Residual Peritoneal Volume and Body Size in Children on Peritoneal Dialysis

Claus P. Schmitt,1 Rouven Dötschmann,1 Markus Daschner,1 Mirjam Zimmering,2 Christine Greiner,3 Michael Böswald,4 Günther Klaus,5 Franz Schaefer1 [Members of the Mid European Pediatric Peritoneal Dialysis Study Group (MEPPS)]

Residual peritoneal volume may play an important role in dialysis efficacy and abdominal compliance in patients on chronic peritoneal dialysis (CPD). In children on CPD, the relationship between residual peritoneal volume and different measures of body size, as well as the day-to-day variability of residual volume, have not been established. We therefore investigated, on two consecutive days, residual peritoneal volume in 25 children on CPD, using the dextran dilution technique. Residual volume was linearly correlated with body size. Residual volume was independent of body size when normalized to body surface area, but decreased with increasing body size when normalized to body weight (r = -0.62, p < 0.001). Mean residual volume was 79 ± 25 mL/m2, with an intra-individual day-to-day coefficient of variation of 21% ± 15%. Residual volume was not correlated with the duration of PD, frequency of peritonitis, or peritoneal permeability as estimated by D/P creatinine or D/D0 glucose.

In conclusion, residual peritoneal volume is constant across the pediatric age range when normalized to body surface area. It accounts for approximately 8% of the usual fill volume in patients on CPD. Residual volume is not a major confounder of the transport status estimation obtained by peritoneal equilibration test.

Key words

Residual volume, dextran indicator method, children


Department of Pediatrics,1 Ruperto-Carola University, Heidelberg; Charité Children's Hospital,2 Humboldt University, Berlin; St. Georg's Hospital,3 Leipzig; Department of Pediatrics,4 University of Erlangen, Erlangen; Department of Pediatrics,5 University of Marburg, Marburg, Germany.


Evidence from animal work and autopsy studies in humans suggests that peritoneal surface area is proportional to body surface area (1-3). Peritoneal transport modeling according to the three-pore theory has confirmed that the total peritoneal "pore area" is a linear function of body surface area (4). These findings provide a rationale to normalize the peritoneal dialysis (PD) fill volume in children for body surface area rather than for body weight. Weight normalization results in relatively lower effective fill volumes in young children. However, even if dwell volumes are normalized to body surface area, young infants show more rapid peritoneal solute equilibration and lower ultrafiltration rates than older children (5). One possible, partial explanation for this difference might be a relatively higher residual peritoneal volume in young children, resulting in a more marked initial rise in solute concentrations and dilution of glucose.

The relationship between residual peritoneal volume and body size in children has not been studied to date. Here, we present results of standard peritoneal permeability analyses (SPAs) performed in a cohort of pediatric patients. This recently introduced variant of the peritoneal equilibration test (PET) includes an assessment of residual peritoneal volume by the dextran dilution technique (6). We adopted the SPA protocol for use in children by standardizing the fill volume to 1000 mL/m2 body surface area. In the present study, residual peritoneal volume was estimated from pairs of SPAs performed on two consecutive days in 25 children on chronic peritoneal dialysis.



Twenty-five children on automated peritoneal dialysis (APD) participated in the study. The patients were treated in five pediatric dialysis units in Germany. Underlying diseases included hemolytic uremic syndrome (n = 4), renal (hypo-)dysplasia (n = 3), obstructive uropathy (n = 3), focal segmental glomerulosclerosis (n = 2), nephronophthisis (n = 2), autosomal recessive polycystic kidney disease (n = 2), prune belly syndrome (n = 2), Jeune syndrome (n = 2), membranoproliferative glomerulonephritis type II (n = 1), Alport syndrome (n = 1), Wegener's granulomatosis (n = 1), and unknown (n = 2). Children with severe pulmonary, cardiac, or liver disease were excluded from the study. The patients were aged 1.1 - 19.3 years (median: 11.7 years). Median duration of PD was 7.0 months (range: 1 - 53 months). Seven patients had experienced one or more episodes of peritonitis; in these subjects, antibiotic treatment had been discontinued at least 6 weeks prior to the study.

Screening for dextran antibodies was negative in all patients. No patient had been treated with dextran in the past.

Study protocol

After overnight APD, the patients underwent two standardized peritoneal permeability assessments (SPAs) on the mornings of two consecutive days. The tests were performed using 1.5% glucose solution. On one day, a 35 mmol/L lactate solution buffered at pH 5.5 [CAPD17 (Fresenius, Oberursel, Germany)] was used; on the other day, a 34 mmol/L bicarbonate solution buffered at pH 7.4 [BIC170 (Fresenius)] was used. To each bag, 1 g/L dextran 70 [Hyskon (Pharma Reusch, Bonn, Germany)] was added. To prevent possible anaphylactic reactions to dextran, 50 mg/kg dextran 1 [Promit (Pharma Reusch)] was given intravenously before instillation of the test solution.

The peritoneal cavity was rinsed with 500 mL/m2 body surface area of the dextran-containing test solution, which was immediately drained by gravity after inflow was complete. Then, 1000 mL/m2 body surface area of the same test solution was instilled and left to dwell for 4 hours. At 10, 20, 30, 60, 120, 180, and 240 minutes, 10% of the dwell volume was temporarily drained to avoid dead-space effects. The dialysate was subsequently re-infused after 5 mL of dialysate had been collected. After complete drainage at 240 minutes, the peritoneal cavity was rinsed with 500 mL/m2 body surface area of the test solution without dextran supplement.

That night, the patients were subjected to the same APD regimen as on the previous night. The next morning, a second SPA using the other PD solution was conducted. The sequence of PD fluid was randomly allocated.

Local ethical committee approval for the study protocol was obtained in each center.


Dextran concentrations were determined in the test bag at 10 minutes and at 240 minutes, and in the effluents of the two rinse solutions. Dextran 70 was measured by a previously published procedure (7). In brief, the samples were deproteinized with trichloroacetic acid and the dextran was pre-extracted using a Sephadex G-25 PD-10 column (Pharmacia LKB, Uppsala Sweden). Thereafter, high-performance liquid chromatography (HPLC) was performed using a Bio-Gel XL guard column (Bio-Rad Laboratories GmbH, Munich, Germany) coupled to a Waters 410 refraction index detector (Waters, Eschborn Germany).


Peritoneal residual volume was determined from the second rinsing procedure using the following equation:
Residual volume = VR × CR/C240

where VR is the volume of the second rinse solution, CR the concentration of dextran 70 in the rinse fluid, and C240 the concentration of dextran 70 in the SPA fluid at dwell time 240 minutes.

Data are described as mean ± standard deviation (SD). Associations between variables were evaluated by Pearson correlation analysis.


Figure 1: Relationship between body surface area and residual peritoneal volume expressed as an absolute value (upper panel), and normalized to body surface area (middle panel) or body weight (bottom panel).

As shown in Table I, residual volumes were estimated without systematic differences between the two tests. The mean intra-individual coefficient of variation was 21%; the test-retest correlation was r = 0.37 (p = 0.06). Absolute residual volume was positively correlated with age (r = 0.64, p < 0.001), weight (r = 0.70, p < 0.0001), height (r = 0.49, p < 0.05), and body surface area (r = 0.69, p < 0.0001) (Figure 1). When normalized to body surface area, residual volume was independent of age and any measure of body size. In contrast, a normalization of residual volume to body weight resulted in significant inverse relationships to age (r = -0.62, p < 0.001), weight (r = -0.60, p < 0.005), height (r = -0.51, p < 0.001), and body surface area (r = 0.62, p < 0.001) (Figure 1).

The mean residual volume normalized to body surface area was not correlated with the D/P creatinine ratio or the D/D0 glucose ratio at any time point of the SPA. D/D0 glucose ratio was, however, correlated with the number of previous peritonitis episodes at 20 minutes (r = -0.46, p < 0.05), at 30 minutes (r = -0.67, p < 0.01), and at 60 minutes (r = -0.41, p = 0.07).

No relationship was observed between residual peritoneal volume and the absolute number or annualized incidence of previous peritonitis episodes. Also, residual volume was not dependent on the total duration of PD at the time of the study.


Few studies of fluid and solute kinetics in children have included estimates of residual peritoneal volume. Using creatinine as an endogenous indicator substance, Warady et al observed a mean residual volume of 215 mL in a large pediatric population with a mean body surface area of 0.98 m2, similar to the cohort studied here (5). Another study using the same marker even estimated an average residual volume of 425 mL/m2 (8). The discrepancy between these findings and our much lower mean residual volume estimate of 80 mL/m2 confirms the results of a direct comparative study using endogenous and exogenous markers in adults (9). Residual volume tends to be grossly overestimated by endogenous indicator substances such as creatinine or urea, owing to immediate diffusive transport during the rinsing procedure. It is therefore essential to use exogenous markers such as dextran 70 when an accurate determination of residual volume is needed. Indeed, two pediatric fluid kinetic studies in children using dextran 70 and autologous hemoglobin yielded mean estimates of 3.2 mL/kg and 3.9 mL/kg, close to the 3.0 mL/kg observed here (10,11).

Our results indicate that residual peritoneal volume in children is a linear function of body size. Residual volume becomes size-independent when the estimate is normalized to body surface area. Hence, when residual volume assumptions must be made in daily clinical practice or in studies on fluid kinetics in children, it is advisable to use a value based on body surface area; 80 mL/m2 may be a good rule of thumb.

Under the controlled conditions of this study, residual peritoneal volume was relatively constant within individuals, with a test-retest coefficient of variation of approximately 20%. A previous study in adults did not find a similar degree of reproducibility (9). This discrepancy might be due to the shorter interval between studies in this trial (24 hours) as compared to the previous trial (up to 1 week), and to our meticulous attention to complete drainage of the abdomen. Of course, a higher day-to-day variability of residual volume must be expected in clinical practice, particularly in patients on automated PD where fixed drainage times are used.

Another frequently raised concern is the possible confounding effect of residual volume on apparent solute transport rates observed in peritoneal equilibration studies. Our analysis of the SPA data showed no significant association between residual peritoneal volume and D/P creatinine ratio or D/D0 glucose ratio even in the early phase of equilibration, suggesting that the small amount of residual volume does not affect apparent equilibration rates to a major degree when a fill volume of 1000 mL/m2 is used.


This study was supported by Fresenius Medical Care (Oberursel, Germany). The contribution of Dr. Judith Kirchgessner is gratefully acknowledged. The authors also thank Ms. Bärbel Philippin and Ms. Ruth Vierling for performing the high-performance liquid chromatography (HPLC) measurements.


  1. Wegner G. Chirurgische Bemerkungen über die Peritonealhöhle mit besonderer Berücksichtigung der Ovariotomie. Arch Klin Chir 1877; 20:51.
  2. Putiloff PV. Materials for the study of the laws of growth of the human body in relation to the surface areas of different systems; the trial on Russian subjects of planigraphic anatomy as a means for exact anthropometry — One of the problems of anthropology. Siberian Branch of Russian Geografic Society, 1884.
  3. Esperanca MJ, Collins DL. Peritoneal dialysis efficiency in relation to body weight. J Pediatr Surg 1966; 1:162-9.
  4. Schaefer F, Haraldsson B, Haas S, Simkova E, Feber J, Mehls O. Estimation of peritoneal mass transport by three-pore model in children. Kidney Int 1998; 54:1372-9.
  5. Warady BA, Alexander SR, Hossli S, et al. Peritoneal membrane transport function in children receiving long-term dialysis. Perit Dial Int 1996; 7:2385-91.
  6. Pannekeet MM, Imholz AL, Struijk DG, et al. The standard peritoneal permeability analysis: A tool for the assessment of peritoneal permeability characteristics in CAPD patients. Kidney Int 1995; 48:866-75.
  7. Koomen GCM, Krediet RT, Leegwater ACJ, Struijk DG, Arisz L, Hoek FJ. A fast reliable method for the measurement of intraperitoneal dextran 70, used to calculate lymphatic absorption. In: Khanna R, Nolph KD, Prowant BF, Twardowski ZJ, Oreopoulos DG, eds. Advances in peritoneal dialysis. Toronto: Peritoneal Dialysis Bulletin, 1991; 7:10-14.
  8. Fukuda M, Kawamura K, Okawa T, Kawahara K, Kamiyama Y, Honda M. The peritoneal equilibration test variables in pediatric CAPD patients. Acta Paediatr Jpn 1994; 36:57-61.
  9. Imholz ALT, Koomen GCM, Struijk DG, Arisz L, Krediet RT. Residual volume measurements in CAPD with exogenous and endogenous solutes. In: Khanna R, Nolph KD, Prowant BF, Twardowski ZJ, Oreopoulos DG, eds. Advances in peritoneal dialysis. Toronto: Peritoneal Dialysis Bulletin, 1992; 8:33-8.
  10. Reddingius RE, Schröder CH, Willems JL, et al. Measurement of peritoneal fluid handling in children on CAPD using autologous hemoglobin. Perit Dial Int 1994; 14:42-7.
  11. Reddingius RE, Schröder CH, Willems JL, et al. Measurement of peritoneal fluid handling in children on continuous ambulatory peritoneal dialysis using dextran 70. Nephrol Dial Transplant 1995; 10:866-70.

Corresponding author:

Franz Schaefer, md, Division of Pediatric Nephrology, University Children's Hospital, Im Neuenheimer Feld 150, Heidelberg D-69120 Germany.