Abstract
Background
The factors that predict baseline peritoneal permeability remain largely unknown. We noticed that patients that adhered to a strict low protein diet (LPD) during the predialysis period seldom showed high peritoneal permeability on the peritoneal equilibration test (PET) at the introduction of peritoneal dialysis (PD). Therefore, we investigated whether a strict LPD during the predialysis period affects peritoneal permeability.
Method
We retrospectively analyzed 37 patients that started PD in a single Japanese center. Patients were divided into group A and group B by the median amount of daily protein intake (PI) during the predialysis period using urine collected over 24 hours.
Results
There were no differences between groups A and B in age, gender, proportion of diabetic nephropathy, blood pressure, body mass index, or body surface area. There were also no differences between the groups in laboratory findings, including hematocrit, serum albumin, and serum creatinine. The PETs showed a significantly lower dialysate-to-plasma ratio of creatinine at 4 hours (Cr D/P) for group A than for group B (p = 0.02). Furthermore, a significant positive correlation between Cr D/P and PI was observed (r = 0.53, p < 0.01).
Conclusion
It is suggested that a strict LPD during the predialysis period may suppress peritoneal permeability at induction of PD.
On the other hand, it has been shown that a low protein diet (LPD) during the predialysis period of chronic kidney disease (CKD) not only improves uremic symptoms and delays the introduction of dialysis, but also prevents progression of CKD (11).
We noticed that patients that had adhered to a strict LPD during the predialysis period of CKD seldom showed high peritoneal permeability in PETs at the time of induction of PD (Figure 1). The mean and median amounts of daily protein intake (PI) in these patients were extremely low: 0.51 g/kg/day and 0.46 g/kg/day respectively. Comparison with the data of Mujais and Vonesh (12) revealed that the percentage of patients showing high peritoneal permeability in PETs at the beginning of PD was significantly lower for these patients. Therefore, we investigated whether a strict LPD during the predialysis period of CKD could affect peritoneal permeability at the time of induction of PD.
Comparison of peritoneal equilibration test (PET) status distribution (Showa University Fujigaoka Hospital vs Mujais&Vonesh). Patients in the present study showed a significantly lower ratio of high peritoneal permeability in PETs at the induction of peritoneal dialysis (PD) than that of Mujais and Vonesh by chi-square test (p < 0.01) (12).
Patients and Methods
Patient Selection and Study Design
We retrospectively analyzed 37 patients selected from 45 consecutive patients that started PD in Showa University Fujigaoka Hospital from 1 April 1998 through 31 March 2003. These selected patients were all advised continuously on a LPD during the predialysis period of CKD (mean and median periods of hospital assessment were 2.87 years and 1.32 years respectively). Their PI was determined using Maroni's formula for urea nitrogen appearance in urine collected over 24 hours (13). Their PI data were measured persistently during the predialysis period and the mean and median amounts of PI were calculated from more than three measured values during 1 year before the start of PD. PET data were analyzed as an indicator of peritoneal permeability. Following Twardowski's method (14), PET was performed within 1 month after PD catheterization and the ratios of dialysate-to-plasma creatinine (Cr D/P) at 4 hours and the ratios of dialysate glucose at 4 hours to the initial dialysate glucose (Glu D/D0) were calculated. The PET category was assigned on the basis of the results of Cr D/P. The 8 of 45 patients without PI data during the predialysis period, the main exposure to be tested in this investigation, were excluded from the analysis.
The 37 remaining patients were divided into group A and group B by median PI (0.46 g/kg/day): group A patients were below the median and group B patients were above the median. We then compared patient characteristics, laboratory findings, and PET data between the two groups. Correlations between PI and Cr D/P, as well as Glu D/D0, were also investigated.
Statistical Analysis
All data are expressed as mean ± standard deviation (SD). Comparisons between the two groups were made using the appropriate t-test after analysis of variance by F test. Ratios were compared between the two groups using the chi-square test. Linear regression analysis was performed for correlation between two variables. Statistical analysis was performed using Stata/SE 9.2 (Stata Corp., College Station, TX, USA). A p value less than 0.05 was considered significant.
Results
Characteristics and Laboratory Findings of the Patients
Table 1 shows the characteristics of the analyzed patients at the time of induction of PD. Mean age was 46.8 ± 15.0 years; 62.2% were males. Diabetic nephropathy was the cause of end-stage renal disease in 3 of the 37 patients.
Patient Characteristics and Laboratory Findings at Induction of Peritoneal Dialysis
PI = daily protein intake determined by urea nitrogen appearance in 24-hour urine.
Group A versus group B, where group A was comprised of patients below the median PI of 0.46 g/kg/day and group B patients were above the median PI.
Values expressed mean±SD.
There were no significant differences between groups A and B with respect to age, gender, proportion of diabetic nephropathy, systolic/diastolic blood pressure, body mass index, or body surface area. The PI was 0.37 ± 0.05 g/kg/day for group A and 0.64 ± 0.13 g/kg/day for group B; the difference was significant (p < 0.01).
Laboratory findings are also shown in Table 1. Serum creatinine (Cr) was 14.4 ± 3.9 mg/dL for group A and 12.6 ± 4.3 mg/dL for group B; the difference was not significant. Neither serum albumin nor hematocrit differed significantly between groups A and B: 3.8 ± 0.4 vs 4.0 ± 0.5 g/dL and 23.2% ± 3.9% vs 24.9% ± 3.6% respectively.
Result of Pet
The comparison of the two groups with respect to PET data is shown in Figure 2. For group A, Cr D/P was 0.52 ± 0.08 and for group B it was 0.62 ± 0.14; it was significantly lower in group A than in group B (p = 0.02). For group A, Glu D/D0 was 0.51 ± 0.17 and for group B it was 0.43 ± 0.12: Glu D/D0 tended to be higher in group A than in group B (p = 0.16).
Comparison of peritoneal equilibration test (PET) results (group A vs group B). PET revealed significantly lower dialysate-to-plasma ratio of creatinine at 4 hours by non-paired t-test in group A than in group B (p = 0.02). Group A was comprised of patients below the median daily protein intake of 0.46 g/kg/day and group B patients were above.
Comparison of the PET status distribution between the two groups based on Cr D/P is shown in Figure 3. In group A, high and high-average groups were not recognized; the ratio of PET status distribution was significantly different between group A and group B (p = 0.03).
Comparison of peritoneal equilibration test (PET) status distribution (group A vs group B). Group A showed a significantly lower ratio of high peritoneal permeability in their PETs based on dialysate-to-plasma ratio of creatinine at the induction of peritoneal dialysis than that of group B by chi-square test (p = 0.03). Group A was comprised of patients below the median daily protein intake of 0.46 g/kg/day and group B patients were above.
Correlation between Pi and Results of the Pet
Correlation between PI and Cr D/P is shown in Figure 4. In linear regression analysis, the correlation coefficient r was 0.53 between PI and Cr D/P and they showed a significant positive correlation (p < 0.01). Correlation between PI and Glu D/D0 is shown in Figure 5. In linear regression analysis, r was -0.35 between PI and Glu D/D0. Albeit weak, there was also a significant negative correlation between these variables (p = 0.04).
Correlation between daily protein intake (PI) and dialysate-to-plasma ratio of creatinine (Cr D/P). A significant positive correlation between PI and Cr D/P was observed (p < 0.01).
Correlation between daily protein intake (PI) and glucose D/D0. A significant negative correlation between PI and Glu D/D0 was observed (p = 0.04).
Discussion
More than a quarter of a century has passed since Moncrief and Popovich et al. (15) first reported the clinical application of PD in the current style. Since then, innovations and improvements in PD catheters, peritoneal dialysate, and dialysis equipment have been made every year and clinical outcome has improved. Now PD plays an important role in renal replacement therapy for end-stage renal disease patients.
The CANUSA study (5) was a large-scale study of the prognosis of PD patients. In that study it was recognized that the higher the peritoneal permeability, the poorer the prognosis. According to the 2-year survival rate with respect to PET status, prognosis was better in the order low > low average > high average > high. Technical survival was also significantly shorter for the patients that showed high peritoneal permeability. These results indicate that high peritoneal permeability is a risk factor for poor prognosis. Since that study, reports concerning poor prognosis for patients with high peritoneal permeability have accumulated (1,2). There is also a report that showed high peritoneal permeability to be an independent risk factor for mortality (16); however, its mechanism has not yet been clarified (17). On the other hand, it has been reported that high peritoneal permeability is not disadvantageous clinically. Rather, it has been suggested that poor prognosis is correlated more significantly with concomitant complications such as chronic inflammation, malnutrition, and excess body fluid (18,19).
Meanwhile, it has been proposed that a LPD during the predialysis period of CKD improves uremic symptoms, delays the induction of dialysis, and suppresses progression of renal failure (11). We also enforced a strict LPD (0.39 g/kg/day), without supplementing amino acids and keto analogs, on very late-stage CKD patients (Cr ≥ 10 mg/dL) and reported that a strict LPD can not only prevent deterioration in renal function and delay induction of dialysis, but can also maintain a reasonable nutritional state (20). However, it is necessary to consider that there may be a risk of such a strict LPD leading to protein-energy malnutrition at the start of dialysis. This kind of malnutrition may be a consequence of uremia itself or related to comorbid conditions (21) and is a strong risk factor for PD patient survival (5,22).
We noticed that the patients that maintained a strict LPD during the predialysis period of CKD seldom showed high permeability in their PET at the time of induction of PD. Therefore, we carried out this study to examine the effects of a LPD during the predialysis period of CKD on peritoneal permeability at the induction of PD.
Consequently, we observed a significant positive correlation between PI and Cr D/P, and a significant negative correlation between PI and Glu D/D0. We also compared Cr D/P at the time of PD induction between two groups based on median amount of PI determined during the predialysis period (0.46 g/kg/day). Creatinine D/P at baseline was significantly lower in the group with less than the median amount of PI during the predialysis period. However, there were no differences between the groups in patient characteristics or laboratory findings. These results suggest that a strict LPD may suppress peritoneal permeability at the time of PD induction. An increase in blood flow in the peritoneal microcirculation and contributions of cytokines such as transforming growth factor-beta have been suggested as possible causes of increased peritoneal permeability (23). Future clinical and experimental studies are expected to further clarify these mechanisms.
It is necessary to consider selection bias in a singlecenter study such as the current investigation. Indeed, the proportion of diabetic nephropathy was very low in this study and adherence to a strict LPD generally presented some difficulty. Therefore, we have to recognize that our results may suffer from selection bias to some extent. We also have to consider that daily PI in this study was estimated by 24-hour urine urea excretion rather than by food records. Accordingly, if the patient was in an anabolic or a catabolic state, PI would be underestimated or overestimated respectively. We should also note that this investigation was observational and not a randomized controlled trial. It is the nature of an observational study that the association being recognized may suffer from confounding factors not measured. In this study, our findings may have been confounded by adherence/compliance issues in medication or diet therapy, which could have had an impact on other areas of the patient's medical care. Additional limitations of this study are its retrospective design and small sample size. However, our data provided some preliminary results on a possible factor that may suppress baseline peritoneal permeability and could improve outcomes in PD patients.
In conclusion, it is suggested that continuation of an adequate low protein diet during the predialysis period of CKD, based on strict protein restriction, may suppress peritoneal permeability at the time of PD induction.
Footnotes
This study was supported in part by grants for PD research from Baxter Japan Healthcare Corporation. There were no restrictions on publication nor conflicts of interest.