[Prognostic significance of programmed dialysis in patients who initiate renal substitutive treatment. Multicenter study in Spain]

Abstract

Background: The aim of our study was to analyse patient characteristics, mortality and costs, all of them in relation to whether starting dialysis was planned or unplanned.

Methods: A total of 362 patients (227 male and 135 female) from five hospitals of the National Health System, who were started on chronic renal replacement therapy (RRT) during 1996 and 1997 were included. Patients who were started on RRT after acute renal failure were excluded. We carried out a retrospective analysis of the demographic characteristics, patients' conditions at the time of initiating dialysis and outcome and costs at six and thirty-six months of treatment. Patients were classified as planned (PL-D) or unplanned dialysis (UNPL-D), depending on whether or not the patient had a vascular or peritoneal access ready to use for initiating RRT.

Results: One hundred and eighty-six patients (51.4%) started on dialysis in the PL-D group whereas 176 (48.6%) did it as UNPL-D. In this latter group, 135 (37.3% of the total) had previously been monitored by a nephrologist, and 41 (11.3%) initiated dialysis without previous nephrological follow-up. UNPL-D was associated with older age (p < 0.001), non-nephrological follow-up (p < 0.001), diabetes (34.7% vs 22.6%) (p = 0.011), haemodialysis as a first mode of RRT (94.9 vs 81.7%) (p < 0.001), higher comorbidity risk (p < 0.001), dialysis initiation with uraemic symptoms or fluid overload (p < 0.001), increased blood transfusion requirement (p < 0.001) and lower serum albumin (p < 0.001), creatinine clearance (p < 0.001), haemoglobin concentration (p < 0.001), and weight (p = 0.002). In the PL-D group the main primary renal diseases were glomerular and polycystic disease, whereas interstitial and diabetic nephropathy were higher in UNPL-D group (p = 0.005). Multivariate analysis showed that previous non nephrological follow-up, uraemic symptoms, interstitial nephritis as primary renal disease correlated with UNPL-D initiation, and it was followed by choosing haemodialysis as first RRT. UNPL-D was also associated with increased number of days of hospitalization at the initiation of dialysis, and during the first 6 months (p < 0.001), increase of hospitalization days (p = 0.009), and increased 6-month-mortality (10.2% vs 3.2%) (p = 0.015, log rank test), and three-year mortality (24.2 vs 36.9%) (p = 0.006, log rank test). The costs of UNPL-D were fivefold that of the PL-D group.

Conclusion: UNPL-D has been associated with worse overall clinical conditions at the initiation of chronic replacement therapy, choosing haemodialysis as first RRT, increased morbi-mortality and subsequent increase of costs.