Angiotensin Receptor Blockers Decrease the Risk of Major Adverse Cardiovascular Events in Patients with End-Stage Renal Disease on Maintenance Dialysis: A Nationwide Matched-Cohort Study

Background Major adverse cardiovascular events (MACE) cause the leading cause of morbidity and mortality in patients with end-stage renal disease (ESRD) on maintenance Hemodialysis (HD) or peritoneal dialysis (PD). Many randomized-controlled trials (RCTs) have proved that angiotensin receptor blockers (ARBs) can reduce the risk of MACE in the people with normal or impaired kidney function without dialysis. This study seeks to clarify whether ARBs therapy could also attenuate this risk in patients with ESRD on maintenance dialysis. Materials and Methods The National Health Research Institute provided a database of one million random subjects for the study. A random sample was taken of 1800 patients ≥18 years y/o with ESRD on dialysis without a history of MACE and use of ARBs within 6-months prior to enrollment. Cox proportional hazard regression analysis was used to identify the risk factors and compute the hazard ratios accompanying 95% confidence intervals. Results In these 1800 patients, 1061 had never used ARBs, while 224 had used them for 1–90 days, and 515 had used them for more than 90 days. We found that ARBs significantly decrease the incidences of acute myocardial infarctions (AMI), coronary artery diseases (CAD) requiring coronary stent or percutaneous transluminal coronary angioplasty (PTCA), peripheral artery disease (PAD) requiring percutaneous transluminal angioplasty (PTA), and acute stroke. Cumulative prescription days of ARBs beyond 365–760 days or more were found to be negatively correlated with incidence of MACEs. For patients with dual comorbidity (i.e., mellitus and hyperlipidemia), 91–365 cumulative prescription days might also attenuate the risk. Conclusions For patients on maintenance dialysis, the use of ARBs could significantly attenuate the risk of major cardiovascular events: AMI, acute stroke, and PAD requiring PTA.


Introduction
At 2,902 per million population, Taiwan's end-stage renal disease (ESRD) rate is the highest in the world [1]. Furthermore, patients with ESRD have a much higher risk of cardiovascular disease [2,3], and are 10-30 times more likely to die of cardiovascular disease than non-dialysis patients of the same age [4,5]. In total, 44% of ESRD patients die from cardiovascular disease [6]. A study in the US found that 63% of patients with chronic kidney disease (CKD) suffer from cardiovascular disease, as opposed to only 5.8% in non-CKD patients [7]. In addition, the incidence of certain cardiovascular diseases increases with CKD severity. For example, the incidence rate of myocardial infarction in stages I and II is 5.3%, but 10.1% in stages III and IV [5]. This high morbidity and mortality of cardiovascular diseases has emerged as a significant challenge in treating ESRD patients.
Hypertension is an important risk factor for cardiovascular disease, and is estimated to afflict up to 80% of the CKD population [8,9]. Angiotensin receptor blockers (ARBs), an effective and well-tolerated orally-active antihypertensive drug, act mainly by blocking the angiotensin II receptor, type 1 (AT 1 ), thereby relaxing the vascular smooth muscle, increasing salt excretion, decreasing cellular hypertrophy and inducing an antihypertensive effect without modifying the heart rate or cardiac output [10,11]. Reports show ARBs have benefits for patients with normal kidney function or CKD on protecting target organs, such as heart [12,13] or kidney [14][15][16]. Among the patients on maintenance dialysis, we do not have adequate evidences to prove whether the ARBs have the same effects, especially on cardiovascular system. We therefore conducted a nationwide, population-based study to clarify whether ARB therapy in chronic dialysis patients could attenuate the risk of major adverse cardiovascular events (MACEs).
The Taiwan National Health Insurance Research Database (NHIRD) is used to clarify whether ARBs could attenuate the risk of MACE in chronic dialysis patients in a five year follow-up (2000)(2001)(2002)(2003)(2004)(2005). As of June 2009, Taiwan's National Health Insurance (NHI) Program covers more than 99% of the population, and 97% of medical providers [17]. Previous studies have found the NHIRD to be a valid resource for medical research [18,19].

Materials and Methods Database
In Taiwan, the patients who need chronic dialysis are made by nephrologists and NHI administration will review then issue a catastrophic illness card to each patient receiving hemodialysis (HD) or peritoneal dialysis (PD).
This population-based cohort study used healthcare data from the Longitudinal Health Insurance Database 2005 (LHID2005), which was randomly sampled from the National Health Insurance Research Database (NHIRD). The NHIRD has prospectively collected data since the implementation of Taiwan's National Health Insurance (NHI) in 1995. It covers both outpatient and inpatient services for approximately 99% of entire 23 million population of Taiwan. The LHID2005 contains 1,000,000 original claims data which were randomly sampled from the year 2005 registry of all beneficiaries under the NHI program. It comprises comprehensive health care data including encrypted patient identification number, demographic data, outpatient/inpatient visits, diagnosis codes and details of prescriptions. Diseases were coded using the ICD-9-CM (International Classification of Diseases, 9th Revision, Clinical Modification) diagnosis codes. Dialysis patients were identified based on diagnosis codes and the Catastrophic Illness Registry. The Catastrophic Illness Registry contains records of patients, who possess catastrophic illness cards issued by NHI administration. The records for chronic dialysis patients receiving either hemodialysis (HD) or peritoneal dialysis (PD) cards are updated every year to ensure the dialysis status. Personal medical information in the database had been encrypted for privacy. There were no statistically significant differences in the distribution of sex, age, number of births per year or health-care costs between patients in the database.

Population
From July 1, 2000 to December 31, 2005 in LHID2005, a total of 2722 ESRD patients were given catastrophic illness cards. As for ESRD diagnosis, we used both diagnosis codes (ICD-9-CM code: 585) and the Catastrophic Illness Registry. Patients were excluded from consideration if they 1) were aged < 18 years, 2) had ARB records within the previous six months as wash-out periods, and 3) had MACE prior to enrollment. The final sample included 1800 chronic dialysis patients.
Patients were divided into ARB user and non-user groups according to ARB prescriptions between follow-ups. For patients who never used ARBs, the index date of the follow-up period was set as the date of beginning dialysis, while that for patients who used ARBs regularly was set as the first ARB prescription date. In ARB user group, we define the treatment of ARBs with 90 or fewer cumulative prescription days as short-term use group (ARB (+)), and those more than 90 cumulative prescription days as long-term use group (ARB (++)). Fig 1 illustrates the population selection process.

Statistical analysis
Differences in distribution of age, gender, and comorbidities between with ARB group and non-ARB group were compared using the chi square test for categorical variables. Kaplan-Meier analysis was used to assess cumulative hazard, with significance based on the log-rank test. Multiple regression analysis was conducted using Cox proportional hazard regression analysis to identify risk factors and compute the hazard ratios (HRs) accompanying 95% confidence intervals (95% CIs). A two-sided P < 0.05 was considered statistically significant. All statistical analyses were performed using SPSS software (Version 19.0, SPSS, Inc., Chicago, IL, USA) and Microsoft SQL Server 2008 (Redmond, Seattle, WA, USA) for data management.

Ethical statement
This study conforms to Declaration of Helsinki protocols, and was approved from a full review by the Institutional Review Board of Cathay General Hospital (permission code: CGH-OP102002). All the data set consists of de-identified secondary data, and was released without restrictions for research purposes. Table 1 shows demographic distributions for the 1800 patients enrolled in this study. Among them, 1061 were in the ARB(-) group (i.e., not treated with ARB at all), 224 were in the ARB(+) group (short-term use group), and 515 were in the ARB(++) group (long-term use group). The sex ratio of ARB(-) group (female 52.6% and male 47.4%) was similar to those in previous studies which showed that female patients had a higher incidence [20,21]. No significant gender difference was found between two groups. However, several comorbidities were more prevalent among the ARB(++) and ARB(+) groups, including hypertension (95.50%, 92.40%, and 50.60%; P<0.001), diabetes (52.80%, 55.40%, and 23.60%; P<0.001), hyperlipidemia (39.20%, 29.90%, and 17.30%; P<0.001), and COPD (21.90%, 24.10%, and 13.10%; P<0.001). ARB users had a higher prevalence of these cardiovascular comorbidities than non-users. The ARB types included losartan, valsartan, irbesartan, candesartan, olmesartan, telmisartan and eprosartan.

Results
During the 5 years follow-up, 851 of the subjects developed incident MACE, including 233 from the ARB(++) group, 121 from the ARB(+) group, and 497 from the ARB(-) group. The Kaplan-Meier curve plot presented in Fig 2 shows the protective effect of long-term ARB use (ARB (++) group) against the MACE occurrence. (Log-rank test P-value < 0.05). Table 2 also shows the ARB (++) group has significantly lower the new incidence of MACE (adjusted HR = 0.68, 95% CI = 0.57-0.82, P < 0.001). However, we cannot find the benefits in the ARB (+) group compared with ARB (-) group in Fig 2 and Table 2.

Discussion
Our aim focuses on the possible effects of MACE protection by the use of ARBs in chronic dialysis patients. The use of angiotensin-converting-enzyme inhibitors (ACEIs) may interfere the results, because both ACEIs and ARBs are the blockades of the renin-angiotensin-aldosterone system. But the use of ACEIs in dialysis patients is not very common due to their disturbing side effects: cough and hyperkalemia. On the other hand, because of lots of generic drugs and brand drugs in Taiwan's medical market, we cannot make a good analysis if we mix these ACEIs. Previous studies suggest that the ARB therapy in patients with a normal kidney function or chronic kidney disease without dialysis might protect heart and kidney from disorders. A randomized controlled trial study showed that ARBs didn't significantly lower the risks of major cardiovascular events or death among patients with hypertension on chronic HD [22]. A metaanalysis study concluded that ARB wasn't associated with a statistically significant reduction in the risk of fatal and nonfatal CV events [23]. An open-label randomized controlled trial study proved ARBs to be effective in reducing nonfatal CVD events in patients undergoing a longterm HD [22]. One study revealed that blockade of rennin-angiotensin system might be effective in heart failure patients on long-term Hemodialysis [24]. This is the first ARB study focused on attenuating the risk of MACE in ESRD patients. In Fig 2, we find the association of ARB use with decreased risk of MACE was significant in the ARB (++) group (more than 90 cumulative prescription days), as compared to the ARB (+) (under 90 cumulative prescription days) or ARB (-) group (no ARB treatment). Furthermore, subdividing the groups by cumulative prescription days shows that longer total treatment duration is crucial for decreasing risk for MACE, with more than 365 treatment days associated with significantly decreased risk for MACE. The use of ARB in patients on maintenance dialysis focuses on blood pressure control. After using ARB, the number of dialysis patients that need to do invasive cardiovascular examination was decreased significantly. However for some reason, such as hyperkalemia, 537 patients with hypertension in this study do not be prescribed ARBs for blood pressure control in Taiwan  (Table 1). We should encourage nephrologists to prescribe ARBs as the first-line medications for blood pressure control. In Table 2, we found that hypertension alone is not an independent factor to the development of MACE in maintenance dialysis patients after adjusted for gender, age, and comorbidities (HR = 1.12, 95% CI = 0.94-1.33, P = 0.196). Besides the hypertension, diabetes mellitus and hyperlipidemia still increase the risk of MACE in these patients (HR = 1.40, 95% CI = 1.21-1.63, P < 0.001; and HR = 1.27, 95% CI = 1.09-1.49, P < 0.05, respectively). Table 3 show the MACE protection by ARBs correlates with the cumulative prescription days of ARBs. In short-term use group ARB (+) there is no difference of MACE occurrences compared with non-user group ARB (-) (HR = 1.00, 95% CI = 0.81-1.23). In long-term use group ARB (++), a significant decrease of MACE was noted when cumulative prescription days of ARB > 365 days. Table 4 show the difference comorbidities influence the effect of ARBs to reduce the risk of MACE. For patients with hypertension only, longer cumulative prescription days (> 1095 days) are needed to reduce the risk (HR = 0.26, 95% CI = 0.08-0.81, P < 0.05). For patients with hypertension and adding diabetes mellitus or hyperlipidemia, cumulative prescription days over 365 days can reach the target to reduce the risk of MACE. However, for patients with hypertension and hyperlipidemia, the benefits vanishes when cumulative prescription days > 1095 days. For patient with hypertension, diabetes mellitus, and hyperlipidemia, benefits of MACE reduction occur earlier after prescription of ARBs over 90 days and these benefits continue.
In Table 5, we find that the long-term use of ARBs can reduce the risks of AMI, acute stroke, PTA for PAD patients, but there are no obvious benefits on patients of CAD for elective PTCA procedures. The mechanisms by which ARBs protect hypertensive dialysis patients from MACE are still not completely clear. Direct blood pressure control is surely a crucial factor [25]. However, other studies have reported several effects independent of blood pressure control, such as modulation of endothelial function or arteriosclerotic plaques [26][27][28]. Some researchers have found that AT 1 -receptor activation could induce superoxide radical release via the NAPDH oxidase pathway, increasing oxidative stress, and thus leading to endothelial dysfunction [29,30]. Therefore, ARBs may play a role in protecting the endothelium by blocking the AT 1 -receptor, thus preventing the formation of endothelial atherosclerotic plaque in the arteries. ARBs could also modulate or decrease atherosclerotic plaque in arteries [31,32]. Protection of the endothelium or modulation of atherosclerotic plaque might contribute to the prevention of AMI and acute stroke. However, this study showed that ARB therapy might not decrease the demand of PTCA for CAD patients. Further studies are needed to clarify why this group of patients seems to not benefit from ARB therapy.
This study is subject to certain limitations. The claim databank provides only the number of ARB prescription days, but contains no data for blood pressure or other risk factors for MACE including obesity, smoking, alcohol consumption, family history, lifestyle, and diet or genetic markers. Although this is an important limitation in our study, we have done comorbidityrelation analysis for MACE as shown in Table 4. The blood pressure records and the comorbidities, such as hypertension, hyperlipidemia and diabetes mellitus, are metabolic risk factors which are related to the factors of obesity, smoking, alcohol consumption, family history, lifestyle, and diet or even genetic markers. Since this study has adjusted these factors, considerable part of these factors should have been adjusted. In addition, we have no way of knowing if patients actually took their prescribed medication regularly. Death certificate data or autopsy reports are not included in the NHIRD dataset, either. Therefore, MACE-related mortality could not be analyzed.

Conclusions
For patients on maintenance dialysis, ARB therapy for hypertension could attenuate the risk of MACE, especially for acute myocardial infarctions, acute strokes, and PAD requiring PTA. More than 365 cumulative ARB prescription days are needed to lower the risk of MACE, except for patients with comorbidities of both diabetes mellitus and hyperlipidemia who benefitted from > 90 cumulative prescription days. Further studies are needed to clarify the ARB mechanisms for risk attenuation for MACE and the effects on MACE-related mortality in maintenance dialysis patients.