Renal replacement therapy in ADPKD patients: a 25-year survey based on the Catalan registry

Autosomal dominant polycystic kidney disease (ADPKD) is an inherited disease caused by mutations in two genes, PKD1 and PKD2, which encode for polycystin 1 and polycystin 2, respectively: mutations in PKD1 are present in 85% of cases, and mutations in PKD2 in the remaining 15% [1‐3]. The abnormalities in these proteins cause the appearance and growth of kidney cysts that gradually impair renal function [1]. ADPKD is the cause of chronic kidney disease (CKD) in 5% of patients undergoing renal replacement therapy (RRT) in the USA [4], a rate that rises to 7.9% in our geographic area [5]. In 75-82% of patients it is additionally associated with hypertension, which often appears before CKD development [6], and with other systemic disturbances such as liver cysts, pancreatic cysts, brain aneurysms or colonic diverticula.

Progression of CKD in ADPKD basically depends on a combination of genetic and environmental factors, with hypertension as a significant prognostic factor. CKD is more severe and appears earlier in patients with PKD1 gene mutations than in those with PKD2 mutations (mean age at RRT initiation: 53 and 69 years, respectively) [3, 7]. Also, the prognosis is worse for those harbouring a truncating PKD1 mutation compared with those with missense mutations [8]. Various studies have shown that controlling hypertension slows down CKD development in patients with ADPKD [9]. A recent trial, however, has indicated that reduced glomerular filtration is a function of older age, and that there are no differences between hypertensive and normotensive patients [10]; hypertensionis present in nearly 90% of ADPKD patients once they have renal failure.

CKD is associated with a high risk of cardiovascular disease: cardiovascular mortality among dialysis patients is 10–30 times greater than among the general population [11]. This fact is explained by the higher prevalence of classic cardiovascular risk factors in such patients: hypertension, diabetes, dyslipidaemia and smoking [12]. Diabetic nephropathy and vascular nephropathy are the primary reasons for initiation of dialysis, specially among the North American population [4, 5]. Moreover, a slowdown in CKD progression has been verified when cardiovascular risk factors are controlled [13]. However, a lower prevalence of these risk factors has been described in patients with ADPKD, with the sole exception of hypertension [14], which is present in nearly 90% of ADPKD patients once they have renal failure.

The objective of this observational study is to analyse the epidemiological, clinical and survival changes that occur over time in ADPKD patients who have started RRT, as compared with non-ADPKD patients on RRT during the last 25 years.

A total of 20,033 patients in whom RRT had been initiated were registered between 1984 and 2009: 1,586 of them (7.9%) had ADPKD and 18,447 (92.1%) had other nephropathies.

As shown by Table 1 and Figure 1, age at initiation of RRT in the ADPKD group was significantly lower than in the remaining patients, for both males and females (p < 0.001). Among ADPKD patients, the mean age over the three periods investigated was 56.7 ± 10.9, 57.5 ± 12.1 and 57.8 ± 13.3 years, respectively, with no significant differences (p = 0.37). Among non-ADPKD patients, however, the mean age increased significantly (p < 0.001) as follows: from 54.8 ± 16.8 years in the first period to 61.7 ± 16.2 years in the second period to 63.9 ± 16.3 years in the third period.

Both groups contained more males than females. The ADPKD group comprised 837 (52.8%) men and 749 (47.2%) women, while the non-ADPKD group comprised 11,641 (63.1%) men and 6,806 (36.9%) women. No significant differences were identified regarding the incidence of males and females at RRT initiation in the ADPKD group (ratio, 1.1–1.2), but in the non-ADPKD group the number of men who started RRT was significantly higher than the number of women (ratio, 1.6–1.8) (Figure 2).

Table 2 shows that cardiovascular diseases, malignancies, chronic respiratory disorders and joint diseases are significantly less frequent among ADPKD patients (p < 0.001). Among the cardiovascular risk factors analysed, diabetes was significantly less frequent (p < 0.001) in ADPKD patients than in the non-ADPKD group (6.76% vs 11.89%), and no differences were observed regarding the prevalence of hypertension (86.24% vs 84.20%).

Most of the patients started RRT with haemodialysis in both groups. However, a prevalence analysis dated 31 December 2009 showed that transplantation was the most prevalent situation among ADPKD patients: 65.4% in the ADPKD group and 49.3% in the non-ADPKD group. The analysis also showed that fewer patients with ADPKD required erythropoietin (p < 0.001) and that they had a lower prevalence of hepatitis C virus (p < 0.001). The CRP levels were lower in the ADPKD group, but this difference did not attain statistical significance (p = 0.124) (Table 1).

Significant differences (p < 0.001) in the use of different types of vascular access at initiation of haemodialysis were also observed in both the periods analysed (1997–2003 and 2004–2009) in the ADPKD group. An increase in arteriovenous fistulae was observed (74.3% vs 78.6%), and the same was true for tunnelled catheters (2.6% vs 7.5%) and prostheses (2.6% vs 7.5%); there was also a reduction in the number of temporary catheters (22.3% vs 11.7%). When analysing patients with kidney transplants, the number of living donors increased only over the most recent years (10–11 per year), even though cadaveric transplants continue to be more frequent (95.1% of all kidney transplants).

The primary causes of death during the study period were cardiovascular, followed by infections, in both groups (Table 3). No significant differences were observed regarding the age of death between ADPKD and non-ADPKD patients (69.0 ± 10.0 vs 69.4 ± 12.6 years) (p = 0.41). Survival over the first 3 years of RRT was higher in the ADPKD group (p < 0.001) as well as in the subgroup of kidney-transplanted patients (p < 0.03) (Table 4; Figure 3). The renal graft survival was also longer in the ADPKD group (p < 0.001) (Table 4; Figure 4a and 4b).

Although there is no specific treatment for ADPKD as yet, the therapeutic advances to slow down CKD might cause one to expect that, over time, onset of ESRD in ADPKD patients will have been delayed until a later age. Our data, however, show that the age at initiation of RRT in these patients has not varied since 1984. This result is in concordance with other articles [15], but some publications [16‐18] do indicate that the age at initiation of RRT in ADPKD patients has indeed increased. According to Orskov et al. [16], age at onset of ESRD among their 693 ADPKD patients, followed up for 18 years (divided into three periods), rose from 55.9 (1990–1995) to 60.6 years (2002–2007), though these findings are subject to a potential bias, given the increasing number of patients in each period and especially the rise in the number of women.

The prognosis of CKD patients who start with RRT has improved over the last few years [19, 20]. In the European Registry, the percentage of patients older than 65 who started RRT increased from 22% in 1980 to 55% in 2005 [21], and this may be attributed to a better control of cardiovascular risk factors. We have not been able to find in the literature any studies comparing the age at onset of ESRD between patients with and patients without ADPKD. Based on our own results, we would stress that while, over the three periods examined, the age at initiation of RRT in non-ADPKD patients rose by more than 9 years, no such change occurred among ADPKD patients.

Epidemiological studies indicate that within the general population, the prevalence of men undergoing RRT is higher than that of women, in accordance with our findings [21]. These data are in agreement with some other publications that associate male gender with earlier initiation of RRT in ADPKD patients [22, 23]. In contrast, and in accordance with other studies [16, 24], we found no significant gender differences in relation to age at initiation of RRT. Our data show that male gender is a risk factor for initiation of RRT among non-ADPKD patients (ratio, 1.6-1.8), but not among ADPKD patients (ratio, 1.1-1.2).

Although most of our ADPKD patients started RRT with haemodialysis, the prevalence study revealed kidney transplantation to be the most widely used method, as expected based on the age at onset of ESRD in this population. The sparse prescription of peritoneal dialysis among ADPKD patients could be due to large kidneys, to the presence of colonic diverticula, to hernias or to massive polycystic liver disease. However, peritoneal dialysis is a first-line technique for this disease, as has been shown by several studies that revealed no differences concerning survival or risk of peritonitis [25, 26]. On the contrary, it was found that ADPKD patients on peritoneal dialysis had a higher survival rate than those on haemodialysis [25].

The factors related to greater ADPKD severity are primarily carrying a PKD1 mutation, especially if truncating, and hypertension [3, 8, 27]. Other relevant factors include: age at diagnosis [28, 29], increased number and volume of the cysts, left ventricle hypertrophy [30], proteinuria [31], haematuria, microalbuminuria, hyperuricaemia [32], low HDL-cholesterol levels, high urinary sodium and an increase in urinary osmolality [33]. Cardiovascular risk factors increase the rate CKD progression [4, 12], with both entities boosting each other, giving rise to a cardiorenal syndrome. ADPKD is a typical model of type IV cardiorenal syndrome [34], as the cause of CKD is primary renal failure and secondary cardiovascular involvement. We have observed a lower prevalence of diabetes among patients with ADPKD, but the differences in comparison with non-ADKPD patients have not been found to be significant with regard to hypertension, similarly to other studies [14, 35‐37]. The data published on associated diseases in ADPKD and non-ADPKD patients are either incomplete [38] or have concerned only cardiovascular events [39]. We have observed that cardiovascular diseases, malignancies and respiratory diseases are more frequent among non-ADPKD patients on RRT. This higher rate of co-morbidities in non-ADPKD patients could be explained by the higher prevalence of diabetes and by initiation of RRT at an older age. The lower cardiovascular morbidity among ADPKD patients could be associated with higher haemoglobin levels in this group [35]. In correspondence with Abbott and Agodoa’s results [35], we found that fewer ADPKD patients on dialysis need erythropoietin compared with the non-ADPKD group (77.9% vs 91.9%). CKD favours a chronic inflammation status with increased inflammatory markers such as C-reactive protein (CRP) [37]. In accordance with other articles, we observed no significant differences between the groups with regard to CRP levels, although they were lower among ADPKD patients on dialysis, probably owing to the lower prevalence of associated diseases [40].

The hugely significant genetic factor in ADPKD could be related to a fatal renal prognosis that is quite difficult to modify with the currently available therapeutic tools which improve cardiovascular factors. Stimulation of the renin-angiotensin system (RAS) has been associated with a sombre ADPKD prognosis owing to faster growth of the renal cysts [24]. The HALT trial is underway to elucidate the contribution of angiotensin-converting enzyme inhibitors (ACEIs) and angiotensin receptor blockers (ARBs) and the strict control of hypertension to the slowing of decline in renal function [41]. Increased use of RAS-blocking anti-hypertensive drugs could imply that ADPKD patients will start on RRT later and later [16]. However, although ACEIs and ARBs are widely used for ADPKD in our population (specific data not available), this assumption is not in accordance with our findings.

Kidney transplantation is for the moment the best therapeutic choice for ADPKD patients with ESRD. Longer survival of both patient and graft has been observed in recipients of kidney transplants with ADPKD [20, 42], in agreement with our results. This increase in survival could be explained by the ADPKD patients’ younger age, lower rate of diabetes, lower prevalence of dyslipidaemia and anaemia, and also a lower prevalence of co-morbidities, mainly cardiovascular ones. It has not been possible to show that immunological factors may play a role in this survival difference. No explanation for a longer graft survival, other than the own patient’s survival, is available. The relatively low but increasing percentage of pre-emptive living donor transplantation is due to the transplantation policies of Spain, which has a highly successful programme for cadaveric transplantation and is encouraging living donor transplantation to overcome the current relative scarcity of cadaveric donors.

The primary cause of death among ADPKD patients on RRT, in our cohort, is cardiovascular disease, followed by infections, as reported by several other authors [43‐45]. Various articles have shown higher survival of ADPKD patients on RRT as compared to non-ADPKD patients, which concurs with our data [46, 47]. We found no differences as to the age of death, even though ADPKD patients started RRT earlier than non-ADPKD patients; this indicates a longer survival of ADPKD patients on RRT. Longer survival of ADPKD patients has been associated with: starting RRT at a younger age, lower prevalence of the cardiovascular risk factors and higher haemoglobin levels, all of which are factors applicable to our results [48, 49]. In another study, ADPKD patients on RRT still showed a higher survival rate than diabetes-free non-ADPKD patients [50], though such a difference was not observed by other authors [51].

Over the last few years initiation of RRT has been delayed in non-ADPKD patients owing to cardiovascular risk factors being better controlled. ADPKD patients have a lower prevalence of cardiovascular risk factors and also less co-morbidity, which favours longer survival. Still, over time no changes in age at onset of ESRD have occurred among ADPKD patients who have started RRT, probably because of the impact of unmodifiable genetic factors in the absence of a specific treatment. However, there is reasonable hope that disease progression in patients with ADPKD can be delayed with newer therapeutic approaches which are currently being tested.

RT chairs the Inherited Renal Diseases Unit at Fundació Puigvert. She also chairs the working group on inherited kidney disorders within the Spanish Society of Nephrology and is the President of the Scientific Committee for the AIRG-E.