Analysis of emergency Department Frequentation among patients with advanced CKD (chronic kidney disease): Lessons to...

The decision to initiate renal replacement therapy (RRT) implies a wide margin of uncertainty. Glomerular filtration rate (GFR) tells us the magnitude of renal damage. Proteinuria indicates the speed of progression. However, nowadays more than 50% of patients are still initiating RRT hastily, and it is life threatening.

By analysing Emergency Department (ED) frequentation and causes of a hurried initiation, we can better schedule the timing of the start of RRT.

Retrospective and observational study of all CKD patients in our outpatient clinic. ED frequentation and hospitalisation (Hos) time were reviewed during a 12-month period. We analysed: 1) time at risk, purpose (modality of RRT), previous comorbidity; 2) causes of ED frequentation and Hos; 3) type of initiation: «scheduled» vs. «non-scheduled», and within these «non-planned» vs. «potentially planned».

Of a total of 267 patients (time at risk 63.987 days, 70±13 years, 67% males, 38% diabetics), 68 (25%) patients came to hospital on 97 occasions: 39 only ED, 46 ED+Hos and 12 only Hos. ED frequentation was one patient every 4.3 days, and bed occupation was almost 3 per day. Main causes: 47% cardiopulmonary (1/3 heart failure), 11% vascular peripheral+cerebral, 11% gastrointestinal: 8/11 due to bleeding (all with anticoagulants/antiplatelet agents). Thirty-one (12%) patients initiated RRT: of these, 14 (45%) were scheduled (6 PD, 6 HD, and 2 living donor RTx), and 17 (55%) were not scheduled or were rushed, all with venous central catheter. Following the objectives of this study, the non-scheduled group were itemised into 2 groups: 9 non-planned (initial indication of conservative management or patient's refusal to undergo dialysis, and diverse social circumstances not controllable by the nephrologist) and 8 were considered potentially planned (6 heart failure, one gastrointestinal bleeding and one peripheral vascular complication). This last group (potentially planned), when compared with the 14 patients who started treatment in a scheduled manner, had significant differences in that they were older, with more previous cardiac events, and GFR almost double that of the other group. All of them started treatment in the ED.

This analysis provides us with knowledge on those patients who may benefit from an earlier preparation in RRT. We suggest that patients with previous cardiac events, especially with a risk of gastrointestinal bleeding, should start the preparation for RRT even with GFR rates of 20-25ml/min. In spite of the retrospective nature of this study, and taking into account the difficulties of carrying out clinical trials in this population, we propose this suggestion as complementary to the current recommendations for a scheduled start using this technique.