Intradialytic Parenteral Nutrition and Intraperitoneal Nutrition

Nonrandomized Studies of Intradialytic Parenteral Nutrition

Key nonrandomized studies are presented in Table 35.5 . In 1975, Heidland and Kult published the first report describing the use of IDPN therapy during a 60-week study period in 18 patients receiving maintenance hemodialysis three times a week. Patients were given 16.7 g of essential amino acids, including histidine, and 250 mL of a mixture of d / l -malic acid, xylitol, and sorbitol during the last one-half hour of hemodialysis. During the first 3 months, some nonessential amino acids were added to the IDPN. After 16 weeks of IDPN therapy, 13 patients were discontinued out of the 18 patients. With every hemodialysis session, about 100 g of protein was also included. The food intake was not constant. After 30 weeks of IDPN therapy, serum albumin and serum total protein levels were reported to increase significantly. Serum transferrin level, complement level, and hemoglobin level decreased. In a study by Piriano of 21 MHD patients who had lost at least 10% of their dry weight, 16 were treated with one type of IDPN mixture for 20 weeks. IDPN was given to these 16 patients who had 400 mL of 50% glucose and 400 mL of 8.5% essential amino acids and nonessential amino acids. Five out of 21 patients who had lost at least 15% of their dry weight were treated with a solution mix of 50% glucose and essential amino acids. The dialysis dose, the volume of IDPN, and the comorbid conditions of these MHD patients were not well described. Neither of these groups gained weight, and neither of these groups had any significant increase in serum albumin level. The only exception is that patients who received essential and nonessential amino acids, which did not have hyperparathyroidism gained weight.

Bilbrey et al. reported that in 47 MHD patients with severe PEW who received IDPN for 3 months, there was a significant increase in serum albumin levels (mean ± SD, from 3.30 ± 0.38 g/dL to 3.71 ± 0.30 g/dL, p < 0.001) and serum transferrin levels (mean ± SD, from 165 ± 37 to 200 g/dL ± 62 mg/dL, p < 0.001) in survivors. There was no increase in serum albumin or serum transferrin levels in the nonsurvivors. The dialysis dose, comorbid conditions, and duration of therapy of IDPN were not reported. Capelli et al. reported in their retrospective study of 81 MHD patients a survival difference between 50 patients who received IDPN and 31 patients who did not. In those patients treated with IDPN, the body weight was 10% less than their desirable body weight and/or who had at a minimum 10% weight loss. All patients in the study had reduced serum albumin levels. In the 31 patients comprising the control group, a history of recent weight loss or low body weight was not consistently present. All the patients in the study group received oral nutritional supplements and/or nutritional counseling at the initiation of the study. In those patients who did not respond to this supplementation of nutrition, IDPN was started. The IDPN solution had a 10% to 20% lipid emulsion (20–500 kcal/dialysis session), variable amounts of d -glucose based on the presence or absence of diabetes, and 50 g of essential amino acids. Mortality rate was 36% in the IDPN-treated patients ( p > 0.05) and 48% in the control group. The time to death was significantly greater (mean ± SD, 16.9 ± 7.9 versus 7.5 ± 4.2 months; p < 0.001) in the nonsurvivors in the IDPN-treated patients. A nonrandomized study reported by Foulks et al. had 72 patients with PEW who failed to respond to dietary counseling who received IDPN. If there was a 10% increase in dry body weight or an increase in serum albumin of ≥ 0.5 g/dL, these patients were stratified as responders while they received IDPN. In these responders, mortality was significantly lower. The responders had a significantly lower serum albumin levels compared to nonresponders (mean ± SD, 2.2 ± 0.7 versus 3.0 ± 0.8 g/dL; p < 0.0001). Both the responders and nonresponders had similar body weights both before and after treatment with IDPN. Responders had higher hospitalization rates during the 6 months prior to the initiation of IDPN ( p < 0.0001), but during IDPN therapy, only 52% of responders were hospitalized as compared to 76% of nonresponders ( p < 0.0001). Only the responders had significantly reduced hospitalization rates. It is a plausible hypothesis that the nonresponders’ increased morbidity reduced their ability to respond to IDPN therapy. It is also conceivable that the improved clinical course of the responders was due to IDPN.

A retrospective study compared 22,517 control MHD patients with 1679 MHD patients who received IDPN. In the 1679 patients who received IDPN, the composition of IDPN was not identical. There was a significant decline in the odds ratio of death at 1 year in those patients with serum albumin levels of ≤ 3.3 g/dL treated with IDPN after correction for case mix and predialysis serum creatinine levels compared to patients who had comparable serum albumin levels who did not receive IDPN therapy. In patients treated with IDPN who had serum albumin levels of ≥ 3.5 g/dL at 1 year, the mortality was higher. In the MHD patients who had a predialysis serum creatinine level of 8.0 mg/dL or lower, the survival effect of IDPN was higher.

Hiroshige et al., in a nonrandomized study, published the results of IDPN therapy in 10 MHD patients. The control group included 18 MHD patients who refused IDPN therapy and were given dietary counseling. The treated group received IDPN for 12 months and consisted of 200 mL of 50% glucose, 200 mL of 20% lipid emulsion, and 200 mL of 7.1% essential amino acids per dialysis session. The baseline nutritional measures between the two groups did not differ significantly. In the treated group, during the IDPN therapy, there was a significant rise in serum albumin and transferrin, triceps skinfold thickness, body weight, and mid-arm muscle circumference. The control group patients showed measurable and significant differences in these parameters. A reduction in plasma essential amino acids was seen in the control group over the duration of the study.

Dezfuli et al., in a prospective cohort of 196 MHD patients with reduced serum albumin levels, examined the predictors of response to IDPN in a multivariate logistic regression model ( Figs. 35.2 and 35.3 ). Patients with severe hypoalbuminemia, defined as a baseline serum albumin level of less than 3.0 g/dL, totaled 134. The average ± SD period of IDPN therapy was 5.8 ± 2.4 months. The baseline level of serum albumin was lower in MHD patients who responded to IDPN (mean ± SD, 2.68 ± 0.47 g/dL). A multivariate logistic regression analysis adjusted the associations for age, gender, diabetes, and IDPN time. The presence of severe hypoalbuminemia (serum albumin, 3.0 g/dL) at baseline was associated with a 2.5 times higher chance of responding to IDPN (95% confidence interval, 1.3 to 4.9; p = 0.006). The same severe hypoalbuminemia was associated with a 3.5 times increased likelihood of serum albumin correction by at least 0.5 g/dL (95% confidence interval, 1.8 to 6.8; p = 0.001). The authors concluded that the degree of response to IDPN correlated with the severity of hypoalbuminemia.

Randomized Prospective Controlled Trials of Intradialytic Parenteral Nutrition

Key randomized trials on IDPN are presented in Table 35.6 . Over 12 weeks, Cano et al. studied 26 MHD patients with PEW. The IDPN-treated group consisted of 12 patients, and the rest were in the control group. The treated group exhibited a significant rise in body weight and serum transthyretin (prealbumin) and serum albumin levels. There were also increases in mid-arm muscle circumference, skin test reactivity, plasma apolipoprotein A1, and plasma leucine levels. None of the control patients experienced a rise in any of these parameters. Of note, the IDPN-treated group had a statistically nonsignificant lower baseline value of many of these measures. This may or may not have predisposed them to increases of these measures during therapy with IDPN. The plasma lipid levels did not change, but the plasma apolipoprotein A1 levels increased. The IDPN therapy provided an estimated 1.6 g/kg body weight of fat and 0.08 g/kg of nitrogen from essential and nonessential amino acids.

In a study by Toigo et al., 18 MHD patients were assigned in a random fashion to three IDPN treatment regimens. These regimens differed according to the amino acid content of the solutions. In these 18 patients, most had PEW. Patients received either only essential amino acids or a combination of nonessential and essential amino acids or an amino acid–free solution with an isocaloric infusion of 5% glucose. All three IDPN regimens contained vitamins, minerals, and trace elements. IDPN was given three times weekly for 2 months. The only change experienced in the three groups was that the patients who received only essential amino acids had gains of body weight. Twenty-one patients on MHD were randomly assigned to receive IDPN therapy for 6 months. Ten patients were given a combination of nonessential, and essential amino acids and 11 patients were given only essential amino acids as the nitrogen source. Baseline energy input before the initiation of IDPN was low, and at baseline, the mean serum albumin level was normal. Protein intake and mean baseline body weight were decreased but not significantly. With the essential and nonessential amino acid acids, there was a reduction in serum albumin levels and an increase in normalized protein nitrogen appearance.

The French Interdialytic Nutrition Evaluation Study (FineS) is, to date, the largest prospective randomized study to examine the benefit of IDPN in dialysis patients. A total of 186 MHD patients aged 18–80 years undergoing dialysis for at least 6 months were randomly assigned to receive IDPN therapy ( n = 93) or not receive therapy with IDPN ( n = 93). The treatment period was 1 year, and IDPN was given during three times a week dialysis. All patients, i.e., both the IDPN-treated group and the control group, received oral nutritional supplementation. All patients in the study had PEW, defined as having at least two of the following: (1) BMI of < 20 kg/m ² ; (2) serum albumin < 3.5 g/dL; (3) edema-free weight loss of greater than 10% over the previous 6 months; and (4) serum transthyretin levels < 30 mg/dL. The exclusion criteria were (1) total parenteral nutrition (TPN) received within the 3 months preceding the study; (2) single pool Kt/V < 1.2; (3) less than 12 hours of dialysis treatment per week; (4) presence of severe comorbid conditions that adversely affect 1-year survival; (5) fasting serum triglyceride levels > 300 mg/dL; and (6) hospitalization at the time of randomization. The follow-up period was 2 years. Energy and nutritional intakes were monitored at baseline and at 3, 6, 12, 18, and 24 months. The IDPN therapy in this study provided at months 3, 6, and 12 equivalents of 6.6 ± 2.2, 6.4 ± 2.1, and 6.1 ± 2.2 kcal/kg (mean ± SD) of amino acids. To estimate the time-averaged dose given daily, one should multiply these estimates by 3 (three sessions of hemodialysis per week) and then divide by 7 (the number of days per week). At months 3, 6, and 12, oral supplements provided 5.9 ± 2.7 kcal/kg/day and 0.39 ± 0.18, 0.38 ± 0.18, and 0.37 ± 0.18 protein/kg/day. There was patient-to-patient variability as well as variability among hemodialysis centers with regard to the intake of nutrients from the IDPN and spontaneous intake. There was no statistically significant difference in mortality, hospitalization rates, and indices of PEW between the IDPN-treated group and the control group. The Karnofsky scores did not change from baseline in either the control group or the group that received IDPN. Both groups received oral nutritional supplements, and this may have played a role in the failure to demonstrate any difference in outcomes between the groups. Of note, the patients in the IDPN-treated group who had a rise in serum transthyretin of greater than 30 mg/L within the first 3 months of the study showed an approximately 50% reduction in mortality in 2 years. The authors note that although this is the largest prospective randomized trial to date, it was still underpowered.

Amino Acids

The amino acid formulation contains essential and nonessential amino acids, and the initiation should be 1.0 to 1.2 g/kg dependent on comprehensive assessment of the patient. It is suggested to avoid protein restriction in patients with liver failure, using the same recommendations utilized for other critically ill patients and using the patient’s usual weight rather than the actual weight in predictive equations due to complications associated with ascites, edema, portal hypertension, and hypoalbuminemia. Amino acids can be delivered in doses as high as 1.8 g/kg or potentially higher depending on the patient’s lean body mass catabolism.

The amino acid’s impact on volume is dependent on its concentration. For example, 100 g of protein in 10%, 15%, or 20% is 1000 mL, 667 mL, or 500 mL, respectively. Thus, a 20% amino acid solution has the least volume, and in the CKD population, which is sensitive to volume management, this is of utmost importance. Of note, the 10% amino acid solution has the lowest osmolality, and in identifying patients with intolerance, this should be taken into consideration. A two-in-one solution with amino acid and dextrose is administered over a 2-week titration schedule, with 50% of the dose administered during week 1 and achieving the goal rate by week 2 in order to assess tolerance to the infusion.