acid-to-creatinine ratio of the NF1 group was slightly higher than that of the control group (0.72 ± 0.43 vs. 0.47 ± 0.34 mg/mg) (p = 0.01). The fractional excretion of uric acid was 13 ± 7% in the NF1 group and 8.3 ± 4.5% in the control group (p = 0.003). The groups did not differ in terms of the percentage fractional excretions of sodium (FeNa) (p = 0.82), potassium (FeK) (p = 0.81), or magnesium (FeMg) (p = 0.78) (Table 1). Glycosuria was not detected in either group.

Discussion

All of hypertension, urinary system abnormalities, and changes in renal glomerular and tubular functions were more common in NF1 than healthy children. The mean systolic and diastolic blood pressures of normotensive NF1 patients were higher than those of healthy children. Renal artery stenosis and middle aortic syndrome are the two most common causes of renovascular hypertension in children; both can be associated with fibromuscular dysplasia, Williams syndrome, NF1, tuberous sclerosis, and certain vasculitic syndromes [25,26,27].

Vasculopathy is a frequently described complication of NF1 associated with hypertension in 13–15% of pediatric patients [26, 28]. Hypertension was detected in 12.6% of our NF1 patients. In another study evaluating 27 NF1 patients aged 4–24 years, the hypertension rate was 18.5% [20]. In a larger cohort of NF1 patients, the hypertension rate was 6.1% [27]. Several case reports on the relationship between hypertension and NF1 have appeared [28, 29].

Uncontrolled systemic hypertension may give rise to stroke; prompt diagnosis and appropriate treatment are important [30]. Antihypertensive medication, percutaneous transluminal angioplasty, or surgery can be used to treat hypertensive NF1 patients [31]. In a study on the congenital anomalies of NF1, hydronephrosis was detected in three and ureter duplication in one of 22 patients [32]. A horseshoe kidney [33, 34], bladder involvement caused by neurofibromas accompanied by urinary incontinence [8], an autosomal-dominant polycystic kidney [35, 36], anuric acute kidney injury caused by renal artery occlusion and a thromboembolism [20], and chronic renal disease attributable to bladder neurofibromas [21] have been (rarely) reported in NF1 patients, as have glomerulonephritis [19], focal segmental glomerular sclerosis [17, 18], C1q nephropathy [37], and Joubert Syndrome with nephronophthisis [38].

We found that urinary system abnormalities were more common in NF1 patients than healthy children. The most common urinary anomaly was hydronephrosis (13%). Urogenital involvement of the bladder and bladder neck caused by neurofibromas arising from the autonomic neural plexus has been reported in NF1 patients [8, 9, 21, 39]. Genitourinary tract involvement may trigger urinary obstruction and hydronephrosis, voiding dysfunction, and enuresis. In agreement with the literature, voiding dysfunction and enuresis were common in our NF1 patients. Rarely, Wilms tumors and angiomyolipomas may develop in NF1 patients [40,41,42,43,44].

The mean eGFR of NF1 children was lower than that of controls; eGFRs < 90 mL/min/1.73 m2 were evident in six patients. We speculate that renal vasculopathy per se causes a slight decrease in the GFR; also, neurofibromas, some of which are not detected by conventional imaging modes such as ultrasonography, may exert pressure on the urinary tract, inflicting obstructive damage.

In terms of tubular function (e.g., the TRP and the urinary fractional solute excretions), the mean phosphorus tubular re-absorption was lower in NF1 patients than controls, but the random urine uric acid-to-creatinine ratio and the fractional excretion of uric acid were higher in patients. These differences may reflect kidney vasculopathy. The urine calcium-to-creatinine ratio of patients was significantly higher than that of controls, perhaps reflecting a change in calcium metabolism caused by mutation of the NF1 tumor suppressor gene [45] or altered calcium channel density [46]. Such speculations require clinical confirmation.

To the best of our knowledge, this is the first study to evaluate glomerular and tubular function in pediatric NF1 patients. Adult NF1 patients may have additional renal issues; these have not been adequately studied. A limitation of our study is that hypertension was not diagnosed via 24-h ambulatory blood pressure monitoring. We may thus have underestimated hypertension (by missing “masked hypertension”) or may have (incorrectly) scored the white coat effect as hypertension. Further studies are needed.

Conclusions

Hypertension and urinary tract anomalies were more common in NF1 children than controls. Renal glomerular and tubular function were compromised in NF1 patients. Such patients must undergo annual blood pressure monitoring and, if a clinical suspicion of secondary hypertension is aroused, the etiology thereof must be elucidated. Clinicians should consider ultrasound screening of NF1 children to detect urinary anomalies in settings of hypertension or urinary abnormalities. Glomerular and tubular function should be checked if needed, and selected cases should be referred to pediatric nephrologists.