Introduction

A reduction in renal mass leads to adaptation in the remaining nephrons. Beyond a certain limit, adaptation results in progression due to glomerular hyperperfusion, hyperfiltration, and hypertension, leading to a gradual decline in renal function [6]. Progression is characterized by proteinuria, hypertension, and a gradual drop in the glomerular filtration rate. This phenomenon is well documented in experimental models and a number of human disorders with reduced renal mass. In this study we asked the following question: does progression occur in patients with a single functioning kidney from birth within a period of 10 years?

Patients and methods

In the Renal Unit of our Department of Pediatrics, all patients with a single functioning kidney underwent renal function evaluation at the age of 2, 5 and 10 years. Physical examination included three blood pressure measurements, the average of which was expressed as age-related percentiles according to the American Task Force on Blood Pressure Control in Children [1]. In a urine sample the protein–creatinine ratio was measured and expressed as milligram per milligram (mg/mg), the normal value being <0.5 in infants and <0.25 beyond infancy. Serum creatinine was measured and expressed in milligram per deciliter (mg/dl). The glomerular filtration rate was measured as inulin clearance using the single shot technique and, once the children were continent, as creatinine clearance measured in 24-h urine collections.

In this paper we summarize our findings in 99 patients with a single functioning kidney observed in our renal unit over the past 25 years. All patients were diagnosed in the first year of life. Ultrasonography was used to define “normal” kidneys and multicystic kidney disease. Isotope scintigraphy was used to identify nonfunctioning kidneys (technetium-99m dimercaptosuccinic acid) and obstructed urinary tracts (technetium-99m mercaptoacetyltriglycine). Traditional voiding cystography was used to document vesico-ureteral reflux and the urethral valves. Patients were divided into three subgroups: A, patients with multicystic kidney disease (MCKD) having a normal contralateral kidney; B, patients with a normal solitary kidney without uropathy; C, patients with a nonfunctioning kidney within the framework of obstructive uropathy: grade IV and V vesico-ureteral reflux (VUR), ureteropelvic junction obstruction (UPJO), ureterovesical obstruction (UVJO), and posterior urethral valves. Patients with an underlying syndrome or other malformations were excluded, e.g. VATER sequence, Turner syndrome, tuberous sclerosis, etc. The Kruskal–Wallis analysis of variance or Mann–Whitney U test were used to analyze the differences among groups A, B, and C. The changes that occurred over 10 years in the patients were analyzed using the paired signed rank test. A p value <0.05 was considered statistically significant.

Results

Patients’ characteristics

Table 1 shows the number of patients according to subgroups and ages at investigation.

Table 1 Number of patients according to subgroups and ages at investigation
Full size table

Group A included 36 patients (23 boys). The multicystic dysplastic kidney was on the left side in 18 patients. Nephrectomy was performed in 4 patients. The following extrarenal abnormalities were present: ventricular septum defect (n = 1), undescended testicle (n = 3), hypospadias (n = 1), and inguinal hernia (n = 3). Urinary tract infections were documented in 3 patients.

Group B consisted of 20 patients (15 boys). The solitary kidney was left-sided in 11 patients. In 5 children there was an extrarenal anomaly: coarctation of the aorta (n = 1), undescended testicle (n = 3), and ureterocele (n = 1). In 1 patient a urinary tract infection was documented.

Group C was made up of 43 patients (27 boys). The nonfunctional kidney was on the left side in 22 patients. It had been removed in 11 children. VUR was observed in 27 patients; it was bilateral in 9 and surgically corrected in 13 children. UPJO was present in 7 patients, UVJO in 4, and urethral valves in another 5 patients. The latter patients had undergone surgery at least once. Extrarenal anomalies were undescended testicle (n = 3) and inguinal hernia (n = 1). Urinary tract infection was documented in 9 patients.

The records of the 10-year-old patients are incomplete because they did not appear at follow-up or were not yet 10 years old (see Table 1) at the time of writing.

Serum creatinine

The increase in serum creatinine levels with increasing age was statistically significant in every group. Group C had a significantly higher median creatinine level in all age categories compared with group A (p < 0.005, p < 0.05, p < 0.005 respectively). At the age of 10 years, the creatinine level was significantly lower in group A compared with patients in group B (p < 0.05); Fig. 1.

Fig. 1
figure 1

Time course of serum creatinine and glomerular filtration rate (GFR) in the different subgroups. Box plots indicate the upper and lower quartiles and ranges

Full size image

Glomerular filtration rate (ml/min per 1.73 m2)

Median glomerular filtration rate (GFR) did not change over time. Group C had a significantly lower GFR compared with group A (p < 0.005, p < 0.05, p < 0.05) respectively and B (p < 0.05, p < 0.005, p < 0.05) at every age (Fig. 1).

In group A, GFR was below 80 ml/min per 1.73 m2 in 2 out of 36 patients at 2 years, in 4 out of 36 patients at 5 years and in 2 out of 16 patients at 10 years. In group B, we noticed a GFR below 80 ml/min per 1.73 m2 in 3 out of 20 patients at 2 years, in 2 out of 20 patients at 5 years and in 1 out of 16 patients at 10 years. In group C, GFR was below 80 ml/min per 1.73 m2 in 14 out of 43 patients at 2 years, in 16 out of 43 patients at 5 years and in 9 out of 31 patients at 10 years.

Blood pressure

In the vast majority of patients studied over the 10-year period, systolic and diastolic blood pressure was below the 95th percentile. Median blood pressure figures did not increase with age. There were no significant differences among the three groups (Table 2).

Table 2 Blood pressure (expressed as age-related centiles): number of patients per subgroup
Full size table

Urinary protein–creatinine ratio

A urinary protein–creatinine ratio >0.25 mg/mg was surprisingly more frequent at the age of 2 years in all three subgroups of patients than later in life, whereas a ratio >0.5 mg/mg was only seen in group C. At 5 and 10 years of age, proteinuria was exceptional in groups A and B, but present in 12 out of 31 patients in group C. In group C, proteinuria increased significantly with age (p < 0.005); Table 3).

Table 3 Urinary protein–creatinine ratio (mg/mg): number of patients per subgroup
Full size table

Discussion

In this study we assessed serum creatinine levels, GFR measured with inulin or creatinine, blood pressure, and urinary protein excretion in 99 pediatric patients with a single functioning kidney which was a normal as judged by ultrasound. These patients, who were diagnosed in the first year of life, were studied at the age 2, 5, and 10 years. The cohort was subdivided into three subgroups according to the underlying disorder: multicystic kidneys (A), solitary kidneys (B), and obstructive uropathy (C). Low GFR, hypertension, and proteinuria were abnormalities found in all three subgroups, but with a significantly higher rate in subgroup C. This is no surprise since patients with obstructive uropathy are likely to be at risk of a contralateral anomaly. Yet, it is important to note that patients in groups A and B were not necessarily without anomalies. These patients were born with one functioning kidney. The main question is: are they at risk of progression? In other words, do they have a reduced number of nephrons?

Having one kidney from birth does not necessarily mean a reduced number of nephrons. It is likely that single kidneys can have extra nephrons as a consequence of intrauterine adaptation of nephrogenesis. This is suggested by findings in human beings [8, 12] and has been documented by elegant experiments in the lamb [7]. Nevertheless, the high incidence of a contralateral anomaly in patients with MCKD explains the variability of renal anomalies at follow-up [10].

There are a large number of publications on the long-term sequelae of nephrectomy in healthy kidney donors and for the sake of conciseness only two are mentioned here [9, 17]. It is felt that this very specific topic does not fall within the scope of this paper.

To our knowledge, nine more or less comparable studies on the fate of solitary kidneys occurring in childhood have been published over the past 22 years (Table 4) [2–5, 11–14, 16]. These studies, however, are difficult to compare for several reasons: the number of individuals studied varied considerably (from 27 to 202) as do the underlying kidney disorders (from MCKD over urethral valves to Wilms’ tumor) and the length of follow-up (from months to decades). Most studies are cross-sectional and there are major differences in methodology. In six reports GFR was measured either as creatinine clearance or inulin clearance, but in three GFR was estimated using the Schwartz formula (based on serum creatinine and body length), which is less accurate. Proteinuria was not included in one study, was assessed by dipstix in two studies, by urinary protein–creatinine ratio in two other studies, and measured in urine collection in the remaining studies. Lastly, the data are presented differently: sometimes as means and SD, sometimes as absolute values with or without ranges.

Table 4 Review of the literature
Full size table

Yet, all studies indicate that, in a variable percentage of patients, there are problems in terms of reduced GFR, arterial hypertension, and proteinuria. The earliest study, by Robitaille et al., is instructive [14]. Patients undergoing unilateral nephrectomy at a young age were studied as young adults. A reduced GFR was found in 3 out of 27, and proteinuria was found in 3 out of 27 former patients. The study that illustrates the best the risks related to solitary kidneys is the one by Baudoin et al. [5] on a heterogeneous cohort of 111 patients who underwent unilateral nephrectomy before the age of 16 years. The authors clearly demonstrate that the incidence of reduced GFR, hypertension, and proteinuria increases over time to become striking with a follow-up of over 25 years.

Only two studies are comparable to ours, since they include only pediatric patients followed longitudinally. Aslam et al. studied 165 children with MCKD [4]. At the age of 10 years, none had hypertension or proteinuria, but the estimated GFR was between 48 and 125 ml/min per 1.73 m2 with 2 out of 31 patients below 60 ml/min per 1.73 m2. Wilson et al. studied a mixed cohort of 74 patients followed over time [18]. These authors did not evaluate proteinuria or blood pressure; GFR measured as inulin clearance was below 85 ml/min per 1.73 m2 in 16 out of 74 patients. Seeman et al. were the only authors to distinguish between normal and abnormal solitary kidneys [15]. As in our study, it was shown that abnormalities were detected mainly in the latter.

Only two studies provide blood pressure data collected by 24-h ambulatory blood pressure measurement (ABPM), which is considered to be the optimal approach to hypertension in general, and to screen for blood pressure problems [13, 15].

From our study, and from other studies, it seems reasonable that individuals with a single functioning kidney for whatever reason deserve long-term follow-up during and after childhood. Additional prospective carefully designed studies are needed to identify those at risk of a progressive decrease in GFR, of which proteinuria and hypertension are the hallmarks. Studies could be optimized in several ways and these are our recommendations.

  1. 1.

    All patients born with a single functioning kidney deserve follow-up at regular intervals during and beyond childhood. The age of 2 years is an excellent moment to evaluate the kidney, which by that time has matured. The next step depends on what is found at age 2 years, but, as our study points out, ages 5 and 10 years are likely to be well chosen for the purpose.

  2. 2.

    Blood pressure should be measured, the best technique being the ABPM over a 24-h period [13]. Yet, this is hard to carry out at a young age and normal values are lacking at this time.

  3. 3.

    Urinalysis should include measuring proteinuria. The easiest way is measuring the protein–creatinine ratio in a random sample. More sensitive is microalbuminuria, which can also be estimated as a ratio over creatinine in a sample, but preferably in timed collections, as is done in children with diabetes mellitus. Microalbuminuria is a more sensitive marker for nephron hypertrophy and hyperfiltration.

  4. 4.

    Glomerular function needs to be estimated. Measuring serum creatinine is elementary, but not enough. For continent children, creatinine clearance measured in 24-hr urine collection is good enough for clinical studies. Other clearance techniques using inulin, iothalamate or chromium EDTA are better, especially in young patients.

  5. 5.

    Imaging is useful to distinguish between normal and abnormal kidneys. We would like to suggest ultrasound for measuring kidney volume rather than kidney length.

  6. 6.

    If signs of progression are documented, treatment with either an angiotensin-converting enzyme inhibitor or an angiotensin receptor antagonist is indicated. It is good to remember that a renal artery stenosis on a single kidney is a contraindication for the use of these drugs.