diabetes research and clinical practice 102 (2013) 233–241
Contents available at ScienceDirect
Diabetes Research and Clinical Practice journ al h ome pa ge : www .elsevier.co m/lo cate/diabres
Angiotensin-converting enzyme inhibitors versus angiotensin receptor blockers for end-stage renal disease/mortality in type 2 diabetes§ Heather M. Campbell a,b,*, Nasreen Khan b,c, Dennis W. Raisch a,b, Matthew E. Borrego b, Mike R. Sather a,b, Glen H. Murata d,e a
VA Cooperative Studies Program, Clinical Research Pharmacy Coordinating Center, Albuquerque, NM, United States University of New Mexico College of Pharmacy, Albuquerque, NM, United States c Oxford Outcomes, Morristown, NJ, United States d New Mexico VA Health Care System, Albuquerque, NM, United States e University of New Mexico School of Medicine, Albuquerque, NM, United States b
article info
abstract
Article history:
Aims: To compare angiotensin-converting enzyme inhibitors (ACEIs) and angiotensin re-
Received 11 April 2013
ceptor blockers (ARBs) for end-stage renal disease (ESRD) development and all-cause
Received in revised form
mortality in veterans with macroalbuminuria and with newly documented type 2 diabetes.
31 July 2013
Methods: A retrospective cohort study utilizing data from the national Department of
Accepted 1 October 2013
Veterans Affairs (VA) databases. The study followed 5166 subjects without a history of
Available online 9 October 2013
use of ACEIs or ARBs. To control for differences in baseline characteristics between groups, comparisons of subjects ACEIs and ARBs were made by incorporating propensity scores
Keywords:
analysis into multivariate logistic regression. This resulted in adjusted odds ratios and 95%
Angiotensin-converting enzyme
confidence intervals for ESRD development and all-cause mortality.
inhibitor
Results: The sample was followed up to five years with a mean follow-up of three years.
ACEI
Subjects taking ACEIs has lower odds of ESRD development (OR, 0.33 [95% CI, 0.13–0.82]) and
Angiotensin receptor blocker
all-cause mortality (OR, 0.10 [95% CI, 0.04–0.21]) than ARBs.
ARB
Conclusions: This study shows that ACEIs are associated with lower ESRD development and all-cause mortality than ARBs. This may have implications for guidelines which currently suggest that these two therapeutic classes provide similar benefits in people with newly diagnosed type 2 diabetes and macroalbuminuria. Published by Elsevier Ireland Ltd.
1.
Introduction
Diabetes mellitus is the leading health condition contributing to incident ESRD cases with 45% of individuals with ESRD
having type 2 diabetes mellitus [1]. The incidence and severity of ESRD in people with diabetes mellitus is expected to increase as the U.S. population becomes more diverse. For example, Hispanics and African Americans with diabetes have a higher age-adjusted rate of acquiring ESRD compared with
§ Disclaimer: Contents are expressed by the authors and do not represent the views of the Department of Veterans Affairs or the United States Government. * Corresponding author at: VA Cooperative Studies Program, Clinical Research Pharmacy Coordinating Center, 2401 Centre Avenue SE, Albuquerque, NM 87106, United States. Tel.: +1 505 248 3203; fax: +1 505 248 3205. E-mail address: [email protected] (H.M. Campbell). 0168-8227/$ – see front matter . Published by Elsevier Ireland Ltd. http://dx.doi.org/10.1016/j.diabres.2013.10.005
234
diabetes research and clinical practice 102 (2013) 233–241
non-Hispanic whites with diabetes [2,3]. Additionally, dialysis patients with diabetes have a 27.2% and 23.3% five-year survival rate for hemodialysis and peritoneal dialysis, respectively [1] and have higher mortality rates than people who developed ESRD from different etiologies [4,5]. Albuminuria has been identified as the strongest predictor of ESRD [6] and is a continuous independent predictor for allcause mortality [7]. ACEIs or ARBs reduce albuminuria. At the molecular level, ACEIs and ARBs work by inhibiting angiotensin II from stimulating angiotensin II type 1 (AT1) receptors. ACEIs prevent conversion of angiotensin I to angiotensin II while ARBs compete with angiotensin II for AT1 receptor sites [8,9]. AT1 receptor stimulation increases urinary albumin excretion through vasoconstriction [10,11]. Vasoconstriction also leads to increased systemic pressure, causing higher blood pressure. Therefore, controlling hyperglycemia, blood pressure, and albuminuria aids in slowing the decline of progressive nephropathy. ACEI or ARB monotherapy attenuates progression of albuminuria throughout the disease process [12–16], up to and including the ability to revert to the previous albuminuric state [12,17,18]. Our literature review of PubMed, 1966–2007, did not identify any head-to-head studies comparing ACEI to ARB monotherapy in people with type 2 diabetes mellitus assessing the development of ESRD or all-cause mortality. We found three previous studies that controlled between group differences in baseline albuminuria [19–21]. However, these studies statistically assessed only surrogate endpoints of ESRD (creatinine clearance, albuminuria, glomerular filtration rate), sample sizes were small (approximately 200 subjects), and only one study had therapy duration of at least one year [19]. Although none of the previous studies found a statistically significant difference in albuminuria reduction between the treatment groups, the relative reduction in albuminuria was 4–30% greater for ACEIs compared with ARBs. Perhaps this is the reason the American Diabetes Association guidelines and VA/ DoD guideline state no preferences for ACEI or ARB monotherapy but rather indicate that either therapeutic class can be used in people with type 2 diabetes mellitus to prevent nephropathy progression [22,23]. Thus, we performed a retrospective cohort study to determine whether ACEI and ARB monotherapy of at least 1 year duration was associated with differences in ESRD development or all-cause mortality in people with newly documented type 2 diabetes mellitus in the Department of Veterans Affairs (VA).
index date. As we required at least one year of follow-up for each subject, the last day someone could first seek care for type 2 diabetes mellitus was September 30, 2006. We used the national VA datasets housed at the National Corporate Data Franchise Center, which provide patient information on demographics, outpatient visits, emergency department visits, and hospitalizations for VA-provided services, including diagnoses, prescription utilization, and laboratory results. These datasets are formed by data extraction from electronic medical records. Accordingly, comorbidities, covariates, and development of ESRD were captured with these datasets. An individual was considered to have developed ESRD when at least one code documented dialysis, kidney transplant, or ESRD-related services (e.g., assessing nutrition adequacy, blood urea nitrogen tests). All-cause mortality was captured with a dataset that extracts death information from the National Death Index, Social Security Administration, the Beneficiary Identification Records Locator Subsystem, and VA hospitalizations. To limit bias associated with differences in therapy duration, patients had to be new users of ACEIs or ARBs. A new user was defined as a person not having a prescription filled for an ACEI or ARB in the six months prior to documentation of type 2 diabetes mellitus. Patients also must have had macroalbuminuria at baseline because they were expected to experience higher event rates compared with those with normoalbuminuria or microalbuminuria. We used the standard definition for macroalbuminuria (>300 mg/g or >20 mg/dL). Since the albumin: creatinine ratio is more accurate, we used it preferentially at baseline and in each year of observation for each patient. When there was no value for the ratio, we used the mg/dL value rather than having missing data. We excluded the following patients: age
Contents available at ScienceDirect
Diabetes Research and Clinical Practice journ al h ome pa ge : www .elsevier.co m/lo cate/diabres
Angiotensin-converting enzyme inhibitors versus angiotensin receptor blockers for end-stage renal disease/mortality in type 2 diabetes§ Heather M. Campbell a,b,*, Nasreen Khan b,c, Dennis W. Raisch a,b, Matthew E. Borrego b, Mike R. Sather a,b, Glen H. Murata d,e a
VA Cooperative Studies Program, Clinical Research Pharmacy Coordinating Center, Albuquerque, NM, United States University of New Mexico College of Pharmacy, Albuquerque, NM, United States c Oxford Outcomes, Morristown, NJ, United States d New Mexico VA Health Care System, Albuquerque, NM, United States e University of New Mexico School of Medicine, Albuquerque, NM, United States b
article info
abstract
Article history:
Aims: To compare angiotensin-converting enzyme inhibitors (ACEIs) and angiotensin re-
Received 11 April 2013
ceptor blockers (ARBs) for end-stage renal disease (ESRD) development and all-cause
Received in revised form
mortality in veterans with macroalbuminuria and with newly documented type 2 diabetes.
31 July 2013
Methods: A retrospective cohort study utilizing data from the national Department of
Accepted 1 October 2013
Veterans Affairs (VA) databases. The study followed 5166 subjects without a history of
Available online 9 October 2013
use of ACEIs or ARBs. To control for differences in baseline characteristics between groups, comparisons of subjects ACEIs and ARBs were made by incorporating propensity scores
Keywords:
analysis into multivariate logistic regression. This resulted in adjusted odds ratios and 95%
Angiotensin-converting enzyme
confidence intervals for ESRD development and all-cause mortality.
inhibitor
Results: The sample was followed up to five years with a mean follow-up of three years.
ACEI
Subjects taking ACEIs has lower odds of ESRD development (OR, 0.33 [95% CI, 0.13–0.82]) and
Angiotensin receptor blocker
all-cause mortality (OR, 0.10 [95% CI, 0.04–0.21]) than ARBs.
ARB
Conclusions: This study shows that ACEIs are associated with lower ESRD development and all-cause mortality than ARBs. This may have implications for guidelines which currently suggest that these two therapeutic classes provide similar benefits in people with newly diagnosed type 2 diabetes and macroalbuminuria. Published by Elsevier Ireland Ltd.
1.
Introduction
Diabetes mellitus is the leading health condition contributing to incident ESRD cases with 45% of individuals with ESRD
having type 2 diabetes mellitus [1]. The incidence and severity of ESRD in people with diabetes mellitus is expected to increase as the U.S. population becomes more diverse. For example, Hispanics and African Americans with diabetes have a higher age-adjusted rate of acquiring ESRD compared with
§ Disclaimer: Contents are expressed by the authors and do not represent the views of the Department of Veterans Affairs or the United States Government. * Corresponding author at: VA Cooperative Studies Program, Clinical Research Pharmacy Coordinating Center, 2401 Centre Avenue SE, Albuquerque, NM 87106, United States. Tel.: +1 505 248 3203; fax: +1 505 248 3205. E-mail address: [email protected] (H.M. Campbell). 0168-8227/$ – see front matter . Published by Elsevier Ireland Ltd. http://dx.doi.org/10.1016/j.diabres.2013.10.005
234
diabetes research and clinical practice 102 (2013) 233–241
non-Hispanic whites with diabetes [2,3]. Additionally, dialysis patients with diabetes have a 27.2% and 23.3% five-year survival rate for hemodialysis and peritoneal dialysis, respectively [1] and have higher mortality rates than people who developed ESRD from different etiologies [4,5]. Albuminuria has been identified as the strongest predictor of ESRD [6] and is a continuous independent predictor for allcause mortality [7]. ACEIs or ARBs reduce albuminuria. At the molecular level, ACEIs and ARBs work by inhibiting angiotensin II from stimulating angiotensin II type 1 (AT1) receptors. ACEIs prevent conversion of angiotensin I to angiotensin II while ARBs compete with angiotensin II for AT1 receptor sites [8,9]. AT1 receptor stimulation increases urinary albumin excretion through vasoconstriction [10,11]. Vasoconstriction also leads to increased systemic pressure, causing higher blood pressure. Therefore, controlling hyperglycemia, blood pressure, and albuminuria aids in slowing the decline of progressive nephropathy. ACEI or ARB monotherapy attenuates progression of albuminuria throughout the disease process [12–16], up to and including the ability to revert to the previous albuminuric state [12,17,18]. Our literature review of PubMed, 1966–2007, did not identify any head-to-head studies comparing ACEI to ARB monotherapy in people with type 2 diabetes mellitus assessing the development of ESRD or all-cause mortality. We found three previous studies that controlled between group differences in baseline albuminuria [19–21]. However, these studies statistically assessed only surrogate endpoints of ESRD (creatinine clearance, albuminuria, glomerular filtration rate), sample sizes were small (approximately 200 subjects), and only one study had therapy duration of at least one year [19]. Although none of the previous studies found a statistically significant difference in albuminuria reduction between the treatment groups, the relative reduction in albuminuria was 4–30% greater for ACEIs compared with ARBs. Perhaps this is the reason the American Diabetes Association guidelines and VA/ DoD guideline state no preferences for ACEI or ARB monotherapy but rather indicate that either therapeutic class can be used in people with type 2 diabetes mellitus to prevent nephropathy progression [22,23]. Thus, we performed a retrospective cohort study to determine whether ACEI and ARB monotherapy of at least 1 year duration was associated with differences in ESRD development or all-cause mortality in people with newly documented type 2 diabetes mellitus in the Department of Veterans Affairs (VA).
index date. As we required at least one year of follow-up for each subject, the last day someone could first seek care for type 2 diabetes mellitus was September 30, 2006. We used the national VA datasets housed at the National Corporate Data Franchise Center, which provide patient information on demographics, outpatient visits, emergency department visits, and hospitalizations for VA-provided services, including diagnoses, prescription utilization, and laboratory results. These datasets are formed by data extraction from electronic medical records. Accordingly, comorbidities, covariates, and development of ESRD were captured with these datasets. An individual was considered to have developed ESRD when at least one code documented dialysis, kidney transplant, or ESRD-related services (e.g., assessing nutrition adequacy, blood urea nitrogen tests). All-cause mortality was captured with a dataset that extracts death information from the National Death Index, Social Security Administration, the Beneficiary Identification Records Locator Subsystem, and VA hospitalizations. To limit bias associated with differences in therapy duration, patients had to be new users of ACEIs or ARBs. A new user was defined as a person not having a prescription filled for an ACEI or ARB in the six months prior to documentation of type 2 diabetes mellitus. Patients also must have had macroalbuminuria at baseline because they were expected to experience higher event rates compared with those with normoalbuminuria or microalbuminuria. We used the standard definition for macroalbuminuria (>300 mg/g or >20 mg/dL). Since the albumin: creatinine ratio is more accurate, we used it preferentially at baseline and in each year of observation for each patient. When there was no value for the ratio, we used the mg/dL value rather than having missing data. We excluded the following patients: age
