Nutritional state correlates with renal function at the start of dialysis.

OBJECTIVE

To determine the effect of renal function at the commencement of dialysis on nutritional state.

DESIGN

Retrospective cohort study.

SETTING

University tertiary referral hospital.

PATIENTS: Patients with end-stage renal disease commencing renal replacement therapy over 8.5 years.

MAIN OUTCOME MEASURES: Total body nitrogen assessment using in vivo neutron activation analysis, expressed as nitrogen index (NI = observed total body nitrogen/predicted total body nitrogen); serum albumin measurement and creatinine clearance estimation, using the Cockcroft-Gault equation, at the start of dialysis. "Early" start group = creatinine clearance > 10 mL/minute (n = 26). "Late" start group = creatinine clearance < or = 10 mL/min (n = 108).

RESULTS

Mean NI was significantly lower [95% confidence interval (CI) 6%-30%, p < 0.0001] in the late start group (NI = 88%) compared to the early start group (NI = 106%). The late start group also had a significantly lower (95% CI 0.06-0.47 g/dL, p = 0.02) mean serum albumin level (3.66 g/dL) compared to that of the early start group (3.93 g/dL). A positive independent correlation existed between renal function and total body nitrogen (r2 = 0.21, p < 0.0001). No other clinical or demographic variable at the start of dialysis was found to predict nutritional state.

CONCLUSIONS

Patients that start dialysis with low levels of renal function are likely to be significantly malnourished. Whether this translates into a survival advantage requires prospective study.