Abstract
Background
Peritoneal dialysis (PD) solutions sterilized at a low pH level contain only minimal amounts of glucose degradation products (GDPs). The latter are known to have an adverse effect on the peritoneal membrane. The present study retrospectively analyzes the effects on the incidences of peritonitis and exit-site infections.
Materials and Methods
Data concerning the frequency of peritonitis and exit-site infections for 120 patients treated with PD were collected. Before 2000, 67 of these patients received conventional dialysates and from 2000 on, 53 patients were treated with the new dialysis fluids. Furthermore, a correlation between the incidence of infections and the duration of treatment with dialysis was established.
Results
It was observed that the use of dialysis solutions low in GDPs resulted in significantly lower rates of peritonitis (p = 0.002) and exit-site infections (p = 0.02). When using the new treatment, peritonitis occurred, on average, after 48 months of treatment and exit-site infections after 34 months of treatment.
Discussion
The result supports the hypothesis that the use of the new, biocompatible, PD solutions contributes to considerable reduction in the rates of peritonitis and exit-site infections. As it is not expected that randomized prospective studies will be conducted in the future, further observational studies should be carried out in order to affirm the observed tendencies.
Materials and Methods
We retrospectively studied 120 patients that were on continuous ambulatory peritoneal dialysis (CAPD) in the renal outpatient clinic of Heidelberg, Germany, between 1990 and 2005. The switch to biocompatible dialysis solutions low in GDPs took place in the year 2000. Of the 120 patients, 67 were treated with the conventional dialysis solutions and 53 received the new dialysates. All patients, including those using conventional dialysates, were already using the double-bag system with the Y-set. During the observation period, neither medical nor nursing standards concerning treatment changed.
Incidences of peritonitis and catheter exit-site infections were assessed by evaluating documented evidence. Diagnosis of peritonitis was based on a significant increase in leukocytes in the dialysis fluid in conjunction with the typical clinical symptoms — with or without a positive test for pathogenic cultures. In order to verify whether the two patient groups to be compared were homogenous, the data obtained for each patient also included such parameters as age at initiation of treatment, sex, presence of diabetes mellitus, and original renal disease.
The Wilcoxon test was used for the statistical evaluation and for analyzing the homogeneity of the two treatment groups. Incidence was calculated and compared with the help of the Fisher test and the chi-square test respectively. A p value < 0.05 was taken to be significant.
Results
Table 1 shows the demographic data and the distribution of the original renal diseases. Table 2 compares both systems and shows the number of exit-site infections per patient, without taking into consideration the duration of CAPD treatment. No statistically significant difference was observed, but when the infection rates per month were compared between the two groups, the first group showed an infection rate of 0.04 per month, while the infection rate of the second group was 0.03 (p = 0.02). Thus, in the group that had been treated with conventional dialysis fluids, an exit-site infection occurred every 23.0 months; patients in the group treated with dialysates low in GDPs presented with an exit-site infection after 33.9 months of treatment.
Demographic Data and Renal Diseases
Exit-Site Infections per Patient
In the group receiving conventional dialysates, exit-site infections were predominantly caused by Staphylococcus aureus (32%), with nonpathogenic corynebacteria (17%) in second place. In the group receiving dialysates provided in multi-compartment systems, the main cause of infection was S. epidermidis (62%), followed by coagulase-negative staphylococcus (21%).
Table 3 shows a comparison of the two treatments and the number of cases of peritonitis. In contrast to exit-site infections, the incidence of peritonitis showed a significant difference (p = 0.006). When taking into consideration the respective durations of treatment, the peritonitis infection rate in patients receiving conventional dialysates was 0.05 per month (this corresponds to 1 case of peritonitis every 20.0 months), and 0.02 (which corresponds to 1 case of peritonitis every 47.6 months) (p = 0.002) in patients treated with dialysates low in GDPs. When conventional dialysis fluids were used, peritonitis was predominantly caused by coagulase-negative staphylococcus (15%) and S. epidermidis (13%). When using dialysates provided in multi-compartment systems, peritonitis was most frequently caused by S. epidermidis (36%), followed by S. aureus (14%) in second place. The number of cases of peritonitis not related to the presence of pathogenic cultures did not differ significantly (8 vs 9). Three cases of peritonitis could be traced back to exit-site infections that had spread. Of these infections, two occurred in patients that had been treated with conventional dialysis fluids (related to S. aureus in both cases) and one occurred in a patient who had been receiving the newer generation of dialysis fluids (related to coagulase-negative staphylococcus).
Peritonitis per Patient
Discussion
The results show that the use of the new dialysis fluids resulted in a significant reduction in the incidence of peritonitis. The calculated peritonitis-free interval of 20 months for conventional treatment systems corresponds well to the values frequently stated by other scientific sources (9). In contrast, when using the new dialysis solutions, peritonitis occurred every 47.6 months. This is probably due to the fact that dialysis solutions low in GDPs lead to fewer alterations in the peritoneal membrane, and thus to a less impaired functioning of the local immune system. This hypothesis is based on in vitro experiments investigating the functioning of peritoneal cells (2,3,6,10) and on animal testing (4,5). Further explanations were given by Alscher et al. (11), who found an improved phagocyte function of the peritoneum, and by Szeto et al. (12), who reported that the use of biocompatible solutions resulted in lower levels of markers of infection.
While both treatment groups showed a homogenous distribution with respect to sex and original renal disease, there were significant differences concerning age, and with reference to diabetes mellitus there were trend differences. Patients treated after the year 2000 were, on average, 5 years older (49 vs 44 years). The percentage of diabetics also increased, from 3% to 11%, but remained low in numbers. One reason is, among other factors, an increased incidence of renal replacement therapy in elderly patients and diabetics. Even more astonishing in this context is the reduced infection rate found in the group treated after 2000. In addition, more and more people arrive at the conclusion that diabetes mellitus per se, in spite of the instillation of glucose-containing fluids, is not a contraindication for peritoneal dialysis treatment. On the one hand, when taking into consideration that diabetics account for more than 50% of patients suffering from end-stage renal insufficiency, the figure of 11% for diabetic patients is still surprisingly low. It can be assumed that both factors (age and diabetes) are unlikely to have contributed to the better results in the second treatment group. There is no doubt that diabetes mellitus is a factor that considerably contributes to the risk of infections, so that even small differences in frequency in both groups might already have an impact on the overall result.
Catheter exit-site infections were also drastically reduced. Based on the presented calculations, the time span between infections rose from 23.0 months for patients treated with the conventional system to 33.9 months for those receiving the new treatment. Although the result is statistically significant, it is less spectacular than the difference in the incidence of peritonitis. Obviously, other factors, such as the presence of pathogenic organisms in other areas of the body, observing hygiene, and correct handling also play an important role in the development of a catheter exit-site infection.
With respect to the conclusions that can be drawn from this study, it must be said that the fact that it was a retrospective analysis, and that there were no control groups, was a limiting factor. In addition, due to the low number of cases, one should exercise caution when drawing conclusions from these results. Furthermore, it cannot be ruled out that the increase in experience gained at the dialysis center and by the persons treating patients also had a positive impact on the second patient group. Furthermore, it needs to be mentioned that these results contradict the publications of Lee et al. (13), whose findings do not indicate that the use of biocompatible dialysis solutions leads to a reduction of the peritonitis rate. In contrast to these findings, Montenegro et al. (14) observed that peritonitis-free intervals in patients treated with conventional dialysates increased from 21 months to 36 months when dialysates low in GDPs were used, and thus reported results comparable to those of the present study.
More studies are necessary in order to back up the described observations that point toward a reduction of the infection rate. Probably this will also allow for the gathering of data about other effects of dialysis fluids low in GDPs. This includes examining possible effects on residual diuresis, on the maintenance of a good peritoneal membrane transport status, on the quality of dialysis (Kt/V), and finally on mortality rates.