Peritoneal Dialysis International: Journal of the International Society for Peritoneal Dialysis

A peritoneal biopsy registry was established to examine morphological and functional changes to the peritoneum during peritoneal dialysis (PD). During the early stages of this study, it became clear that surgical trauma to the peritoneum at the time of biopsy could cause a variety of changes to the surface. We examined the effects of surgical trauma in a rat biopsy model.

Rat peritoneum was subjected to a variety of traumas that might occur at biopsy and compared with peritoneal biopsies that had been collected, using the suture method described here, from PD patients. Changes in the quality of non-PD biopsies taken before and after the development of the suture technique were evaluated.

In the rat model, external massaging of the peritoneum induced moderate loss of microvilli. Brief light touching caused distortion of the mesothelial surface. Pressing resulted in mesothelial denudation and thin strands of presumed cellular remains. Rubbing caused complete loss of mesothelial cells and their basement membrane. Air drying caused progressive loss of microvilli and eventual cellular distortion. Comparison with peritoneal biopsies from PD patients revealed similarities with certain types of trauma, namely, air drying and pressing. Collection of peritoneal biopsies using the suture method significantly improved specimen quality compared with specimens taken before its introduction (p < 0.025%).

These results illustrate the sensitivity of the mesothelium to mechanical trauma, the possibility of confusing trauma with genuine pathology, and, hence, the necessity of employing a trauma-free method of biopsy collection, such as the technique described here.

To determine if a diet complemented with calcium caseinate is better than a natural high protein diet for increasing serum albumin levels in patients on continuous ambulatory peritoneal dialysis (CAPD).

A 4-month clinical trial involving 100 patients older than 18 years was performed. Patients were randomized into two groups: group A, high protein diet (1.4 g natural protein/kg target weight/day and 35 kcal/kg target weight/day); and group B, calcium caseinate (0.7 g calcium caseinate plus 0.7 g natural protein diet/kg target weight/day and 35 kcal/kg target weight/day). Serum levels of albumin, total proteins (TP), BUN, creatinine, glucose, urea, sodium, and potassium, and hematocrit, leukocytes, erythrocytes, and hemoglobin were analyzed at baseline and every 30 days.

The final mean albumin value was, for group A, 3.04 ± 0.39 g/dL, and for group B, 3.1±2 0.41 g/dL (p < 0.05); TP for group A, 6.29 ± 0.47 g/dL, and for group B, 6.49 ± 0.51 g/dL (p < 0.05); leukocytes for group A, 688 ± 8 1282/mm³, for group B, 7288 ± 1878/mm³ (p = 0.05); BUN for group A, 47 ± 11 mg/dL, for group B, 50 ± 16 mg/dL (p = 0.05). Regression analysis showed a treatment effect in serum albumin and TP levels from the third month in both groups. In group B, a constant elevation of serum albumin of 0.19 mg/dL and TP of 0.27 mg/dL was observed in every month of treatment with calcium caseinate. In the regression analysis of group A we observed a smaller increase in serum albumin, 0.06 mg/dL, and in TP, 0.11 mg/dL, in each month of treatment with the high protein diet. Both differences are significant (p < 0.05).

Calcium caseinate used in CAPD patients suffering from malnutrition increases serum albumin levels.

The most widely used peritoneal function test, the peritoneal equilibration test (PET), is performed with a 2.27% glucose solution. Recently, the International Society for Peritoneal Dialysis committee on ultrafiltration failure (UFF) advised performing the test with 3.86% glucose solution because it is more sensitive for detecting clinically significant UFF. Because no reference values for this test were available, we analyzed the results of standard peritoneal permeability analyses (SPAs) using 3.86% glucose.

The tests were performed in our center on 154 clinically stable peritoneal dialysis (PD) patients that were free of peritonitis for at least 4 weeks. For the assessment of reference values, we used two approaches. In approach A, patients with UFF, defined as net ultrafiltration (UF)< 400 mL/4 hours, were excluded. In approach B, only patients within their first 2 years of PD treatment were included, regardless of net UF. Means and 95% confidence intervals (95% CI) were calculated for the transport parameters of the PET and SPA.

Means of normal distribution with 95% CI in approach A were as follows: for 2.0-L exchanges, mass transfer area coefficient (MTAC) for creatinine 8.8 mL/minute (4.7 – 12.7 mL/min), dialysate/plasma ratio (D/P) creatinine 0.70 (0.52 – 0.88), glucose absorption 58% (44% – 72%), dialysate₂₄₀/initial dialysate ratio of glucose (D _t /D₀) 0.28 (0.18 – 0.38), net UF 675 mL (375 – 975 mL), and maximal dip in D/P sodium after correction for diffusion from the circulation 0.110 (0.050 – 0.164); for 1.5-L exchanges, MTAC creatinine 7.4 mL/min (3.8 – 11.0 mL/min), D/P creatinine 0.69 (0.52 – 0.86), glucose absorption 62% (52% – 72%), D _t /D₀ glucose 0.25 (0.17 – 0.32), net UF 551 mL (430 – 670 mL), and maximal dip D/P sodium 0.120 (0.048 – 0.166). In approach B, most of the transport values were similar; however, values for lymphatic absorption were significantly higher [1.52 mL/min (2-L) and 1.40 mL/min (1.5-L), p < 0.01] and values for the maximum dip in D/P sodium were lower [0.101 (2-L) and 0.112 (1.5-L), p > 0.05]. This was probably the result of including patients with UFF in approach B, since these parameters can be causative factors of UFF.

A peritoneal transport function test using 3.86% glucose provides data on various aspects of transport. This study gives normal reference values that can be used for analysis of causes of UFF.

Despite improvements in peritoneal dialysis (PD) technique, peritonitis continues to be one of the most frequent complications of PD. Nonresolving peritonitis remains a risk for severe anatomical peritoneal changes that may limit the viability of the membrane for dialysis purposes. We have observed remarkably poor outcome of peritonitis caused by Escherichia coli in the past 6 years. With its very low response rate to broad-spectrum antibiotics, the increased severity of E. coli peritonitis deteriorates peritoneal function and affects patient outcome.

Retrospective study.

Two large PD units in two university hospitals.

The total number of patients reviewed was 456. The records of 49 E. coli peritonitis episodes were studied. The observation period started in 1980 and ended in March 2001. Sixteen males and 19 females were included. Severity was defined in terms of days of peritoneal inflammation, lack of response to a potentially useful antibiotic, requirement for catheter removal, and/or laparotomy. Study cases (study group) were those episodes appearing after 1996 (when the first severe cases appeared) and historic controls were episodes occurring before1996.

In the study group, 18 peritonitis episodes developed in 15 patients. In the control group, 31 peritonitis episodes developed in 20 patients. There were no significant differences in clinical presentation; however, the outcome was significantly poorer for the later period. A severe outcome occurred in 50% of study versus 10% of control patients. In fact, 68% of the episodes registered before 1996 were cured in 3 days or less. Concurring with this trend, the numbers of surgical interventions and catheter removals were also higher in the study group. Strikingly, E. coli did not show changes in in vitro susceptibility testing to antibiotics, although the in vivo response was much worse.

We describe a change in the virulence of E. coli peritonitis episodes over the past 5 years leading to a high percentage of treatment failure, which does not depend on antibiotic sensitivity and seems to be dependent on changes in host response mechanisms.

To examine gram-negative exit-site infection and peritonitis rates before and after the implementation of Staphylococcus aureus prophylaxis in peritoneal dialysis (PD) patients.

Prospective data collection with periodic implementation of protocols to decrease infection rates in two PD programs.

663 incident patients on PD.

Implementation of S. aureus prophylaxis, beginning in 1990.

Rates of S. aureus, gram-negative, and Pseudomonas aeruginosa exit-site infections and peritonitis.

Staphylococcus aureus exit-site infection and peritonitis rates fluctuated without significant trends during the first decade (without prophylaxis), then began to decline during the 1990s subsequent to implementation of prophylaxis, reaching levels of 0.02/year at risk and zero in the year 2000. Gram-negative infections fell toward the end of the 1980s, due probably to the implementation of better connectology. However, there have been no significant changes for the past 6 years. There was little change in P. aeruginosa infections over the entire time period. Pseudomonas aeruginosa is now the most common cause of catheter infection and catheter-related peritonitis.

Prophylaxis against S. aureus is highly effective in reducing the rate of S. aureus infections but has no effect on gram-negative infections. Pseudomonas aeruginosa is now the most serious cause of catheter-related peritonitis.

To determine the incidence and significance of peritoneal eosinophilia (PEo) during peritoneal dialysis (PD)-related peritonitis.

Retrospective observational study.

Tertiary-care public hospital.

We performed a cytological study of dialysate at the start of 465 cases of peritonitis diagnosed between January 1987 and May 2002. Cases associated with PEo (> 10% eosinophils) were classified according to their infectious or seemingly noninfectious origin. We compared the two groups, trying to disclose differentiating patterns of presentation.

We found PEo in 42 cases. Infectious peritonitis was the final diagnosis in 22 of the 42 cases; a diagnosis of idiopathic eosinophilic peritonitis was finally established in 20 cases. The etiologic spectrum of infectious peritonitis with PEo did not differ markedly from the global spectrum of peritonitis in our unit. Infectious peritonitis with PEo tended to appear later in the course of PD therapy, presented with more severe clinical symptoms, displayed higher total peritoneal leukocyte and neutrophil counts, and showed lower degrees of PEo than idiopathic eosinophilic peritonitis, but overlap between the groups was significant.

Peritoneal eosinophilia is infrequent but not rare during infectious PD-related peritonitis. Our findings agree with established concepts on idiopathic eosinophilic peritonitis, but overlap in presentation with infectious eosinophilic peritonitis is significant, which should be taken into consideration at the time of planning therapy for this condition.

The International Society for Peritoneal Dialysis (ISPD) guidelines recommend empiric therapy with cefazolin and ceftazidime for peritoneal dialysis (PD)-related peritonitis. Empiric cefazolin therapy may have diminishing efficacy because of emerging methicillin resistance in gram-positive bacteria (GPB). Western Australia also has large numbers of Aboriginal and isolated regional patients, where giving these antimicrobials can be impractical.

To evaluate, based on local antimicrobial resistance patterns, the feasibility of following ISPD guidelines in Western Australia and to identify any subgroups of PD peritonitis patients that may benefit from alternative empiric intraperitoneal antibiotics (e.g., vancomycin).

Retrospective study of all PD peritonitis episodes in Western Australia from 1 February 2000 to 31 January 2001.

Three adult tertiary referral university hospitals and their PD patients in metropolitan Perth and regional Western Australia.

All adults on PD in Western Australia.

Isolates and antibiograms were analyzed versus patient characteristics, including race and patient demographics.

293 patients (28% Aborigines, 32% regional patients) received PD. 145 episodes of PD peritonitis occurred during the study. The overall PD peritonitis rate was 1 episode/16 patient months, with Aborigines having 1 episode/10.5 patient months versus non-Aborigines having 1 episode/17 patient monthsp (< 0.001). 36% of isolates from PD peritonitis episodes were resistant to cefazolin or ceftazidime. 22% were methicillin-resistant GPB (MR-GPB) [18% coagulase-negative staphylococci (CoNS), 1.6% MR Staphylococcus aureus]; 2.5% were multidrug-resistant gram-negative bacteria (MDR-GNB); 5.7% were polymicrobial (MR-GPB and/or MDR-GNB); and 5.7% were fungal. 63% of CoNS were methicillin resistant. Non-Aboriginal patients yielded MR-GPB in 22% of isolates versus 23% in Aborigines (p = 0.9). Six of seven cases of fungal peritonitis occurred inAboriginal patients (p < 0.001).

In our study population the ISPD guidelines were appropriate for 64% of patients with PD peritonitis. We could not identify specific patient subgroups where empiric cefazolin use could be more effective. High proportions of MR-GPB PD peritonitis episodes, along with local factors, make empiric cefazolin unsuitable for many regional PD patients in Western Australia.

To determine the impact of dialysate flow rate (DFR) on cefazolin pharmacokinetics (PK) in peritoneal dialysis (PD) patients.

A meta-analysis of published reports, identified by MEDLINE search (1966-2002) and other sources, containing information on cefazolin PK data in PD patients was conducted. Data were analyzed based upon low DFR (≤ 5.50 mL/minute) or high DFR (> 5.50 mL/minute). Data available were from North American (NA) (n = 45) and Singaporean (n = 10) patients. Complete data sets were available for 33 patients (CDS patients). Data were analyzed with respect to data origin and data set completeness: all patients (ALL), NA, and CDS. Analysis of log-transformed cefazolin PK data was performed to determine coefficient of determination (r²) between DFR and cefazolin elimination rate constant (k_el), clearance total (Cl_T), and clearance peritoneal (Cl_PD). Clearance total data were extrapolated to DFR observed in continuous flow PD.

Published literature provided data on 55 PD patients (12 high DFR, 43 low DFR). Regardless of data origin (ALL, NA, or CDS), a prominent coefficient of determination (p < 0.0001) existed between DFR and all cefazolin PK data except Cl_PD. The p value for DFR correlation to Cl_PD was 0.953, 0.011, and 0.036 for ALL, NA, and CDS patients, respectively. Cefazolin Cl_T and Cl_PD increased at higher DFRs.

These findings demonstrate that an increased DFR leads to an increased rate of cefazolin clearance in NA PD patients. The impact of Asian descent on cefazolin Cl_PD warrants further investigation. Clinicians dosing cefazolin in PD patients using a higher DFR than that used to determine cefazolin PK should use increased doses or prescribe lower/comparable DFRs. Data are not yet available for patients prescribed very high DFRs (e.g., continuous flow PD); extrapolation of our results demonstrates significant influences on clearance and risk for underdosing.

Concerns regarding the impact of ultrafiltration failure on peritoneal dialysis and the effect of hypertonic glucose on the peritoneal membrane have lead to a search for alternative dialysates. Computer simulations based on the three-pore theory suggest that a combination of 1.36% glucose and 7.5% icodextrin (glucose polymer) offers an improved ultrafiltration profile. The aim of the present study was to investigate the ultrafiltration profile of this combination fluid.

Prospective open study comparing 1.36% glucose, 3.86% glucose, 7.5% icodextrin, and the combination fluid (1.36% glucose/7.5% icodextrin).

Sheffield Kidney Institute, Northern General Hospital, Sheffield, UK.

11 patients currently using peritoneal dialysis not previously exposed to icodextrin.

Intraperitoneal volume was measured using a radioisotope dilution method.

The combination fluid showed a biphasic ultrafiltration profile, with a steep initial increase in intraperitoneal volume, then a maintained plateau phase for the duration of the study dwell (7 hours). The final volume was greater than that with the 1.36% glucose dwell and the 7.5% icodextrin dwell. The fluid was well tolerated by the patients.

These findings are in keeping with computer simulations using the three-pore model. The combination fluid offers an improved ultrafiltration profile, with a final volume similar to 3.86% glucose, while avoiding exposing the peritoneal membrane to high glucose concentrations. It may have a role as a long dwell to optimize ultrafiltration and possibly prolong peritoneal dialysis technique survival.

The goal of this paper was to review the viability of peritoneal dialysis (PD) in patients with spina bifida and/or ventriculoperitoneal shunt (VPS).

Pediatric dialysis unit in a tertiary-care hospital.

The course and outcome in 9 children, 5 from the authors’ experience and 4 from reported experience, are analyzed.

One patient died of a cause unrelated to PD or VPS, 2 were transferred to hemodialysis because of recurrent peritonitis, 1 discontinued PD transiently, 2 were transplanted, and 3 continue on PD. Six of these 9 children had a functioning VPS, and none presented evidence of ventriculitis or VPS dysfunction, even though 4 had PD-related peritonitis. One child presented with a massive PD-related hydrothorax.

(1) Having a VPS is not an absolute contraindication to PD; the available data support the viability of PD in patients with spina bifida and/or a VPS. (2) If cerebrospinal fluid diversion is needed simultaneously or after starting PD, an extraperitoneal site should be a better choice than VPS. This should avoid the risk of intra- and postoperative infection in the PD catheter secondary to surgical intervention for VPS insertion. (3) Loss of peritoneal function is a potential late risk related to cerebrospinal fluid and PD. (4) Spina bifida patients on PD present specific diagnostic challenges due to overlapping symptoms (e.g., vomiting, abdominal tenderness, fever) secondary to PD- or VPS-related complications (e.g., peritonitis, visceral injury by devices) or primary disease (e.g., neurogenic bladder, pyelonephritis), with inherent risks of delaying adequate treatment. Cloudy peritoneal effluent is an early indication of peritonitis, although it is not specific. (5) Early evaluation by a pediatric surgeon and a neurosurgeon is required for effective management of complications and selection of more efficient individualized therapeutic alternatives. Prompt treatment of complications is crucial. A registry of children with spina bifida on PD and the accumulation of a large population followed up for longer periods will provide an objective assessment of their problems and management.

The aim of the study was to assess the influence of peritoneal membrane permeability on bone metabolism in dialyzed children.

24 children with end-stage renal failure and being treated with peritoneal dialysis (PD) were studied. The children were divided into two groups based on the results of a standard peritoneal equilibration test: group I, high peritoneal transport [ratio of dialysate glucose concentration at 4 hours to dialysate glucose concentration at 0 hours (D/D₀) < 0.26, dialysate-to-serum ratio of creatinine concentration at 4 hours (D/P) > 0.81], 10 children aged 9.9 ± 2.9 years; group II, other peritoneal transport types (D/D₀ > 0.26, D/P < 0.81), 14 children aged 11.4± 2.7 years. Serum levels of calcium (sCa), phosphorus (sP), protein, albumin, alkaline phosphatase (AP), and parathormone (PTH) were measured, and bone biopsies were performed in all children. Alfacalcidol and calcium carbonate doses were adjusted to sCa, sP, and PTH levels in all patients.

No statistically significant differences (NS) between the two groups were found in age, duration of PD, sCa, sP, AP, PTH, protein, or albumin levels. The mean alfacalcidol dose was 0.055 ± 0.057 μg/kg body weight/week in group I and 0.099 ± 0.065 μg/kg/week in group II (p = NS). In group I, the high peritoneal transport group, significantly lower osteoclast surface (OcS/BS) and bone formation rate (BFR/TV) were found compared with group II (3.1% ± 1.7% vs 4.6% ± 2.0%, and 483.5 ± 329.2 μm³/mm²/day vs 913.9 ± 558.3 μm³/mm²/day, respectively, p < 0.05). We also found significant positive correlation between D/D₀ and BFR/TV and OcS/BS (r = 0.45, p < 0.05).

Bone turnover in children treated with PD may depend on peritoneal permeability.

To evaluate the impact of the “flush before fill” technique on the frequency of peritonitis in children receiving automated peritoneal dialysis (APD).

Randomized prospective multicenter study.

Participating pediatric dialysis programs of the Pediatric Peritoneal Dialysis Study Consortium.

121 pediatric (< 21 years of age) patients that had received peritoneal dialysis for > 2 months and that were currently receiving APD were randomized to use (flush group) or non-use (no flush group) of the “flush before fill” option. 66 patients were followed for> 12 months.

Peritonitis rates.

Overall, patients enrolled in the flush group experienced a peritonitis rate of 1 infection every 16.8 patient months; patients in the no flush group experienced a rate of 1 infection every 12.6 patient months (p= 0.193). However, analysis by gender revealed the peritonitis rate of females in the flush group (1 infection every 44.7 patient months) to be significantly better than females in the no flush group (1 infection every 12.4 patient months) (p ≤ 0.01). There was no difference noted in the male patients.

The use of the “flush before fill” option in pediatric patients receiving APD is associated with a marked improvement in the peritonitis rate of female but not male patients. Further study is indicated to explain the gender differences.

To determine the feasibility of reinstitution of continuous ambulatory peritoneal dialysis (CAPD) in patients with malignant hepatic tumors after partial hepatectomy.

Retrospective analysis of 2 CAPD patients.

Dialysis unit of a university teaching hospital.

Two CAPD patients with malignant hepatic tumors who had undergone partial hepatectomy.

Serum biochemistry, Kt/V, peritoneal equilibration test (PET) results before and after hepatectomy.

One patient was able to resume CAPD 4 weeks after partial hepatectomy. The other patient was successfully resumed on CAPD after resting the peritoneum for 3 months following partial hepatectomy. The serum biochemistry, Kt/V, and PET results of the 2 patients did not change significantly before and after partial hepatectomy.

Reinstitution of CAPD after partial hepatectomy in patients with malignant hepatic tumors is feasible.