This study demonstrates that a high proteinuria level is associated with malnutrition and rapid RRF or urine volume decline. A simultaneous decline in proteinuria level and RRF or urine volume appears to improve initial low GNRI. However, the RRF and urine volume decline led to decreased dialysis adequacy.
Proteinuria is a well-documented risk factor for the progression of chronic kidney disease in patients who are not treated with dialysis
. However, the specific effects of proteinuria level on RRF in patients on dialysis remain elusive. Jansen et al. showed that greater urinary protein loss is negatively associated with RRF
. However, using multivariate analysis, another study did not show that 24-hr proteinuria is associated with a faster RRF decline
. This study demonstrates that high proteinuria levels in patients on PD are associated with a rapid RRF decline. In this study, the RRF decline was 0.00 (−9.99 to 7.10) mL/min/1.73 m2 in the A group and −0.81 (−7.88 to 6.58) mL/min/1.73 m2 in the B group; in contrast, it was −2.41 (−23.41 to 2.31) mL/min/1.73 m2 in the C group. Although RRF decline occurred in the B and C groups, the RRF in the patients with high proteinuria levels tended to show a steeper decline. The multivariate regression analysis showed similar results.
Malnutrition is associated with mortality and morbidity in patients on dialysis
. Hypoalbuminemia or protein energy wasting in PD patients is related to increased catabolism, decreased protein synthesis, and external protein loss from peritoneum or urine
[18–20]. External protein loss can have an identical pathogenic effect on nephrotic syndrome
. The loss of peritoneal protein in PD patients averages 5 g/day
[21, 22]. One study demonstrated that peritoneal protein loss is related to mortality and malnutrition in patients on PD
. However, protein losses in the peritoneal effluent are usually compensated for by increased protein synthesis
. Balafa et al. showed that baseline peritoneal albumin and protein clearances are associated with comorbidities, but this is not a risk factor for decreased survival per se
. In the present study, the peritoneal albumin loss was less in the C group than in the A group at the time of PD initiation. High proteinuria levels are associated with hypoalbuminemia, which may be associated with decrease in peritoneal albumin loss. However, no significant difference was observed in peritoneal albumin loss among the 3 groups at the end of follow-up period. Compared with changes in proteinuria, those in peritoneal albumin loss were small (delta peritoneal albumin loss was −233 ± 995 mg/day in A group, –59 ± 1370 in B group, and 324 ± 1477 mg/day in C group). These data demonstrate the lack of a relationship between proteinuria and peritoneal albumin loss. In linear regression, the peritoneal albumin loss was negatively correlated with the nutritional status at the end of the follow-up period. Our data show that the initial nutritional status was associated with proteinuria alone, that follow-up nutritional status was associated with proteinuria and peritoneal albumin loss. High proteinuria levels may have a greater effect on nutritional status than peritoneal albumin loss.
The patients on PD who had high proteinuria levels showed rapid RRF decline and eventual improvement of nutritional status. The GNRI scores increased from 86.4 ± 9.2 to 94.9 ± 7.3 in the C group. The C group demonstrated a significantly higher initial protein loss in the urine accompanied by malnutrition. However, the follow-up GNRI was not significantly different between the C group and the other 2 groups. The impact of the initial high proteinuria level on malnutrition was attenuated, along with a decreased RRF with increasing duration of PD. The change in the serum albumin level was similar to that in GNRI.
Preservation of RRF has been consistent independent predictor of survival in patients on PD
[3–5, 24]. The benefits include improved control of blood pressure and anemia, removal of uremic toxin, superior phosphate control, and dialysis adequacy
. Wang et al. showed that RRF preservation is associated with better nutritional status according to appetite improvement or the removal of inflammatory mediators
. However, uncontrolled high proteinuria levels are related to malnutrition. In fact, patients with uncontrolled high proteinuria levels who are not on dialysis could be treated with medical nephrectomy using iatrogenic nephrotoxic drugs. This study shows that RRF combined with heavy proteinuria is associated with malnutrition and that a rapid RRF decline leads to improved nutritional status. This study also shows that RRF declines lead to decreased dialysis adequacy. Because there were no significant differences in peritoneal Kt/V between the 3 groups at baseline or follow-up, the decline of the weekly Kt/V may be associated with the decreased renal clearance. However, there was no significant difference in number of patients with inadequate dialysis.
Previous studies have shown that RRF loss is associated with malnutrition in patients on PD
[24–26]. However, these studies have limitations that differ from those of the present study. First, they had a cross-sectional design. Anuria or low RRF as a baseline characteristic was associated with poor nutritional status, but there are few data on the effects of RRF decline on nutrition. Second, patients were divided into groups according to RRF status. Patients with heavy proteinuria may be enrolled into groups along with patients with RRF. If patients with RRF were divided into groups with respect to proteinuria, subgroup analyses may show similar results. The present study had an observational design and simply monitored changes in RRF status. Patients with RRF were divided into 3 groups according to proteinuria status.
The patients with nephrotic syndrome and normal renal function have increased protein synthesis rates but inadequate responses to normalized serum albumin levels
. Patients on dialysis also experience increased protein synthesis in response to protein loss, but the response to protein loss in patients on dialysis is greater than that in patients with nephrotic syndrome and normal renal function
[2, 28]. Therefore, this malnutrition change may be more closely associated with increases in protein catabolic rate than renal function. The present study did not show whether the nutritional improvement in patients on PD with nephrotic-range proteinuria levels is associated with changes in protein catabolic rate or decreases in proteinuria levels. Further investigations are needed to determine the definite association between improved nutrition and RRF decline or decreased proteinuria levels.
In the present study, MAP was not associated with a decline in RRF, and at the end of the follow-up period, MAP in C group was higher than that in the other groups. Previous studies have shown that MAP in the post-dialysis period is not associated with a decline in RRF in patients on PD
[29, 30]. However, diastolic blood pressure at predialysis period is related to decline in RRF
. The relation between blood pressure and RRF remains controversial. The higher MAP observed in C group at the end of follow-up period may be associated with rapid decline in RRF or underlying disease (high incidence of DM in the C group). In our cohort, MAP at the end of the follow-up period was 102 ± 14 mmHg in patients with RRF < 1 mL/min/1.73m2 and 97 ± 12 mmHg in those with RRF ≥ 1 mL/min/1.73m2 (P = 0.020). The RAS blockade is a well-known therapeutic target associated with RRF preservation
. Most patients in the present study received a RAS blockade, and we were unable to analyze the association between RAS blockade use and RRF decline.
It is important to evaluate whether aminoglycoside use is associated with RRF decline in patients on PD. Some previous studies have shown variable results regarding the association between aminoglycoside use and rapid RRF decline
[17, 31, 32]. A multicenter study recently revealed that empiric treatment with aminoglycoside for peritonitis was not associated with a rapid RRF decline
. In present study, all patients with peritonitis received aminoglycoside as first-line antibiotics. There was no significant difference in peritonitis rate or aminoglycoside use among the 3 groups. Investigations that enroll only patients with peritonitis are needed to determine the effect of aminoglycoside on RRF decline.
Serum albumin is not the gold standard of nutrition in patients on PD
. Serum albumin is greatly influenced by variable factors such as malnutrition, hydration status, and inflammation. This variable is included in the GNRI equation. Therefore, it is difficult to consider GNRI as a reliable measure of nutrition in patients with heavy proteinuria levels. At the time of PD initiation, the differences in GNRI among the 3 groups may be not associated with nutritional status. The present study shows data for nPNA, lean mass index, and fat mass index using DEXA. The lean mass index in the C group was greater than that in the other 2 groups at the time of PD initiation and the end of the follow-up period. However, there was no significant difference in lean mass index change from the time of PD initiation to the end of the follow-up period. However, because overhydration causes overestimation of lean mass by DEXA, the initial lean mass levels in C group may have been overestimated. Therefore, the increases in lean mass in C group may be higher than those in the other groups. There was no significant difference in fat mass index among the 3 groups, but we did detect a greater increase in fat mass index in the C group (delta fat mass index: of 0.76 ± 1.03, 0.75 ± 1.87, and 1.20 ± 1.49 in the A, B, and C groups, respectively). The same trend was seen in nPNA.