The prevalence of chronic kidney disease (CKD) in Spain has increased by 20% in the last ten years and it increases progressively with age.1 According to data from the 2016 Spanish Registry of Renal Patients, the prevalence of CKD in people aged over 64 is 25% and it is 22% in those older than 75 years.2

In addition to the ageing of the population, high prevalent diseases such as hypertension (HTN), diabetes and cardiovascular disease are contribute to the increase in CKD3; this means that it tends to occur in patients with significant comorbidity.

We are therefore increasingly faced with the clinical decision of whether to start renal replacement therapy (RRT) or conservative kidney management (CKM), having to weigh up the benefits of each intervention, in older patients with high comorbidity.4

Although overall survival tends to be higher in patients who undergo dialysis versus those who do not, the advantage is lost in patients over 80 with a greater comorbidity, especially if patients have coronary heart disease.5,6 In addition, the burden of symptoms (pain, fatigue, anorexia, dyspnoea) that these patients have on dialysis is high,7,8 their health-related quality of life (HRQoL) is often deficient,9,10 and many have progressive secondary functional deterioration.11

In view of the evident limitations of RRT for these patients, there is currently increasing interest in determining whether CKM or active management without dialysis of advanced CKD (ACKD) is a valid therapeutic option.

CKM is a comprehensive way of managing the patient which should include: interventions to delay the progression of kidney disease and minimise the risk of adverse events; shared decision making; active management of symptoms; detailed communication; advanced care planning; psychological support; and social, family and spiritual support.12

It can be difficult for us, as nephrologists, to initiate these conversations,13 in part due to the lack of prognostic information to help us advise our patients in the decision-making process.

The majority of studies report that patients on CKM have a worse prognosis, shorter survival14,15 and reduced HRQoL. Most of these studies are cross-sectional and retrospective.16

In order to provide information about the utility of CKM to patients and healthcare professionals and facilitate the process of shared decision making, we carried out a detailed analysis of patients choosing CKM. Our aims were:

• 1.

  To identify variables predictive of mortality in patients opting for CKM.

• 2.

  To analyse the changes in HRQoL in this population on conservative management.

A total of 82 patients on CKM were included: 56 patients in 2015, 23 in 2016 and three in 2017. Over the course of the follow-up, one patient was excluded from the analysis after deciding to abandon CKM and start haemodialysis and two patients because of loss to follow-up.

The causes of ACKD were: nephroangiosclerosis (57.4%); diabetic nephropathy (25.7%); chronic tubulointerstitial nephropathy (6.1%); cardiorenal syndrome (4.9%); glomerulonephritis (2.4%); and other (3.7%).

Of the 82 patients in the study, the main reason for opting for CKM was: a serious illness unlikely to improve with RRT in 61 patients; severe functional dependence in eight patients; advanced cognitive impairment in five patients; own decision in five patients; limiting psychiatric disease in two patients; and metastatic cancer in one patient.

The baseline characteristics of the sample overall and of the deceased and non-deceased patients are shown in Table 1.

The mean age was 84.79±3.99 years, 50% of the patients were male and 56% had history of previous vascular event. A Charlson index score above 8 shows a high degree of comorbidity.

Analysis of mortality rates and factors predicting death

During the follow-up period 38 patients died: 26 (68.4%) in hospital and 12 (31.6%) at home. The causes of death were cardiovascular disorders in 15 patients (39.5%), progression of ACKD in 14 patients (36.8%), infections in four patients (10.5%), metastatic cancer in three patients (7.9%), trauma in one patient (2.6%) and unknown in one patient (2.6%).

We analysed the difference between patients who died and those who were still living after the follow-up period. Patients who died had a higher prevalence of previous vascular events, lower eGFR and albumin, and higher NT-proBNP levels. They also had a higher degree of dependence and frailty, a worse score in the physical component of the SF-36 and, at the limit of statistical significance, a poorer nutritional status and a higher Charlson comorbidity score.

The total mortality rate was 23/1000 patient-months (Fig. 1). Fig. 2 shows the mean monthly mortality rate by semesters over the course of the follow-up; after the first six months the mortality rate remained uniform for the rest of the follow-up period.

Fig. 1.

Survival curve of the cohort over the course of follow-up.

(0.05MB).

Fig. 2.

Mean monthly mortality rate by semester over the course of follow-up.

(0.05MB).

Median survival was 26.9 (95% CI: 19.6–34.2) months. There were significant differences (Fig. 3) in patient survival depending on whether or not they had a previous vascular event (mortality rate 36.7 vs 14.8, p=0.028), Charlson score ≥10 versus <10 (mortality rate 42 vs 17; p=0.002), and a high degree of functional dependence (mortality rate 48.4 vs 19; p=0.002), and whether or not they were frail (mortality rate 27 vs 10, p=0.05).

Fig. 3.

Survival of patients according to previous vascular event, Charlson Comorbidity Index, degree of functional dependency and frailty.

(0.29MB).

Table 2 shows the variables significantly associated with increased mortality rates by univariate analysis. The variables cognitive impairment by the Pfeiffer test and frailty were at the limit of statistical significance. Renal involvement parameters (GFR and proteinuria), a previous vascular event, the Charlson comorbidity index, malnutrition-inflammation parameters (albumin and MNA score), degree of dependency, physical quality of life, and phosphorus and calcium metabolism parameters (increase in PTH) were all predictors of mortality. In multivariate analysis, a previous vascular event, comorbidity measured by the Charlson index, descending albumin and elevation of PTH were independent predictive variables related to mortality in older patients receiving CKM.

Table 2.

Variables associated with higher mortality rates: univariate and multivariate analysis.

	Univariate analysis		Multivariate analysis
	HR (95% CI)	p	HR (95% CI)	p
eGFR (ml/min/1.73m2)	0.81 (0.72–0.91)	<0.001
Proteinuria (g/l)	1.50 (1.26–1.92)	<0.001
Albumin (mg/dl)	0.34 (0.17–0.68)	0.002	0.23 (0.10–0.52)	<0.001
PTH (pg/ml)	1.002 (1.000–1.005)	0.05	1.004 (1.001–1.007)	0.007
Previous vascular event (presence)	2.11 (1.06–4.2)	0.03	2.70 (1.05–6.92)	0.028
Charlson comorbidity index (points)	1.40 (1.11–1.77)	0.004	1.56 (1.14–2.12)	0.005
Barthel index (points 0–100)	0.98 (0.97–0.99)	0.005
Pfeiffer (points)	1.08 (0.99–1.19)	0.07
Frailty (%)	0.37 (0.13–1.05)	0.06
MNA (points)	0.86 (0.75–0.99)	0.04
SF-36 physical component	0.94 (0.89–1.00)	0.05
SF-36 mental component	0.99 (0.96–1.03)	0.84

We found in our study that the patients treated with CKM were patients whose kidney disease had become chronic and who had multiple problems26 due to associated comorbidity (56% of patients had had a previous vascular event, 38% were diabetic and the mean Charlson comorbidity index score was 9). Many of these patients were complex chronic patients as, in addition to having different progressive illnesses, they were also on polypharmacy, were dependent for basic activities of daily living and had poor social status (13.4% of our patients had moderate dependence, 2.4% severe and 9.8% total, and 20% had socio-family problems). CKM should not mean “no treatment” or less specialised care. It must provide comprehensive care which aims to delay the progression of the kidney disease, manage the symptoms and, when the time comes, plan advanced palliative care.6

Our results show that the median survival in CKM patients was 26 months. Wong et al.15 reported a survival time of 23.4 months, but included patients with GFR below 30ml/min/1.73m2. Chanda et al.,27 reported mean survival of 21.2 months, Murtagh et al.,5 18 months and Ellam et al.,14 21 months, but in these studies the GFR was 15ml/min/1.73m2.

The above studies, all with similar heterogeneous inclusion criteria, analysed the survival of older people on CKM compared to RRT. The majority showed that dialysis increased survival compared to conservative management, but being over 80,28 having comorbidity27,29 and coronary heart disease5 substantially reduced these differences.

The mortality rate was consistent throughout the follow-up period. The fact that the mortality rate did not increase in a given period, particularly during the first months of treatment, and that the main causes of death were cardiovascular and progressive deterioration in kidney function underline that these patients did have the benefit of adequate follow-up and multidisciplinary care, without resorting to other more aggressive interventions.

The geriatric assessment30–33 provides an overall approach to older people through clinical assessment (with identification of chronic diseases and nutritional aspects), functional assessment, cognitive assessment, psychosocial assessment and frailty, and is becoming increasingly important in patients with CKD.18 Assessment and treatment require a multidisciplinary approach in patients with CKD, with strategies that identify factors for worse prognosis which can help in decision making.

In our study, we found that the patients with the shortest survival were those who were frail, with the greatest initial comorbidity, a previous vascular event and the greatest degree of dependence. In the univariate analysis, eGFR, proteinuria, albumin, nutritional status and increased PTH were also associated with mortality. In the final multivariate analysis, a high degree of comorbidity, a previous vascular event, a decrease in albumin and an increase in PTH all maintained statistical significance.

Although the mean albumin and PTH values in our patients were adequate, our analysis shows that the higher the albumin levels and the lower the PTH levels, the better the prognosis.

Previous studies describe comorbidity as a prognostic factor for mortality.15 Ellam et al.14 found that albumin levels above 3.5mg/dl were associated with longer survival. Joly et al.34 analysed patients over 80 on CKM and RRT and found that octogenarians who did not go on dialysis differed from those on RRT in terms of degree of dependence, comorbidity and diabetes. In the Cox analysis, the independent predictors of death on dialysis were nutritional status, functional dependence and the presence of macroangiopathy defined by peripheral vascular disease.

The prevalence of vascular calcification is documented in patients with ACKD.35 It is possible that it has a multifactorial aetiology and is due to ageing, comorbidity and mineral and bone metabolism (MBM) disorders. MBM disorders also lead to abnormal bone architecture,36 fractures37 and decreased mobility in these patients, which could explain the relationship we found between elevated PTH levels and mortality.

Frailty was not an independent predictor of mortality in our study, but the results were at the limit of statistical significance. One explanation could be the limited number of patients, which subtracts statistical power, and the superposition of comorbidity and dependence already described in previous studies in a cohort of dialysis patients.38

Most of the studies that analyse quality of life in patients with ACKD are focused on patients on dialysis, especially haemodialysis, and are of a cross-sectional observational design.9,10 We analysed the changes in HRQoL in CKM patients and our results show that the most affected areas were the physical dimensions and the difficulty with their social relationships. However, the patients perceived less limitation in daily activities due to emotional problems and had a good perception of their mental health from the point of view of a feeling of well-being and being calm.

In a recent study of survival and changes in HRQoL39 conducted with patients on CKM and RRT, physical quality of life in conservative management was sustained or improved after 12 months of treatment in less than half of the patients, but mental health was sustained or improved in somewhat more than half. Our results show that patients on CKM do perceive the physical limitations due to the passage of time and progression of their disease, but they also feel continued greater satisfaction in psychological dimensions, with a quality of life that does not worsen over time but remains stable, which is the primary aim of CKM.

As far as the contributions made by our study are concerned, we would like to emphasise that it was a prospective study which included assessment of renal function parameters, functional status, frailty, nutritional status and HRQoL. Our analysis of survival and quality of life in patients who have already made the decision to have conservative management may provide a useful tool for informing and guiding the patient in the shared decision process. We also performed a comprehensive analysis of older patients, with this being an increasingly numerous but not very visible group because of a lack of registries on conservative management.20

The limitations of our study include the lack of a comparative group, in which we could have analysed survival and quality of life in patients recommended for dialysis, and the small sample size, which reduces the statistical power of our results. We used version 2 of the SF-36 instrument which includes improvements in the metric characteristics of two dimensions, some simpler statements to facilitate reading and completion and a calculation programme which makes it easier to estimate missing responses. However, although it is translated into Spanish and its use enables greater cultural comparability, it has not been validated for the Spanish population, so we cannot compare our results with the general population in Spain.

In conclusion, in our group of patients on CKM we found the mortality rate to be high and uniform over time, independent predictors of which were a previous vascular event, comorbidity measured by the Charlson index, descending albumin levels and elevated PTH. We also found that quality of life remained stable during treatment. We therefore believe that adequate CKM could be a recommended therapeutic option in these patients, who will also receive active treatment of hyperparathyroidism and prevention and treatment of malnutrition, and autonomy and physical rehabilitation will be encouraged.

We consider necessary to have multidisciplinary teams (palliative care teams [doctors and nurses], geriatricians, nutritionists, physiotherapists and primary care) in place to provide better care in CKM patients. Also necessary is more coordination with palliative care and home support teams, especially in the later stages of the disease, in order to maintain or even improve HRQoL for these patients and their families.

The authors declare that they have no conflicts of interest.