Factors affecting survival in advanced chronic kidney disease patients who choose not to receive dialysis.

INTRODUCTION: Non-dialytic treatment (NDT) has become a recognized and important modality of treatment in end stage renal disease (ESRD) in certain groups of chronic kidney disease (CKD) patients. However, little is known about the prognosis of these NDT patients in terms of hospitalization rates and survival. We analyzed our experience in managing these NDT with a multidisciplinary team (MDT) approach over a three-year period.

PATIENTS AND METHODS: The Renal Unit at the Royal Liverpool University Hospital set up a dedicated MDT clinic to manage NDT patients in January 2003. Patients approaching end stage chronic kidney disease who chose not to dialyse were recruited from other nephrologists. The study group was classified according to age band (<70 years, 71-80 years, and >80 years), estimated glomerular filtration rate (eGFR) (<10 ml/min, 11-20 ml/min, and >20 ml/min) according to the Modified Diet In Renal Disease formula and Stoke comorbidity grade (SCG). The SCG is a validated scoring system for the survival of patients on renal replacement therapy. We also used the ERA-EDTA primary renal diagnosis codes. As there are no existing standards for NDT patients, we used the U.K. national set for haemodialysis patients as a reference and target for our NDT patients. Data was collected prospectively.

RESULTS: The median age was 79 years and the male: female ratio was approximately 1. The most common primary cause of kidney disease in the NDT study population was chronic renal failure of unknown cause n = 22 (31%), but the most common identifiable cause was diabetic nephropathy, n = 20 (28%). The most common comorbidity was ischaemic heart disease n = 25 (34%). Those achieving the standards for anaemia were 78% at referral. Only 30% of the NDT patients achieved the standard for blood pressure (<130/80 mmHg) at referral. Forty-three patients (60%) had no admissions at all. There were a total of 30 patients admitted on 58 occasions. Thirty-one (53%) of these were due to a non-renal cause. The median length of stay for the other NDT patients was 10 days. The median overall survival (life expectancy) was 1.95 years. The one-year overall survival was 65%. SCG was an independent prognostic factor in predicting survival in NDT patients studied (p = 0.005), the hazard ratio being 2.53, for each incremental increase in the SCG. At one year, the survival for comorbidity grade 0, 1 and 2 were 83%, 70% and 56% respectively. Of the 28 patients who died, 20 did so at home (71%).

DISCUSSION: The NDT of ESRD has become an important alternative modality in renal replacement therapy. With the emergence of epidemic proportions of CKD, more elderly patients with progressive renal disease will need to make informed decisions regarding renal replacement therapy. There is likely to be increasing number of elderly patients that will tolerate dialysis badly and who will be very dependent on others. We believe that there should be a multidisciplinary approach to assist the ESRD patients in choosing their modality of renal replacement therapy, and with an agreed care plan to support these patients in managing their chosen modality to achieve the best possible quality of life. There should be integrated services with primary care, community nurses, and palliative care teams to enable the majority of the patient's treatment to be carried out at home and to allow a dignified death. However. there was a statistically significant trend for shorter survival among those with greater comorbidities, as determined by the SCG. This is the first report of the potential importance of SCG as an independent prognostic factor in NDT patients. This will help us to counsel our patients in the future about their prognosis if they choose NDT as their modality of renal replacement therapy.

CONCLUSION: Our prospective study is the first and currently the largest observational study of a multidisciplinary approach in the management of NDT patients. SCG was an independent prognostic factor in predicting survival. In those patients who chose not to dialyse, SCG provides a potentially useful indication of expected prognosis.