Statins and atherosclerotic cardiovascular outcomes in patients on incident dialysis and with atherosclerotic heart disease

Abstract

Background:

Statins failed to reduce cardiovascular (CV) events in trials of patients on dialysis. However, trial populations used criteria that often excluded those with atherosclerotic heart disease (ASHD), in whom statins have the greatest benefit, and included outcome composites with high rates of non-atherosclerotic CV events that may not be modified by statins. Here, we study whether statin use associates with lower atherosclerotic CV risk among patients with known ASHD on dialysis, including in those likely to receive a kidney transplant, a group excluded within trials, but with lower competing mortality risks.

Methods:

Using data from the United States Renal Data System including Medicare claims, we identified adults initiating dialysis with ASHD. We matched statin users 1:1 to statin nonusers with propensity scores incorporating hard matches for age and kidney transplant listing status. Using Cox models, we evaluated associations of statin use with the primary composite of fatal/non-fatal myocardial infarction and stroke (including within pre-specified subgroups of younger age (<50 years) and waitlisting status); secondary outcomes included all-cause mortality, and composite of all-cause mortality, non-fatal myocardial infarction or stroke.

Results:

Of 197,716 patients with ASHD, 47,562 (24%) were consistent statin users from which we created 46,186 matched pairs. Over a median 662 days, statin users had similar risk of fatal/non-fatal myocardial infarction or stroke overall (HR 1.00, 95% CI 0.97, 1.02), or in subgroups [age <50 years (HR = 1.05, 95% CI 0.95, 1.17); waitlisted for kidney transplant (HR 0.99, 95% CI 0.97, 1.02)]. Statin use was modestly associated with lower all-cause mortality (HR 0.96, 95% CI 0.94, 0.98; E-value =1.21) and similarly, a modest lower composite risk of all-cause mortality, non-fatal myocardial infarction or stroke over the first two years (HR 0.90, 95% CI 0.88, 0.91), but attenuated thereafter (HR 0.98, 95% CI 0.96, 1.01).

Conclusions:

Our large observational analyses are consistent with trials in more selected populations and suggest that statins may not meaningfully reduce atherosclerotic CV events even among incident dialysis patients with established ASHD and those likely to receive kidney transplants.

INTRODUCTION

In patients with established atherosclerotic heart disease (ASHD), statins significantly reduce the risk of atherosclerotic cardiovascular (CV) events (1–4). In contrast, three well-conducted randomized controlled trials of statin therapy in patients with end-stage kidney disease (ESKD) on maintenance dialysis have not identified a CV benefit of statins in this population (5–7). These trials included a relatively heterogeneous population of patients on dialysis who may not have had a history of ASHD, the strongest indication for statin therapy; additionally, a large proportion of trial events were statin non-modifiable. Trials of patients on dialysis have also comprised participants of lower risk, impacting the applicability of their findings to the general population (8). Therefore, the role of statins in patients on dialysis with ASHD remains a major gap in the field (9), but additional clinical trials are unlikely.

Existing clinical practice guidelines underscore uncertainty in the overall community about the role of statins in secondary prevention for patients on dialysis. The American College of Cardiology/American Heart Association (ACC/AHA) Secondary Prevention Guidelines universally endorse the use of statin for all such patients (10). The Kidney Disease: Improving Global Outcomes (KDIGO) Guidelines suggest neither initiating nor discontinuing statins in patients at or after dialysis initiation (11), but recommend their use in patients who have received kidney transplants. Specific evaluation in ASHD populations and groups that are more likely to receive a kidney transplant in the future (e.g. younger age or actively waitlisted for transplantation) could help inform guidelines. In this observational comparative study, we evaluate atherosclerotic CV outcomes among patients with ESKD new to dialysis who have ASHD at baseline, with specific evaluation of a priori subgroups who are waitlisted for kidney transplant and of younger age.

METHODS

Study population and timelines

The study cohort for this analysis included adult patients, ≥ 18 years of age with ESKD who initiated dialysis between January 1, 2006 and December 31, 2014 as registered in the United States Renal Data System (USRDS) Standard Analytic Files (12). To identify the patient subgroup who would be dependent on dialysis, we first excluded patients who died, discontinued dialysis for any reason, including recovered renal function, and those who received a kidney transplant, within the first 180 days after dialysis initiation. To ensure complete ascertainment of comorbidities, medications and outcomes, all included patients utilized Medicare Parts A and B as the primary payer, and had continuous Part D coverage from day 90–180 after dialysis initiation. All patients had ASHD indicated by either the Center for Medicare and Medicaid Services (CMS) Form 2728 or an administrative claim for ASHD up to day 180 after dialysis initiation (outlined in Supplemental Table 1). The first 180 days from dialysis initiation comprised the baseline period during which we ascertained patient comorbidities. Days 90 – 180 from dialysis initiation comprised the exposure period during which we defined statin use. Day 181 after dialysis initiation through to December 31, 2015 comprised the follow up period. Our study protocol was approved by the Duke University Health System IRB and data use was approved by the USRDS. We report our study design and findings in accordance to the reporting guidelines for propensity score analysis (13).

Baseline comorbidities, exposure and outcome ascertainment

Patients on dialysis have a high burden of both CV and non-CV comorbidities that may influence statin prescription patterns and confound association with long-term outcomes. We used a validated comorbidity scoring approach and index to describe and adjust for prognostically significant comorbidity differences in patients on dialysis (14). Comorbidities were ascertained as conditions indicated on CMS Form 2728 and from administrative claims using ICD-9-CM and CPT coding (Supplemental Table 2). Additional covariates including demographics, cause of renal failure, body mass index, prior nephrology care, smoking status, and functional status were ascertained from CMS Form 2728. Dialysis modality was ascertained from the USRDS modality files, waitlisting status from the KI_waitlist and KP_waitlist files, and selected medication classes from Part D data as previously described (15).

The primary exposure in this study was statin use at dialysis initiation and defined as the use of a statin medication for ≥80% outpatient days covered between days 90 to 180 post dialysis initiation using common medication names and Medicare Part D prescription fills (Supplemental Table 3). Hospitalized days were removed from the denominator during the statin ascertainment window because it is assumed that during the in-patient stay, the patient would not be using their home statin supply. New users of statin drugs could not be consistently determined because of lack of Medicare Part D coverage prior to the exposure window in most patients.

The primary outcome of this analysis was a CV composite of fatal or non-fatal myocardial infarction, or fatal or non-fatal stroke. Sudden cardiac death was not included because the pathogenesis in ESKD may not be mediated by atherosclerotic disease and it may not be statin-modifiable. Secondary outcomes included all-cause mortality and a composite of all-cause mortality, non-fatal myocardial infarction or stroke.

Outcomes were ascertained using a combination of ICD-9-CM or ICD-10-CM codes and the Death Notification form, CMS Form 2746, as described in Supplemental Table 4. For the primary outcome, patients were right censored during the follow-up period for non- myocardial infarction or stroke-related death, kidney transplantation, loss of Medicare as primary payer, or at the end of observation. For the secondary outcome of all-cause mortality, patients were right censored for kidney transplantation, or at the end of observation.

Statistical Analysis

Propensity score matching

This analysis was based on an intention to treat principle categorized by statin use during the exposure window. To balance patient characteristics, propensity score matching was used. Propensity scores were estimated using a non-parsimonious multivariable logistic regression model with baseline variables outlined in Supplemental Table 5. Models included fine categorization of covariates to account for potential nonlinear relationships between continuous exposures and propensity. Indicator variables were used to handle missing data, which was limited to <3 % for all included variables. We had an a priori interest in age and kidney transplant waitlist status as subgroups and therefore these two variables were hard-matched. We used a greedy nearest neighbor 1:1 matching without replacement and with a caliper width of 0.2 of the pooled standard deviation of the logit of the propensity score. Standardized differences for all covariates were evaluated pre- and post-matching.

Kaplan Meier curves and Cox proportional hazards models incorporating robust standard error estimates to account for correlation among the matched groups was then used to evaluate outcomes. Younger age (<50 years) and presence on the kidney transplant waitlist were pre-specified and additional outcome analyses were performed including main and interactive effects of these variables. The proportional hazards assumption was evaluated in all models and showed modest time-dependence for the secondary endpoints of all-cause mortality, myocardial infarction or stroke outcome. All P-values were two-sided, and a value of <0.05 was considered significant. All analyses used SAS version 9.4 (SAS Institute, North Carolina).

Sensitivity Analyses

To ascertain the impact of potential unmeasured confounding, we report E-value. The E-value provides an estimate of the minimum strength of association required to fully explain away the observed association between statin use and an outcome. Smaller E-values suggest little unmeasured confounding could generate the observed estimate (16).

RESULTS

Statin Use, Propensity Scores and Balance

The derivation of our analytic cohort is described in Figure 1. Approximately one of every four (n= 47,562/197,716) eligible incident dialysis patients with established ASHD were on statin therapy with ≥80% outpatient days covered by Medicare Part D in the 90–180 days after dialysis initiation. Compared with patients not on statin therapy, statin users were more likely to be older (≥ 65 years age), white, and have a greater burden of clinical comorbidity. They were also more likely than not to be under the care of a nephrologist, and concomitantly on an angiotensin converting enzyme inhibitor/angiotensin receptor blocker or beta-blocker therapy (Table 1). No difference in statin use was evident across dialysis modalities or in patients waitlisted for kidney transplant.

Figure 1: Derivation of the analytic cohort.

Cohort is built from the total adult population initiating dialysis for end-stage kidney disease (ESKD) in the United States between 2006–2014, according to the United States Renal Data System Standard Analytic Files. Participants had to remain on hemodialysis 180 days after initiation with Medicare Parts A and B as the primary payer and enrollment in Part D during 90–180 days after dialysis initiation.

Table 1:

Baseline patient characteristics before and after propensity score matching

Variable % or median (IQR)	Overall (n=197,716)	Statin use before propensity score matching		Std. diff	Statin use after propensity score matching		Std. diff
		Yes (n=47,562)	No (n=150,154)		Yes (n=46,186)	No (n=46,186)
Age (years)	66 (54,75)	69 (61,77)	65 (52,74)	0.409	69 (61,77)	69 (61,77)	0.015
Age categories (years)				0.452			0
<30	2.5	0.4	3.1		0.4	0.4
30 – 39	5.2	1.9	6.3		1.9	1.9
40 – 49	10.3	6.1	11.7		6.1	6.1
50 – 54	7.9	5.8	8.5		5.8	5.8
55 – 59	9.4	8.0	9.9		7.9	7.9
60 – 64	9.6	9.6	9.6		9.7	9.7
65 – 69	15.5	18.7	14.5		18.7	18.7
70 – 74	13.8	17.9	12.5		17.7	17.7
75 – 79	11.7	14.8	10.7		15.0	15.0
80 – 84	8.6	10.8	8.0		10.8	10.8
≥ 85	5.5	6.1	5.3		6.1	6.1
Female sex	48.6	51.6	47.7	0.080	51.6	51.6	0.002
Race				0.223			0.018
White	63.2	69.6	61.2		69.5	70.1
Black	30.9	23.7	33.1		23.9	23.6
American Indian/Alaska Native	1.2	0.9	1.3		0.9	0.8
Asian	3.7	4.8	3.3		4.7	4.4
Native Hawaiian/Pacific Islander	0.8	0.9	0.7		0.9	0.9
Other	0.3	0.2	0.3		0.2	0.2
Body mass index	27.9 (23.8, 33.7)	28.7 (24.5, 34.5)	27.6 (23.5, 33.4)	0.108	28.7 (24.5, 34.4)	28.4 (24.2, 34.3)	0.002
Diabetes mellitus	67.2	76.4	64.3	0.267	76.3	75.8	0.011
Current smoking	7.2	5.8	7.7	0.074	5.8	5.8	0.002
Hypertension	89.3	90.6	89.0	0.539	90.6	90.3	0.010
Primary cause of kidney failure				0.247			0.014
Diabetes mellitus	48.6	56.8	46.0		56.6	56.0
Hypertension/large vessel disease	30.3	28.2	30.9		28.3	28.8
Primary glomerulonephritis	5.6	4.3	6.0		4.3	4.4
Other	15.6	10.7	17.1		10.7	10.9
Dialysis type				0.065			0.005
In-center hemodialysis	92.8	91.5	93.2		91.6	91.4
Home hemodialysis	0.3	0.3	0.3		0.3	0.3
CAPD/CCPD/Other	6.9	8.2	6.5		8.1	8.3
Comorbidity index	4 (2, 8)	5 (3, 9)	4 (1, 7)	-	5 (3, 9)	5 (3, 9)	-
Comorbidity category				0.236			0.015
0–1	24.0	17.7	25.9		17.5	17.1
2–3	17.2	15.8	17.7		15.8	16.0
4–6	25.4	26.5	25.1		26.6	26.9
7–9	18.9	22.5	17.8		22.5	22.2
>10	14.5	17.5	13.5		17.6	17.8
Nephrologist care prior to dialysis	62.4	71.3	59.5	0.259	71.3	71.8	0.012
Inability to ambulate	6.9	6.5	7.0	0.021	6.5	6.5	0.001
Inability to transfer	3.3	2.8	3.5	0.041	2.7	2.7	0.003
Institutionalized	8.6	8.3	8.8	0.018	8.2	8.2	0.001
Need assistance with activities of daily living	13.1	14.0	12.8	0.036	13.9	14.0	0.003
ACE-I/ARB	19.1	31.2	15.3	0.384	30.5	30.1	0.008
Beta-blocker	30.6	51.7	24.0	0.595	51.2	51.3	0.002
Waitlisted for transplant	4.6	4.9	4.5	0.018	4.7	4.7	0

ACE-I, angiotensin converting enzyme; ARB, angiotensin receptor blocker; CAPD, continuous ambulatory peritoneal dialysis; CCPD, continuous cycling peritoneal dialysis;

To assess the validity of assignment of statin use at baseline, we reassessed statin adherence over available follow up in our cohort. We observed that approximately two-thirds of statin non-users had no additional Part D claims for statins over follow up. Similarly, in statin users approximately two-thirds of patients had part D claims for statins for ≥ 50% outpatient days in follow up. These statin adherence results were consistent in the study cohorts prior to- and after propensity score matching (Supplemental Tables 6 and 7, respectively).

Of the 47,562 patients on statin therapy, 46,186 statin users were 1:1 propensity score matched to statin non-users (Table 1). The distribution of the estimated propensity for statin use in patients on a statin compared with those not on a statin is illustrated in Figure 2 and demonstrates a large overlap in the propensity scores between statin users and non-users, highlighting the uncertainty in clinical practice which allows for more generalizable interpretations of our inferences. Following propensity score matching, the balance in baseline comorbidity across the two statin use categories (Table 1 and Supplemental Figure 1) demonstrates relatively small imbalances in key variables, with standardized differences <10% for all tested variables (Supplemental Figures 2-4).

Figure 2: Distribution of propensity scores for statin use.

A histogram of scores is presented for statin users (blue) and statin non-users (red), with areas of overlap indicated in purple. The large overlap between the two propensity distributions indicates equipoise with similar individuals receiving different therapies over the full range of scores.

Primary and secondary outcomes

Over a median of 622 days (interquartile range spanning 299 to 1224 days) in the propensity matched population, statin use was not associated with the composite risk for fatal or non-fatal myocardial infarction or stroke (11,359 vs 10,743 events, hazard ratio (HR) 1.00, 95% confidence interval (CI) 0.97, 1.02, Figure 3). Statin use was associated with a lower risk for all-cause mortality (15,274 vs 14,877 events, HR 0.96, 95% CI 0.94, 0.98). The relatively small E-value= 1.21 suggests that little unmeasured confounding would be needed to produce this association between statin use and all-cause mortality. Statin use compared with non-use was associated with a similarly lower risk for the composite of all-cause mortality, non-fatal myocardial infarction or stroke over the first two years of follow-up (HR 0.90, 95% CI 0.88, 0.91) and attenuated thereafter (HR 0.98, 95% CI 0.96, 1.01) (Supplemental Figure 5).

Figure 3: Event-free survival for the composite of fatal and non-fatal myocardial infarction or stroke by statin use.

Curves are derived using a Kaplan-Meier estimator for the matched statin user and non-user population. P-value is from the log-rank test.

Subgroup analyses

Regardless of statin use, younger age (<50 years) and active kidney transplant waitlist status correlated with a lower risk of the primary composite (<50 years versus ≥ 50 years, p<0.0001, Figure 4A; waitlisted versus non waitlisted, p<0.0001, Figure 4B). However, the association between statin use and the risk of the primary composite did not differ across age groups (p-interaction = 0.33, Figure 4A) or waitlist status (p-interaction = 0.09, Figure 4B). Evaluating all 4 generated subgroups there were no associations between statin use and the primary outcome (<50 years and not waitlisted: HR 1.04, 95% CI 0.95, 1.17; ≥50 years and not waitlisted: HR 0.99, 95% CI 0.96, 1.02; <50 years and waitlisted: HR 1.19, 95% CI 1.00, 1.43; ≥50 years and waitlisted: HR 1.13, 95% CI 0.97, 1.32).

Figure 4: Event-free survival for the composite of fatal and non-fatal myocardial infarction or stroke by statin use and subgroups of age (A) and kidney transplant waitlist status (B).

Curves are derived using a Kaplan-Meier estimator for the matched statin user and non-user population with hard matches for age and waitlist status. P-value for main effects of age or waitlist status, and interactive effects with statins are from Cox proportional hazards models. (Young, <50 years age, old ≥50 years; WL, waitlisted for kidney transplant, NWL, not waitlisted for kidney transplant)

DISCUSSION

In this comparative effectiveness analysis, we did not find an association between statin use and risk of atherosclerotic CV outcomes in a population of patients with incident ESKD on dialysis and established ASHD. We found similar results in subgroups defined by younger age and active kidney transplant waitlisting. At dialysis initiation, only one in four patients with established ASHD were consistently using statin therapy based on Medicare Part D claims between 90–180 days; the excellent overlap in propensity scores among statin users and non-users suggests equipoise in the field. This real-world analysis extends prior negative results from clinical trials in heterogeneous patient populations with ESKD on dialysis to a secondary prevention population with ASHD.

Despite the high cardiovascular risk in patients on dialysis, statin trials in ESKD populations have consistently demonstrated an absence of incremental risk reduction related to statin therapy (5, 6). The combination of simvastatin and ezetimibe reduced the risk of a major atherosclerotic CV events in the combination of patients with dialysis and non-dialysis dependent chronic kidney disease, however the outcomes specifically in those on dialysis were not significant (7). For several reasons, consideration is required in generalizing the findings of these trials to all patients on dialysis. First, while the benefit of statins is greatest in patients with documented ASHD (1, 4), a large proportion of these patients were excluded from the statin trials in ESKD populations. For instance, the SHARP trial universally excluded patients with baseline coronary artery disease (CAD)(17), while the 4D trial excluded patients with a myocardial infarct in the preceding 3 months (5). Additionally, both the 4D and AURORA trials excluded patients on a statin in the 3–6 months preceding eligibility (5, 6) and as such, only 20–30% of patients on dialysis collectively enrolled within 4D, AURORA and SHARP trials had established ASHD at baseline. Second, the primary outcomes in 4D, AURORA, and SHARP each included a disproportionately large burden of non-atherosclerotic events. SHARP modified its primary outcome after initiating the trial because of this finding. Considering both the relatively low patient proportion with established ASHD, and the fact that up to one-third of all primary outcome events were non-atherosclerotic, statin trials in ESKD populations were likely underpowered to detect statin-related differences for atherosclerotic events in patients with baseline ASHD.

Outside of ESKD, statin related benefits accrue over the long term and traditionally well beyond the median life expectancy of patients on dialysis (18–20). However specific dialysis subgroups such as younger patients and those waitlisted for kidney transplant are healthier and may have longer anticipated survivals. Both AURORA and SHARP trials excluded patients who anticipated receipt of a kidney transplant in the near-term, making it challenging to generalize to this important group (6, 17). Subsequent trials have demonstrated that ESKD patients benefit from statins after receiving a kidney transplant (21, 22). For this reason, guidelines recommend continuing or re-initiating statins in these patient’s post-transplant, but little data is available to guide their pre-transplant management (11).

Several other smaller observational studies have been conducted and described beneficial associations between statin use and some CV outcomes, including in those transitioning to dialysis (23–26). However, in many of these studies, the primary outcomes included general CV mortality that encompassed a spectrum of non-statin modifiable deaths. Our results also found marginal association between statins and lower risk of all-cause mortality and the composite of all-cause mortality, non-fatal myocardial infarction or stroke. These observed associations may in part relate to residual confounding (as highlighted by the small E-value) in which participants at high risk of non-atherosclerotic mortality are not treated with statins. Another possibility is a modest protective effect that could have been underestimated from drug cross-overs and misclassification. Importantly however, our specific evaluation of atherosclerotic CV events among those with baseline ASHD appear to support negative trial findings.

The lack of statin benefit on atherosclerotic CV outcomes observed in patients with incident ESKD on dialysis and with established ASHD could be explained by a combination of factors. First, declining renal function associates with deranged metabolic, inflammatory and mineral homeostasis resulting in accelerated atherosclerosis and vascular calcification. It is possible that by the time ESKD is established, and despite statin-related LDL-lowering, the burden of vascular calcification and vascular remodeling is statin non-modifiable (27). Second, only about 20–30% of all patients in our study were concomitantly on an angiotensin converting enzyme inhibitor (ACE-I)/angiotensin receptor blocker (ARB) or a beta-blocker. While the sub-optimal use of these established secondary prevention pharmacotherapies certainly predisposes to recurrent atherosclerotic CV events, their lower than anticipated use could well relate to patients’ inability to tolerate these medications from underlying multimorbidity, such as hypotension, orthostasis, poor functional status, and hyperkalemia. Half of our study population had a comorbidity index of at least 4, and nearly 15% were unable to independently perform activities of daily living. This “frail” phenotype likely reflects the global vascular and non-vascular burden of disease in patients on dialysis, and regardless of statin use, contributes to risk of atherosclerotic CV outcomes. Third adverse cardiovascular outcomes in patients on dialysis are influenced by concomitant risk profiles (and not exclusive to) such as glycemic control, severity of coronary disease, hypertension, and obesity. We did not have information on treatment of these factors, and thus other treatments or comorbidities could confound our non-randomized analysis.

Patients on dialysis with ASHD are often co-managed in multidisciplinary teams. However, the equipoise around statin use in patients with established ASHD has resulted in non-uniform recommendations between the KDIGO and ACC/AHA guidelines. This uncertainty is reflected in the large overlap in the propensity for statin use among users and non-users, an indicator of substantial practice variation in current clinical practice (Figure 2). Our findings from a large, well characterized, comparative effectiveness analysis are clinically important as dedicated statin trials in patients on dialysis with ASHD specifically evaluating atherosclerotic CV outcomes are now unlikely. Moreover, with efforts aimed at limiting polypharmacy in this patient population, these results align with the statin trials in ESKD populations in suggesting a limited role for continued statin therapy in this patient population.

Study Limitations

Our results need to be considered in the context of the following limitations. First, while the study cohort was constructed to include patients with a diagnosis of ASHD, this was based on an ASHD claim on CMS Form 2728 or a claim for coronary revascularization within the first six-months of dialysis initiation. It is therefore possible that some patients may have had an ASHD diagnosis or coronary revascularization prior to dialysis initiation that was not reported on CMS Form 2728. Second, we identified statin use in a 3-month window and evaluated outcomes on an intention-to-treat basis; however, while we describe general stability in the classification of statin use/adherence over follow-up, we performed an intention-to-treat analysis and did not evaluate changing exposure over time. Additionally, we did not evaluate statin intensity or different formulations that could theoretically have different pharmacokinetics in dialysis patients. With limited overall effectiveness of statins, it is not likely that there were subgroups with higher intensity dosing that benefitted meaningfully. All three statin trials in patients on dialysis have compared low-moderate potency statin dosing with placebo (28), and it is likely that most patients included in this analysis were on low-moderate potency statin dosing. Third, we do not include the need for coronary revascularization within our evaluated outcomes. This was primarily due to our pre-specified concern regarding cardiac evaluation in patients listed for kidney transplantation. These patients often undergo coronary evaluation and optimization as part of their transplant evaluation, which could induce a diagnostic bias. Fourth, the median duration of follow-up in our analyses is much shorter compared with clinical trials of statins in patients on dialysis; the longer-term outcomes of statins in patients with ASHD on dialysis are therefore unknown. Finally, despite detailed assessment of comorbidity and metrics of functional status, unmeasured confounding and bias is possible and could potentially influence the observed differences between outcome composite evaluating cause-specific compared with all-cause mortality. Our use of E-values in interpretation is meant to temper conclusions about marginal associations that may be driven by a small degree of uncontrolled confounding.

CONCLUSION

In patients with ESKD and ASHD incident to dialysis, statin users and non-users experienced similar risk of a fatal or non-fatal MI or stroke. These findings similarly extend to younger patients and those waitlisted for kidney transplant. While modest associations with all-cause mortality are observed, the primary results of this non-randomized analysis provide further evidence to re-consider the role of statins in the secondary prevention of atherosclerotic CV events even in patients with established ASHD on dialysis.

HIGHLIGHTS.

• In patient’s incident to dialysis and with established atherosclerotic heart disease, statins do not appear to associate with a lower atherosclerotic risk of fatal or non-fatal myocardial infarction or stroke.

• Aligned with clinical trial results, and now expanding to patients with known atherosclerotic heart disease (including younger age and those waitlisted for kidney transplant), the secondary atherosclerotic prevention role of statins in patients on dialysis needs to be re-considered.