Pediatr Nephrol (2009) 24:11651172 DOI 10.

1007/s00467-008-1110-9

ORIGINAL ARTICLE

Circadian rhythms of diuresis, proteinuria and natriuresis in children with chronic glomerular disease
Amira Peco-Anti & Jelena Marinkovi & Divna Krui & Dusan Paripovi

Received: 13 August 2008 / Revised: 15 December 2008 / Accepted: 16 December 2008 / Published online: 28 January 2009 # IPNA 2009

Abstract The aim of our study was to examine diurnal variation in urine volume (UV) output, proteinuria (UPRT), urine creatinine (UCr) and urine sodium ion excretion (UNa) in children with chronic glomerulopathy. In 56 patients (20 boys/36 girls, aged 11.70.6 years) samples for UPRT, UCr and UNa were collected during the day and night, with continuous ambulatory blood pressure (BP) monitoring. On the basis of creatinine clearance (CrCl) the patients were divided into group I (n =44, with CrCl 131 3.6 ml/min per 1.73 m2 body surface area), or group II (n = 12, with CrCl 44.67.7 ml/min per 1.73 m2 body surface area). Nocturnal polyuria was defined as night time UV 35% of the 24 h UV. Age, gender, body mass index of the patients, 24 h UV, UCr and UNa were similar in both groups. However, arterial hypertension and nocturnal polyuria were widespread (P <0.01) in group II. In addition, proteinuria was higher (P <0. 05) in group II. The nocturnal decline in CrCl, UV, UPRT and UNa was significantly attenuated (P <0.005) in patients in group II compared with those in group I. The night time mean arterial pressure (MAP), as well as the night/day ratios of MAP, UV, UPRT and UNa, showed negative associations with CrCl. Our findings strongly suggest that renal function diurnal variation and nocturnal MAP are related to decreased glomerular filtration rate at the time of examination.

Keywords Circadian rhythms . Chronic kidney disease . Blood pressure . Natriuresis . Proteinuria . Urinary flow rate

Introduction Both urinary flow and urinary excretion of protein and sodium ions exhibit circadian rhythms similar to that of blood pressure (BP), being higher in the active phase (day time) than in the resting phase (night time) [14]. Recent interest in circadian renal functions has been stimulated by the fact that deviation from their usual variation may indicate pathological relevance [57]. Deterioration of renal function in adults has been associated with a loss of nocturnal BP depression and a non-dipping pattern of diuresis, proteinuria and natriuresis [5, 8, 9]. However, broad variation in the subjects ages in the abovementioned studies could have confounded the analyses, because older subjects tend to have higher night/day ratios of urinary protein and sodium ion excretion rates than younger subjects [1012]. Children diagnosed with chronic glomerulopathy and who are not taking medication that could confound tubular function analysis appear to be suitable human models to clarify the effect of kidney injury on the circadian patterns of diuresis, proteinuria and natriuresis. The aim of this study was to assess urine flow rate and urinary excretion rates of protein, creatinine and sodium ions during the day and the night and the prevalence of nocturnal polyuria in children with chronic kidney disease due to glomerulopathy. In addition, the study was designed to test the hypothesis that a nondipping pattern of diurnal variations in renal functions is related to the glomerular filtration rate at the time of examination.

A. Peco-Anti (*) : D. Krui : D. Paripovi University Childrens Hospital, 11000 Belgrade, Serbia e-mail: amira@udk.bg.ac.yu J. Marinkovi Institute for Medical Statistics and Informatics, School of Medicine, Belgrade, Serbia

1166

Pediatr Nephrol (2009) 24:11651172

Patients and methods Fifty-six patients with chronic glomerulopathy (36 girls and 20 boys) aged 419 years (median age 12.0 years) with a mean body mass index of 19.50.5 kg/m2 were studied during hospitalisation. All had chronic kidney disease (CKD) classified as stage 1 to stage 5 according to the Kidney Disease Outcomes Quality Initiative (K/DOQI) CKD criteria [13]. Their histological diagnoses were as follows: diffuse mesangial proliferation (23.2%), focal segmental glomerular sclerosis (16.1%), minimal change disease (16.1%), lupus nephritis (14.3%), immunoglobulin A (IgA) nephropathy (12.5%), microscopic polyangiitis (7.1%), thin basement membranes (5.4%), HenochSchnlein purpura nephritis (1.8%), haemolytic uraemic syndrome (1.8%) and Alport syndrome (1.8%). Patients with massive oedema, those taking diuretics or those that had received renal replacement therapy (RRT) were not included, and nor were patients with nocturnal enuresis. Almost half of the patients (48.2%) did not receive any medication. Other patients (41.1%) were taking stable doses of immunosuppressive therapy [corticosteroids and/or cyclosporin (CSA), either mycophenolic acid or cyclophosphamide] with (nine patients) or without an angiotensin-converting enzyme inhibitor (ACEI) with or without angiotensin-receptor blocker (ARB). The remaining patients (10.7%) were receiving only antihypertensive therapy (ACEI and/or beta or calcium ion blockers). The study was conducted at the Nephrology Department of the hospital. Forty-nine patients were examined within 3 days prior to kidney biopsy; five patients were studied during a 2-day hospitalisation visit for evaluation of renal function, and an additional two patients were investigated within 1 month prior to their starting chronic haemodialysis. All patients adhered to a no salt-supplemented diet for at least 4 weeks and were allowed to perform normal daily activity and consume regular water intake. Blood and urine samples were collected during the first 24 h, and 24 h ambulatory blood pressure monitoring (ABPM) was performed during the second 24 h. On the day prior to and during the study period none of the patients were given fluids intravenously or were prevented from enteral intake. Investigations started at 06:00 and ended at the same time the next day. The study imposed a daily regime of getting up at 06:00 and going to sleep at 22:00. During the daytime the patients were not confined to bed, rather they were encouraged to walk around as much as possible. The day time urine collection started after the first morning urination. The night time collection included urine obtained if the subjects got out of bed in the middle of the night and the first morning urination the following day. BP was monitored every 15 min during the day (06:00 to 22:00) and every 30 min during the night (22:00 to 06:00) by an automatic oscillometric device (model 90207, SpaceLabs). An addi-

tional study entry criterion was the availability of good quality readings from ABPM. Samples for urinary protein (UPRT), urinary creatinine (UCr), urinary sodium ions (UNa) and urinary osmolality (UOsm) were collected during both periods. Blood samples for the determination of serum creatinine levels were collected once daily in the early morning under fasting conditions. Urine volume (UV) was measured, providing a day volume (dUV) and a night volume (nUV) and was expressed as hourly excretion per body surface area (millilitres per square metre per hour). UPRT was determined with pyrogallol red and molybdate on a Selecta bioanalyser, while plasma and UCr were determined by Jaffes method on a Dimension autoanalyser (Dade Behring) and UNa was measured by ion selective analysis on an AVL 9180 electrolyte analyser (Roche). The calculation of excretion rate (excretion rate = concentration volume/ collection time) required accurate measurement of day and night urine volumes. This was facilitated by motivated parents in combination with health care personnel. To facilitate comparison between subjects of varying size we normalised the results to 1 m2 of body surface area. In other words, UCr and UPRT were expressed as milligrammes per square metre per hour, UNa as millimoles per square metre per hour and UPRT/UCr were calculated as the ratios of UPRT and UCr and expressed in milligrammes per milligramme. UOsm [milliosmoles per kilogramme water (H2O)] was calculated as 2 (sodium + potassium) + glucose + urea. Sodium ions (millimoles per litre), potassium ions (millimoles per litre), glucose (millimoles per litre) and urea (millimoles per litre) were measured in both day and night urine. Sodium ion excretion was also expressed as fractional sodium ion excretion (FeNa) over 24 h, for the day time and the night time periods. Kidney function was assessed by measurement of the clearance of endogenous creatinine (CrCl) over 24 h, day time and night time periods. The percentage of nocturnal change in the investigated parameters was calculated as follows [(day time average night time average)/day time average] 100%. Nocturnal polyuria was defined as night UV 35% of total daily UV. The diurnal variation in mean blood pressure (MAP), UV, UPRT, UCr, UNa and UOsm, as well as the occurrence of nocturnal polyuria, were compared within and between the following groups of patients: CKD1 patients (group I), CKD25 patients (group II), patients receiving ACEI/ARB/ CSA therapy (group III), or those not receiving such therapy (group IV). The whole study was approved by the University Childrens Hospital Ethics Committee. The parents of the patients signed informed consent documents. Statistical analysis The results are expressed as mean standard error of the mean (SEM). Comparisons of normally distributed variables

Pediatr Nephrol (2009) 24:11651172

1167

within and between groups were performed with an independent or paired t-test. For groups with variables not normally distributed, non-parametric MannWhitney U and Wilcoxon W matched-pairs tests were performed. The significance of differences in gender distribution, as well as the distribution of the patients nocturnal polyuria, between groups I and II was examined by the 2 test. Significant day night differences (nocturnal reduction) for each parameter in groups IIV, in addition to the presence of their interactions, was tested by repeated measures two-way analysis of variance (ANOVA). Correlation analysis was performed with Pearsons or Spearman's correlation coefficients, depending on data distribution within groups. We used regression analysis to model the effect of the investigated parameters on the night time sample to predict the diurnal value and to examine the effect of urine flow rate during day and night time on the corresponding urinary osmolality and urinary excretion rate of protein, creatinine and sodium ions. All regression analyses were adjusted in groups according to the CKD stage (I and II) and ACEI/ARB/CSA therapy (III and IV). The minimum level of significance was set when P < 0.05. All statistical analyses were performed with SPSS 15.0 for Windows (SPSS, Chicago, IL, USA).

CKD25 patients had higher nocturnal excretion rates of UV, UPRT, UCr, UNa and UPRT/UCr than did CKD1 patients. The night/day ratios of all the parameters, except UOsm, were significantly increased in CKD25 patients. Day time UV, UCr and UOsm were comparable in both groups (Table 2). A significant difference in the measured CrCl (P <0.0001) and FeNa (P <0.0001) between the day and night time periods was found in CKD1 patients and for both periods when CKD1 and CKD25 patients were compared (Table 2). There were no significant differences in all the variables, except for heart rate, when subjects receiving ACEI ARB/CSA were compared with those without such medication. A strong relationship between day time and night time UV, UPRT, UNa, UOsm and MAP was revealed in all patients by regression analyses adjusted according to CKD stage and therapy (ACEI/ARB/CSA) (Table 4). Therefore, by using the observed corresponding coefficient (b) and the regression equation (y = a+ bx) we were able to calculate (predict) day time values. In contrast, night time and day time UCr were not directly related. Nocturnal polyuria Nocturnal polyuria was much more widespread in CKD25 patients than in CKD1 patients (91.7% and 9.1%, respectively; P <0.0001). In addition, the night time/24 h ratio of UV was higher in CKD24 patients than in CKD1 patients (P <0.0001). Day time UV correlated significantly with UOsm (r= 0.51; P <0.01) and UNa (r=0.27; P <0.05). Night time UV correlated significantly with UCr (r=0.29; P =0.038), UPRT (r=0.54; P <0.0001), UNa (r=0.54; P < 0.0001) and UOsm (r= 0.65; P <0.0001). However, when the data were adjusted for groups according to CKD stage (groups I and II) and to therapy (groups III and IV), a significant relationship was found to exist only between UV and UNa during the day (P =0.021) and the night (P = 0.004) as well as between UV and UOsm during the day and the night (P <0.0001) (Table 5). Clearly, day time diuresis and, particularly, night time diuresis are dependent on natriuresis and urinary osmolality. Associations between creatinine clearance and parameters of renal function Correlations between CrCl and all the other parameters investigated were examined. Only significant correlations are presented in Table 6. CrCl showed positive correlation with day time UCr, nocturnal change of UV and the night/ day ratio of urine osmolality. Significant negative correlation with night time MAP, UV, UPRT, UNa and UOsm, day and night time FeNa and day time UPRT, as well as the night/day ratios of MAP, UV, UPRT and UNa, were noted.

Results Patients characteristics According to the K/DOQI classification [13], 44 patients had CKD1, with normal creatinine clearance (CrCl) (131 3.6 ml/min per 1.73 m2 body surface area). The remaining 12 patients had CKD25, with CrCl of 44.67.7 ml/min per 1.73 m2. Age, gender, body mass index, the percentage of patients receiving ACEIs and/or ARBs with or without CSA (ACEI/ARB/CSA), 24 h UV and 24 h UCr were similar in both groups. However, arterial hypertension and nocturnal polyuria were more widespread in the CKD25 group (P <0.01). In addition, proteinuria was higher in CKD25 subjects (P <0.05). Further patients characteristics are given in Table 1. Excretion rate of urine flow, urinary protein, creatinine and sodium ions during day and night time Day and night time values of the parameters investigated are given in Table 2, and their night/day ratios in Table 3. A highly significant nocturnal decline in UV, UPRT, UNa, as well as a nocturnal increase in UOsm, was found only in CKD1 patients (P <0.005). In contrast, in CKD25 patients, UPRT and UOsm did not change during the night, while the excretion rates of UV, UCr and UNa showed reversed diurnal rhythm (being higher at night than during the day).

1168 Table 1 Patients characteristics. Values are mean SEM or number (%) of patients. NS not significant Characteristic CKD1 (n =44) Mean SEM 11.490.64 15/29 144.783.33 19.560.54 116.023.56 1522.38187.73 52.505.89 1.700.19 0.240.02 decrease 51.405.69 608.99209.23 18.345.84 0.940.33 775.2759.86 18.240.72 24.401.18 3.410.26 82.031.11 89.191.90 12 (27.27%) 4 (9.09%) 11 (25%) 17 (38.64%) 44 (100%)

Pediatr Nephrol (2009) 24:11651172

CKD25 (n =12) Mean SEM 12.251.21 5/7 144.466.05 19.391.20 44.567.74 1441.25194.66 48.895.56 1.610.21 0.470.03 increase 46.3012.81 3555.371026.35 60.9129.21 5.063.54 722.47115.34 16.481.41 21.992.27 4.340.62 92.183.65 87.913.93 9 (75%) 11 (91.67%) 4 (33.33) 6 (50%) 0 4 6 0 2 (0%) (33.33%) (50%) (0%) (16.67)

Age (years) Male/female Height (cm) Body mass index (kg/m2) CrCl (ml/min per 1.73 m2) UV (ml/24 h) (ml/m2 per hour) (ml/kg per hour) Night/24 h ratio of UV Nocturnal change in UV (%) Proteinuria (mg/24 h) (mg/m2 per hour) Proteinuria/urinary creatinine (mg/mg) UCr (mg/24 h) (mg/kg per 24 h) (mg/m2 per hour) UNa (mmol/m2 per hour) MAP (mmHg) Heart rate (beats/min) Patients with arterial hypertension (%) Patients with nocturnal polyuria (%) Patients receiving ACEIs and/or ARBs CSA (%) Patients receiving immunosuppressive therapy (%) Patients with CKD stage (%) 1 2 3 4 5

NS NS NS NS <0.0001 NS NS NS <0.0001 <0.0001 0.003 0.021 0.004 NS NS NS NS 0.001 NS 0.005 <0.0001 NS NS

Limitations of the study Firstly, a limitation of the study was the small number of patients diagnosed with chronic renal failure. This was due to the low number of children with nonterminal renal failure with glomerulopathy (according to the local health registry, 13.4% with glomerular disease compared with 60% patients with congenital anomalies of the kidney and urinary tract (CAKUT) [14]) and our strict exclusion criteria. Secondly, the assessment of 24 h ABPM during hospitalisation for kidney biopsy was not optimal, although, during the day, the patients were not confined to bed, rather they were encouraged to move around as much as possible. Even though BP dipping was not interpreted in this study, our results should be confirmed by 24 h ABPM outside of the hospital setting. Thirdly, some of the patients in both groups were taking ACEIs and/or ARBs with or without CSA. Such medication could have influenced natriuresis, BP and CrCl [15,

16]. Possible confounding effects of such medication were excluded by the non-significant difference in all the renal parameters between the subjects taking ACEI/ARB/CSA and those not taking the medication and by adjusted regression analysis in both patient groups (either taking medication or not). Finally, as our study was primarily cross-sectional, a prospective study would be required to examine the values of abnormal renal diurnal variations as markers of progressive renal disease.

Discussion Diurnal variation in BP has been intensively studied since the advent of 24 h ABPM. A normal BP circadian rhythm is characterised by a nocturnal dip of more than 10% of its day time value and is a well-established prognostic marker for the functional integrity of the cardiovascular, renal and autonomic nervous systems [1721]. In contrast, circadian

Pediatr Nephrol (2009) 24:11651172

1169

Table 2 Day Day time time and and night night time time values valuesof ofmean meanblood bloodpressure, pressure,heart heart rate, clearance, urinary flow rate, urine osmolality and urinary Table 2 (a) creatinine I, II (CKD 1:CKD25) and (b) III, IV (with or without ACEI/ARB/ :CKD ) and (b) III, IV (with or without excretion rates of creatinine, protein and sodium ions in groups (a) I, II (CKD 1 2 5 rate, creatinine clearance, urinary flow rate, urine osmolality and CSA medication). Values are means SEMs. NS notACEI/ARB/CSA significant medication). Values are SEMs. NSand not sodium significant urinary excretion rates ofmeans creatinine, protein ions in groups Parameters Group I Day (a) MAP (mm Hg) Heart rate (beats/min) CrCl (ml/min per 1.73 m2) UV (ml/m2per hour) UCr (mg/m2 per hour) UPRT (mg/m2 per hour) UPRT/UCr (mg/mg) UNa (mmol/m2 per hour) FeNa (%) UOsm (mosmol/kg H2O) (b) Parameters Night P Group II Day Night P Comparisons between groups I and II Day time Night time P value P value 0.003 NS <0.0001 <0.0001 NS <0.0001 0.002 <0.0001 <0.0001 0.036

84.491.17 76.841.14 93.431.95 79.222.02 130.165.18 99.245.28 68.818.00 29.111.85 25.557.14 0.920.25 4.760.41 0.790.07 356.50 29.65 Group III Day Night 80.152.76 79.851.82 78.439.74 28.363.87 19.851.25 13.104.26 0.730.24 2.020.27 0.440.06 469.58 35.84

0.002 94.273.98 <0.0001 92.003.46 <0.0001 47.278.11

88.093.16 79.365.03 47.368.95

0.001 NS <0.0001 NS NS <0.0001 0.006 0.031 NS NS 0.017 NS NS NS 0.004 0.008 0.003 NS <0.0001 NS

<0.0001 41.904.53 59.927.97 <0.0001 19.821.76 25.353.04 <0.0001 100.3243.54 107.77 47.71 <0.0001 5.463.56 5.343.79 <0.0001 4.170.51 5.470.70 <0.0001 4.231.93 6.113.20 <0.0001 302.7528.09 312.08 35.43 Group IV P Day Night 79.141.51 79.032.54 88.646.98

MAP (mm Hg) Heart rate (beats/min) CrCl (ml/min per 1.73 m2) UV (ml/m2 per hour) UCr (mg/m2 per hour) UPRT (mg/m2 per hour) UPRT/UCr (mg/mg) UNa (mmol/m2 per hour) FeNa (%) UOsm (mosmol/kg H2O)

89.382.98 98.082.59 100.55 12.46 76.8518.34 24.311.91 37.5913.18 1.480.45 3.640.45 1.080.24 331.86 52.62

<0.0001 91.262.03 <0.0001 91.262.03 NS 112.668.44 58.235.97 28.292.08 43.7215.54 1.951.04 4.920.39

Comparisons between groups III and IV P Day time P Night time P value value <0.0001 NS NS <0.0001 <0.05 NS 0.003 NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS

41.598.42 0.023 21.392.17 NS 28.5010.74 0.012 1.140.31 2.500.60 0.780.29 403.57 56.92 NS NS

33.264.44 0.001 20.731.49 0.007 35.7315.86 0.002 1.951.21 3.000.40 1.981.01 449.30 37.76 NS 0.001 0.018 0.002

<0.0001 1.700.61 0.048 347.368 26.31

rhythms of kidney function and their relevance to clinical medicine are not as well understood. The transition from day to night is normally associated with a pronounced decline in the glomerular filtration rate (GFR) and urinary excretion of protein, electrolytes and other end products of metabolism [4, 22, 23]. Therefore, day time and nocturnal urine samples are generally not comparable. Kidney biomarker excretion is widely used as a clinical tool for screening or follow-up in kidney disease. Examples of such biomarkers include albumin, -glutamyl transpeptidase and, to some extent, N-acetyl-D-glucosaminidase, all of which show higher levels in day time samples than in overnight urine samples [4]. Higher excretion rates are directly linked to effects on

urinary flow rates [4]. Other factors, such as being in the upright position, exercise and diet, might have additional effects on the higher day time excretion rate. To date, there is no complete explanation for the origins of circadian renal functions. In contrast to the circadian rhythm of arterial BP that is largely dependent on external factors [24, 25], the circadian GFR rhythm appears to be more dependent on endogenous factors. For example, during continuous bed rest, circadian BP variation is attenuated, while the GFR rhythm hardly changes [2]. Buijsen and co-workers demonstrated that the GFR circadian rhythm in renal transplant patients occurred independently of autonomic regulation [26]. Furthermore, GFR circadian variations in healthy subjects are not directly

1170

Pediatr Nephrol (2009) 24:11651172

Table 3 Night/day ratio of mean blood pressure and urinary excretion rates of protein, creatinine, sodium ions and osmolality in groups I, II (CKD1:CKD25) and III, IV (with or without ACEI/ARB/CSA medication). NS not significant Parameter Group I (n =44) Mean SEM ratio ratio ratio ratio ratio ratio ratio ratio of of of of of of of of MAP UV UCr CrCl UPRT PRT/Cr (mg/mg) UNa UOsm 0.910.01 0.470.06 0.590.06 0.790.06 0.780.06 0.830.10 0.590.09 1.420.09 Group II (n =12) Mean SEM 0.940.02 1.460.13 1.040.09 1.160.09 1.080.13 0.920.06 1.380.09 1.040.08 I vs II P NS 0.000 0.001 0.003 0.024 NS 0.000 0.019 Group III (n =15) Mean SEM 0.920.11 0.690.10 0.810.09 0.880.08 0.880.09 0.890.05 0.760.15 1.260.01 Group IV (n =41) Mean SEM 0.920.01 0.690.10 0.660.07 0.890.07 0.840.07 0.830.11 0.780.11 1.350.10 III vs IV P NS NS NS NS NS NS NS NS

Night/day Night/day Night/day Night/day Night/day Night/day Night/day Night/day

related to fluctuations in systemic BP and cardiac output, but they can be at least partly explained by fluctuations in hormonal factors such as plasma atrial natriuretic peptide and plasma renin activity [3]. It is plausible that the circadian rhythms of renal function have multifactorial origins that will be fully understood in the coming years. From the clinical point of view it is very important for one to examine whether a non-dipping pattern of renal rhythms, analogous to a non-dipping BP rhythm, could be a prognostic marker of progressive renal disease. To date, this has barely been studied, particularly in children. Relatively small studies have mainly examined the relationship between day and night urinary sodium ion excretion and systemic BP [9, 11, 12]. Although there is no complete consensus, increased nocturnal sodium ion excretion appears to be associated with a non-dipping BP rhythm [2729]. On the other hand, a nondipping circadian BP rhythm has been suggested to be a consequence of renal dysfunction rather than its cause [30]. Renal dysfunction, even in a mild form, as seen in healthy kidney donors, is capable of causing a non-dipping BP phenomenon without affecting its absolute levels [31]. Furthermore, renal dysfunction is followed not only by a

disturbed circadian BP rhythm, but also by nocturnal poly nocturia, increased nocturnal natriuresis and proteinuria [8]. The aforementioned studies were conducted on small numbers of adults. None of the studies included any children. Our study revealed that decreased creatinine clearance in children with chronic glomerulopathy is associated with disturbed diurnal variations in diuresis, natriuresis and proteinuria. In contrast, children with normal creatinine clearance and chronic glomerular disease had normal diurnal variations in these parameters. Bearing in mind that 24 h ABPM was performed under non-optimal conditions (during hospitalisation) and that antihypertensive therapy was administered to some of the patients, we decided not to emphasise the interpretation of BP dipping in our patients. However, nocturnal polyuria was highly and significantly widespread in CKD25 patients compared with CKD1 patients, despite the fact that their 24 h UVs and daytime UVs were not different. Our findings confirmed the notion that nocturnal polyuria was associated with increased nocturnal natriuresis. Fukuda and colleagues suggested that, in chronic kidney disease, polynocturia is related to natriuresis rather than to water diuresis [9].

Table 4 Regression analysis of night and day time urine flow rate, urinary creatinine, protein and sodium ion excretion rate, night and day time urine osmolality and MAP all adjusted according to groups I, II (CKD1:CKD25) and III, IV (with or without ACEI/ARB/CSA medication) Dependent variable Independent variables Variables Constant b P 95% Confidence interval Lower bound UV day time (ml/m2 per hour) UCr day time (mg/m2 per hour) UPRT day time (mg/m2 per hour) UNa day time (mmol/m2 per hour) UOsm day time (mosmol/kg H2O per square metre per hour) MAP day time (mmHg) UV night time (ml/m2 per hour) UCr night time (mg/m2 per hour) UPRT night time (mg/m2 per hour) UNa night time (mmol/m2 per hour) UOsm night time (mosmol/kg H2O per square metre per hour) MAP night time (mmHg) 98.47 38.39 29.18 5.91 14.07 13.50 1.12 0.04 0.96 0.37 0.61 0.93 <0.0001 0.820 <0.0001 0.034 <0.0001 <0.0001 0.73 0.32 0.86 0.03 0.45 0.74 Upper bound 1.55 0.40 1.05 0.71 0.77 1.12

Pediatr Nephrol (2009) 24:11651172

1171

Table 5 Regression analysis of creatinine, protein, sodium ion urinary excretion rates and urine osmolality on urinary flow rate during day and night, adjusted according to groups I, II (CKD1:CKD25) and III, IV (with or without ACEI/ARB/CSA medication) Dependent variable Independent variables Variables Constant B P 95% Confidence interval Lower bound UV day time (ml/m2 per hour) UV night time (ml/m2 per hour) UNa day time (mmol/m2 per hour) UOsm day time (mosmol/kg H2O per square metre per hour) UCr night time (mg/m2 per hour) UPRT night time (mg/m2 per hour) UNa night time (mmol/m2per hour) UOsm night time (mosmol/kg H2O per square metre per hour) 58.37 157.47 13.33 1.60 4.98 46.9 6.00 0.15 0.65 0.08 5.52 0.08 0.021 <0.0001 0.140 0.085 0.004 <0.0001 0.95 0.22 0.22 0.01 1.88 0.11 Upper bound 11.06 0.09 1.52 0.16 9.15 0.05

Increased natriuresis was also found in nocturnal polyuria in the elderly [32], in adults with essential hypertension [33], and in children with monosymptomatic nocturnal enuresis refractory to desmopressin [34]. It has been proposed that such conditions share the same pathophysiological determinants, including decreased production of nitric oxide and/or increased prostaglandin synthesis [32, 34]. In agreement with Fukuda and colleagues [9], we also found nocturnal mean arterial pressure and night/day ratios of diuresis, proteinuria, and natriuresis to be negatively correlated with creatinine clearance. These findings are compatible with our proposal that a non-dipping pattern of diurnal variations in renal functions is related to the glomerular filtration rate at the time of examination. The standard method for quantitative determination of urinary flow rate, proteinuria, urinary creatinine and sodium ion excretion is to measure the parameters in 24 h urine. However, it is not easy to collect all the urine from children for 24 h. An alternative and easier method to determine these urinary markers is to measure them in night time urine and then predict their day time values (shown in Table 4). The regression equations from this study look promising, but they need to be validated in a separate new cohort of patients. Furthermore, nocturnal urine is dissimilar between healthy children and children diagnosed with impaired renal function. In the former the urinary flow rate is lower during night and excretory rates of protein, sodium ions and creatinine are also lower. In contrast, children with chronic renal failure have increased rates of nocturnal urinary flow and enhanced excretion rates of sodium ions, protein and creatinine. Bearing in mind that nocturnal urine samples are often preferred over 24 h urine samples, the use of creatinine-corrected excretion rates of urine markers appears to be a good choice when they are measured in nocturnal urine collected from children with chronic kidney disease due to glomerulopathy, as well as from healthy subjects [4].

Conclusion Our results support the hypothesis that a non-dipping pattern of diurnal renal functions in children with chronic glomerulopathy is related to decreased creatinine clearance, as children diagnosed with chronic renal failure, unlike those with normal creatinine clearance, exhibited a nondipping pattern of renal functions. Moreover, the degree of renal dysfunction measured by creatinine clearance indicated positive association with a reversed rhythm of diuresis,
Table 6 Significant correlations between creatinine clearance and renal function parameters (N/D night/day ratio) CrCl 24 h (ml/min per 1.73 m2 body surface area) Day time CrCl (ml/min per 1.73 m2) Day time UCr (mg/min per square metre per hour) Day time UPRT (mg/min per square metre per hour) Day time FeNa (%) Night time CrCl (ml/min per 1.73 m2) Night time MAP (mmHg) Night time UV (ml/min per square metre per hour) Night time UPRT (mg/min per square metre per hour) Night time UNa (mmol/min per square metre per hour) Night time FeNa (%) Night time UOsm (mosmol/kg H2O) N/D MAP N/D UCr N/D UNa N/D UOsm UV change during night (%) Pearson correlation coefficient (r) 0.89 0.34 0.37 0.59 0.84 0.39 0.46 0.45 0.48 0.57 0.65 0.31 0.39 0.32 0.45 0.65 P

<0.0001 0.017 0.008 <0.0001 <0.0001 0.009 0.001 0.001 0.001 <0.0001 <0.0001 0.042 0.005 0.029 0.002 <0.0001

1172

Pediatr Nephrol (2009) 24:11651172 16. Hansen JM, Foght-Andersen N, Christensen NJ, Strandgaard S (1997) Cyclosporine-induced hypertension and decline in renal function in healthy volunteers. J Hypertens 15:319326 17. Ohkubo T, Hozawa A, Yamagauchi J, Kikuya M, Ohmori K, Michima M, Matsubara M, Hashimoto J, Hoshi H, Araki T, Tsuji I, Satoh H, Hisamichi S, Imai Y (2002) Prognostic significance of the nocturnal decline in blood pressure in individuals with and without high 24-blood pressure: the Ohasama study. J Hypertens 20:21832189 18. Pickering TG (1990) The clinical significance of diurnal blood pressure variations. Dippers and non dippers. Circulation 81:700 702 19. Verdecchia P, Schillaci G, Guerrieri M, Gatteschi C, Benemio G, Boldrini F, Porcellati C (1990) Circadian blood pressure changes and left ventricular hypertrophy in essential hypertension. Circulation 81:528536 20. Carvalho MJ, van den Meiracker AH, Boomsma F, Lima M, Freitas J, Veld AJ, Falco de Freitas A (2000) Diurnal blood pressure variation in progressive autonomic failure. Hypertension 35:892897 21. Portualuppi F, Montanari L, Ferlini M, Gilli P (1990) Altered circadian rhythms of blood pressure and heart rate in nonhemodialysis chronic renal failure. Chronobiol Int 7:321327 22. Agarwal R (2007) Relationship between circadian blood pressure variation and circadian protein excretion in CKD. Am J Physiol Renal 293:F655F659 23. Kamperis K, Hansen MN, Hagstroem S, Hvistendahl G, Djurhuus JC, Rittig S (2004) The circadian rhythm of urine production, and urinary vasopressin and prostaglandin E2 excretion in healthy children. J Urol 171:25712575 24. Floras JS, Jones JV, Johnston JA, Brooks DE, Hassan MO, Sleight P (1978) Arousal and the circadian rhythm of blood pressure. Clin Sci Mol Med Suppl 4:S395S397 25. Sundberg S, Kohvakka A, Gordin A (1988) Rapid reversal of circadian blood pressure rhythm in shift workers. J Hypertens 6:393396 26. Buijsen JG, van Acker BA, Koomen GC, Koopman MG, Arisz L (1994) Circadian rhythm of glomerular filtration rate in patients after kidney transplantation. Nephrol Dial Transplant 9:13301333 27. Uzu T, Kazembe FS, Ishikawa K, Nakamura S, Takashi I, Kimura G (1996) High sodium sensitivity implicates nocturnal hypertension in essential hypertension. Hypertension 28:139142 28. Uzu T, Ishikawa K, Fujii T, Nakamura S, Inenaga T, Kimura G (1997) Sodium restriction shifts circadian rhythm of blood pressure from nondipper to dipper in essential hypertension. Circulation 96:18591862 29. Uzu T, Kimura G (1999) Diuretics shift circadian rhythm of blood pressure from nondipper to dipper in essential hypertension. Circulation 100:16351638 30. Brenner BM, Garcia DL, Anderson S (1988) Glomeruli and blood pressure. Less of one, more the other. Am J Hypertens 1:335347 31. Goto N, Uchida K, Morozumi K, Ueki T, Matsuoka S, Katayama A, Haba T, Tominaga Y, Fukuda M, Nakao A, Kimura G (2005) Circadian blood pressure rhythms is disturbed by nephrectomy. Hypertens Res 28:301306 32. Mc Keigue PM, Reynard JM (2000) Relation of nocturnal polyuria of the elderly to essential hypertension. Lancet 355:436488 33. Bulpitt CJ, Dollery CT, Corne S (1976) Change in symptoms of hypertensive patients after referral to hospital clinic. Br Heart J 38:121128 34. Kamperis K, Rittig S, Jorgensen KA, Djurhus JC (2006) Nocturnal polyuria in monosymptomatic nocturnal enuresis refractory to desmopressin treatment. Am J Physiol Renal Physiol 291:F1232F1240

natriuresis and proteinuria. Therefore, we propose that renal dysfunction is an important determinant of abnormal renal rhythms and that enhanced nocturnal natriuresis, creatinuria and proteinuria could represent biomarkers of progressive glomerular disease.
Acknowledgement The study was supported by the Ministry of Science and Environmental Protection, Government of Serbia, grant no. 145028G.

References
1. Wesson LG (1964) Electrolyte excretion in relation to diurnal cycles of renal function. Medicine (Baltimore) 43:547592 2. Koopman MG, Koomen GCM, Krediet RT, de Moor EA, Hoek FJ, Arisz L (1989) Circadian rhythm of glomerular filtration rate in normal individuals. Clin Sci 77:105111 3. Voogel AJ, Koopman MG, Hart AAM, van Montfrans GA, Arisz L (2001) Circadian rhythms in systemic hemodynamics and renal function in healthy subjects and patients with nephrotic syndrome. Kidney Int 59:18731880 4. Trachtenberg F, Barregard L, McKinlay S (2008) The influence of urinary flow rate in children on excretion of markers used for assessment of renal damage:albumin, -glutamyl transpeptidase, N-acetyl--D-glucosaminidase, and alpha1-microglobulin. Pediatr Nephrol 23:445456 5. Hiller P, Knapp MS, Cove-Smith R (1980) Circadian variations in urine excretion in chronic renal failure. Q J Med 49:461478 6. Bianchi S, Bigazzi R, Baldari G, Sgherri G, Campese VM (1994) Diurnal variations of blood pressure and microalbuminuria in essential hypertension. Am J Hypertens 7:2329 7. Hansen HP, Hovind P, Jensen BR, Parving HH (2002) Diurnal variations of glomerular filtration rate and albuminuria in diabetic nephropathy. Kidney Int 61:163168 8. Fukuda M, Munemura M, Usami T, Nakao N, Takeuchi O, Kamiya Y, Yoshida A, Kimura G (2004) Nocturnal blood pressure is elevated with natriuresis and proteinuria as renal function deteriorates in nephropathy. Kidney Int 65:621625 9. Fukuda M, Masahiro M, Miyagi S, Sengo K, Takeshi U, Yoshida A, Kimura G (2006) Polynicturia in chronic kidney disease is related to natriuresis rather than to water diuresis. Nephrol Dial Transplant 21:21722177 10. Kirkland JL, Lye M, Levy DW, Banerjee AK (1983) Patterns of urinary flow and electrolyte excretion in healthy elderly people. BMJ 287:16651667 11. Sachdeva A, Weder AB (2006) Nocturnal sodium excretion, blood pressure dipping, and sodium sensitivity. Hypertension 48:527533 12. Staessen JA, Birkenhager W, Bulpitt CJ, Fagard R, Fletcher AE, Lijnen P, Thijs L, Amery A (1993) The relationship between blood pressure and sodium and potassium excretion during the day and night. J Hypertens 11:443447 13. National Kidney Foundation (2002) K/DOQI clinical practice guideline for chronic kidney disease: evaluation, classification, and stratification. Am J Kidney Dis 39(2 Suppl 1):S1S266 14. Peco-Anti A, Bogdanovi R, Muli B, SEPNWG (2007) Serbian pediatric chronic kidney disease registrySPEKID. BANTAO J 5:3236 15. Grnhagen-Riska C (1986) Effects of ACE inhibition on regulation of salt and water. Acta Med Scand Suppl 707:9599