Abstract: SA-PO741
Factors Affecting Mortality in the Very Elderly on Haemodialysis – A Single Centre Study
Session Information
- Geriatric Nephrology
October 27, 2018 | Location: Exhibit Hall, San Diego Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Geriatric Nephrology
- 1100 Geriatric Nephrology
Authors
- Roper, Tayeba, Epsom & St Helier NHS Trust, Surrey, United Kingdom
- De'Ath, Thomas G., Epsom and St Helier NHS Trust, London, United Kingdom
- Steele, Maggi, Epsom & St Helier NHS Trust, Surrey, United Kingdom
- Wang, Joe, Epsom & St Helier Hospital NHS Trust, London, United Kingdom
- Makanjuola, David, St. Helier Hospital, Carshalton, surrey, United Kingdom
Background
The number of elderly patients currently managed on haemodialysis (HD) is increasing. We aimed to determine the outcome of very elderly patients (over the age of 85 years) on HD and to identify any factors associated with mortality in this group.
Methods
Data were retrospectively collected from a single UK centre over 11 years (2003-2014). HD dependence was defined as those requiring HD for more than 90 days. All patients starting HD at the age of 85 years or over were included. Demographics, reasons for starting HD, access at initiation and co-morbidities were recorded. The Davies Co-morbidity Score (DCS) was used to quantify co-morbidities in patients. Outcomes were of death, recovery of function and continued HD. Chi-square tests were used for categorical analysis of DCS.
Results
Of 149 patients, 144 were included (5 excluded due to lack of data). Age ranged from 85 to 98 years (modal age 85.04). 73.6% were male. Time on HD ranged from 95 - 2598 days (median 546 days). Overall mortality at 1-year was 18.8% (18.2% for 85-90 years and 25% for over 90 years). 74.3% started HD with a line and 25.7% with an arterio-venous fistula (AVF): A higher 1-year mortality (16.7%) was seen in the former group compared to the latter (2.1%). Overall 72.2% were non-diabetic (ND) with mortality at 1-year being 16.0% compared to 2.8% in diabetic patients. Mortality according to DCS is shown in Table 1. There was no significant difference in 1-year mortality between DCS groups (p >0.05).
Conclusion
Our HD cohort demonstrates better overall 1-year survival than that of the UK Renal Registry 2015 (71% in 85+ years). Type of HD access was the most important factor affecting mortality. Though the reasons for this are varied, HD via a line may reflect a more acute decline in renal function. DCS and diabetes were not independent predictors of mortality in this cohort. It may be that, in this age group, traditional markers of renal progression are no longer relevant.
Further work is needed to look at other markers of survival and HD benefit in this age group, such as frailty and quality of life scores, to aid decision making with regards to HD in the very elderly.
Table 1
| Davies Co-morbidity Score | 1-year Mortality (%) |
| 0 | 0.7 |
| 1 | 16.0 |
| 2 | 2.1 |