Elderly patients represent a growing population among people suffering from ESRD. So far only limited data on actual survival benefits of elderly adults initiating dialysis have been published. Besides the high burden of preexisting comorbidities, dialysis treatment itself may be associated with a further deterioration in functional status in this population. We retrospectively analyzed the Austrian Dialysis and Transplant Registry and identified 8, patients who started maintenance hemodialysis after the age of 65 years between and All patients who died of malignant disease were excluded from this analysis. The risk of mortality was analyzed using multivariable Cox proportional hazards models.
|Published (Last):||14 August 2018|
|PDF File Size:||14.72 Mb|
|ePub File Size:||6.11 Mb|
|Price:||Free* [*Free Regsitration Required]|
Elderly patients represent a growing population among people suffering from ESRD. So far only limited data on actual survival benefits of elderly adults initiating dialysis have been published. Besides the high burden of preexisting comorbidities, dialysis treatment itself may be associated with a further deterioration in functional status in this population.
We retrospectively analyzed the Austrian Dialysis and Transplant Registry and identified 8, patients who started maintenance hemodialysis after the age of 65 years between and All patients who died of malignant disease were excluded from this analysis. The risk of mortality was analyzed using multivariable Cox proportional hazards models. Furthermore, a parametric model of time to event analysis was used for visualization of changing risk over time and precise calculation of time to equal risk assuming a Weibull distribution.
Hemodialysis treatment was associated with a decreased risk for death with a HR of 0. The time to event analysis however showed, that although survival was initially superior in the hemodialysis group, hazards crossed thereafter.
Time to equal risk was 2. Elderly patients with ERSD did benefit from initiation of hemodialysis, as the conservative group showed a very high initial mortality rate. This survival benefit of dialysis treatment however did not persist beyond the first two months compared to survivors of the conservative group.
This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Over the last decades, end stage renal disease ESRD patients became significantly older and sicker [ 1 ]. Today almost every other dialysis patient in industrialized countries is older than 65 years and patients older than 75 years represent the fastest growing age group of prevalent dialysis patients [ 2 ].
This development is mainly driven by the continuing increase in life expectancy in these countries and a more liberal access to renal replacement therapy for older patients [ 3 ]. For conservatively managed patients no such data repositories exist [ 5 ]. Dialysis is a life-saving procedure for patients with ESRD. However, it is associated with considerable comorbidities and a high mortality that is driven by frequent infections, chronic inflammation, accelerated atherosclerosis and malnutrition.
In patients with severe comorbid illnesses including cardiovascular disease and diabetes the short-term mortality is often very high despite the initiation of renal replacement therapy [ 6 — 8 ]. Additionally, initiation of hemodialysis in elderly patients may further deteriorate functional status and reduce quality of life [ 9 — 11 ]. Transportation to and from hemodialysis units is often an all-day affair for patients with limited mobility, leaving very little time for other activities on hemodialysis days including reduced nutritional intake [ 12 ].
When evaluating elderly patients for renal replacement therapy RRT several additional parameters have to be taken into consideration including the remaining live expectancy of the patient on dialysis which is most often limited by comorbidities including diabetes and advanced vascular disease. In elderly patients or younger patients in a palliative setting ESRD may adequately be managed conservatively without the initiation of RRT especially in individuals with relatively physical wellbeing in terms of ESRD related symptoms [ 12 ].
This includes correction of acid base and electrolyte disorders, fluid and blood pressure control and anemia management. It has been shown that hemodialysis itself leads to a progressive loss of kidney function [ 13 ].
However, conservative management can only be continued as long as clinical symptoms of uremia are controlled by supportive medical treatment. This retrospective analysis has been approved by the ethics committee of the Medical University of Vienna. We only received deidentified data from the transplant registry.
Data obtained from the medical records of the outpatient department was deidentified following identification of subjects that met the inclusion criteria.
Only the first author had access to identifying data as part of the initial data mining process before deidentification of the data set that was performed by the first author. The IRB consented to a retrospective data analysis without need for an individual consent.
Data evaluation and deidentification was performed according to the IRB requirements. We thereby identified a cohort of conservatively managed patients with CKD stage 5 aged 65 and older.
In order to adjust for the different numbers of patients between the groups, we calculated a propensity score by applying bootstrap resampling with runs including all patients of the conservative group and the same number of randomly chosen patients from the dialysis group. The propensity score included comorbidities COPD, diabetes, hypertension, heart disease, neoplasia, liver disease and vascular disease as well as age and sex. Demographic variables were compared by the two-sample t-test.
For categorical variables the chi-square test was applied. Kaplan-Meier KM plots were used to visualize the association of dialysis treatment with patient mortality. Significance was calculated by log-rank test.
A multivariable Cox model was computed, which included modality of treatment, age, sex and all comorbidities as covariates. To investigate the change of the hazards over time Cox models including the same covariates were computed for two sequential time periods 0—2 months since study inclusion and after 2 months since study inclusion. Administrative censoring at 2 months was applied for the first model.
Only patients who survived the initial period entered the model for the subsequent period. Furthermore, we calculated a parametric model assuming a Weibull distribution for survival analysis stratifying by treatment modality dialysis vs.
A p-value less than 0. For all analyses SAS for Windows 9. The demographic data for both groups are presented in Table 1. Both groups also significantly differed by gender. Patients in the conservative group had significantly more comorbidities. Values are presented as mean standard deviation or as percentage.
Fig 1 shows a Kaplan-Meier KM plot to visualize patient mortality stratified by treatment modality. Median survival time was Median survival in the hemodialysis group decreased from Hemodialysis treatment was associated with a survival benefit with a HR for death of 0. Age and comorbidities such as diabetes mellitus, heart disease, liver disease and vascular disease were all associated with an increased mortality, whereas hypertension had a beneficial effect on survival in our analysis Table 2.
To further assess changing hazards over time we calculated hazard ratios for two time periods. In the initial two months the HR for death comparing dialysis treatment vs. For the subsequent period this survival benefit did not persist.
There was a non-significant trend to better survival in the conservatively managed group with a HR of 1. The parametric analysis showed, that although survival was initially superior in patients treated with hemodialysis, hazards crossed thereafter Fig 2. Time to equal risk was calculated at 2. For male patients time to equal risk decreased from 1. Dialysis treatment vs. E marks the time to equal risk the slopes of the two survival curves are the same. Our data shows an overall survival benefit for elderly patients started on hemodialysis treatment compared to conservative management.
This hazard ratio is driven by the very high initial mortality in the conservatively managed patients with a median survival of only one month. Our results are in line with previous published data as all studies so far showed an overall survival benefit for elderly patients treated with RRT compared to maximum conservative management alone [ 14 — 19 ].
However, selected individuals opting for a conservative management show an unexpected longevity [ 18 , 20 ]. Besides, an increase in survival time alone may not be the main objective in the care of elderly patients with ESRD [ 5 ]. Older patients on hemodialysis have a great symptom burden pain, fatigue, pruritus, constipation , which can only be poorly addressed by the dialysis treatment and often remain unchanged [ 12 ].
Across countries, mortality rates among hemodialysis patients aged above 65 years show a great variance. Over the last decade several studies on the survival of elderly patients with ESRD opting for conservative treatment have been published.
Joly et al. Similar findings were reported by Hussain et al. In our cohort median survival in the conservatively managed patients was very poor compared to most previously published data. In the comparative statistics, the conservative group showed a higher burden of comorbid illnesses and was significantly older.
A gender discrepancy in prevalent dialysis patients has been observed before [ 24 ]. Potential factors driving this observation are currently investigated in large registry analyses. The factors contributing to this gender discrepancy remain elusive. The observed high mortality in the conservative group may further be driven by the data mining approach based on serum creatinine values that was applied to identify CKD patients with renal failure in a clinical database.
This may result in a substantial bias by indication including severely ill patients that did not qualify for hemodialysis in the first place. We therefor addressed the selection bias by excluding all patients who died due to a malignant disease and by calculating a propensity score for treatment based on all clinical covariates that are available in the dialysis registry data.
We are therefore aware that patients in the conservative group may have more comorbidities resulting in an indication bias that cannot be completely addressed in an observational study design.
Another limitation of this analysis is the low number of patients with conservative treatment that remain in the study for longer than one year. We therefore focus on results within the first few months after study inclusion. Comparing survival in conservatively managed patients between different cohorts and to patients on hemodialysis is difficult as there is no clear starting point for analysis and thus a potential lead time bias.
Carson et al. The right time to start on chronic renal replacement therapy in elderly patients remains elusive and recommendations of international societies have changed over the last decade [ 25 , 26 ].
In the general population early initiation of renal replacement therapy RRT did not provide a survival benefit [ 27 ].
Additionally, eGFR alone as a continuous parameter remains a poor indicator for the timing of dialysis initiation. In our population, no information on uremic symptoms were available for both groups.
Lead time bias is therefore difficult to assess and may be another source of confounding in an observational study. Despite an overall survival benefit in the dialysis group, a time to event analysis showed that hazards for death crossed after two to three months resulting in a better survival in conservatively managed patients who survived for more than two months.
Using time to equal risk analysis allowed us to calculate the time point at which a conservatively treated patient had the same risk for death compared to a patient on hemodialysis.
Time to equal risk is the time point when the slopes of the two survival curves are the same. After this time point the risk to die is higher for a dialysis patient than for one treated conservatively. Time to equal survival on the contrary is the time point, when the two survival curves cross. Time to equal risk analysis help to identify the time point when one treatment becomes superior to another when analyzing cohorts with changing hazards over time.
Register for a free account
Log in to view full text. If you're not a subscriber, you can:. Colleague's E-mail is Invalid. Your message has been successfully sent to your colleague. Save my selection. Maryann Kelly is an assistant professor of anesthesiology and assistant director of the graduate program in nurse anesthesi at Baylor College of Medicine, Houston, TX. You may be trying to access this site from a secured browser on the server.
Approach to the management of end-stage renal disease
Identify and refer patients at risk of end-stage renal disease ESRD. Renal transplantation is the best and most cost-effective renal replacement therapy for suitable patients. Create an arteriovenous fistula in preparation for haemodialysis to avoid using central line access. The two terms end-stage renal disease ESRD and failure ESRF are used to describe the irreversible loss of kidney function which, without treatment by dialysis or kidney transplantation, is likely to lead to fatal complications such as hyperkalaemia or pulmonary oedema over a period of days or weeks. Early recognition of patients destined for ESRF is also necessary to allow time for them to become fully informed about their treatment options and to institute therapy electively. All patients in the UK starting renal replacement therapy RRT ie dialysis or renal transplantation are registered by the UK Renal Registry and their subsequent progress recorded year on year.