Angiotensin-Converting Enzyme Inhibitors and Amyotrophic Lateral Sclerosis Risk: A Total Population–Based Case-Control Study

Importance  Although several studies have shown that use of angiotensin-converting enzyme inhibitors (ACEIs) potentially decreased amyotrophic lateral sclerosis (ALS) risk in animal models, to our knowledge, there has been no human study in the literature discussing this issue.

Objective  To investigate the association between the use of ACEIs and the risk for developing ALS.

Design, Setting, and Participants  This case-control study was conducted using the total population of Taiwanese citizens seen in general medical practice; therefore, the findings can be applied to the general population. The case group comprised 729 patients with newly diagnosed ALS and a severely disabling disease certificate between January 1, 2002, and December 31, 2008. These cases were compared with 14 580 sex-, age-, residence-, and insurance premium–matched control individuals.

Exposures  Use of ACEIs was analyzed using a conditional logistic regression model that controlled for other antihypertensives, aspirin, steroids, nonsteroidal anti-inflammatory drugs, Charlson Comorbidity Index score, length of hospital stay, and number of outpatient visits. The cumulative defined daily dose (cDDD), which indicates the exposed duration of drug use, was estimated as the sum of dispensed DDD of drug and compared with the risk for ALS.

Main Outcomes and Measures  All patients with ALS fulfilled El Escorial criteria in this study. Medical claim data past 1 to 5 years of ALS first diagnosis date for patients and claim data from their matched control individuals were included in the analysis.

Results  There was a dose-dependent inverse association between ACEI use and the risk for developing ALS. When compared with patients who did not use ACEIs, the adjusted odds ratios were 0.83 (95% CI, 0.65-1.07; P = .15) for the group prescribed ACEIs lower than 449.5 of the cDDD and 0.43 cDDD (95% CI, 0.26-0.72; P = .001) for the group with a cumulative ACEI use of greater than 449.5 cDDD. The association was most predominant in men older than 55 years.

Conclusions and Relevance  Use of ACEIs exhibited a dose-dependent inverse association with ALS. This study demonstrated a 57% risk reduction in the chance for developing ALS in people who used ACEIs greater than 449.5 cDDD in 4 years.

Amyotrophic lateral sclerosis (ALS) is the most common form of motor neuron disease. It causes degeneration of both lower motor neurons in the brainstem and spinal cord and upper motor neurons in the cerebrum, thereby resulting in the progressive atrophy of associated muscle tissues and supporting cells. The incidence of ALS is higher among white individuals than among those of African, Asian, and Hispanic races/ethnicities.¹ The lifetime risk for developing ALS by age 70 years is between 1 in 400 and 1 in 1000; in general, most patients with ALS die within 3 to 5 years after the onset of symptoms.²

The only medication used for treatment, riluzole, improves the quality of life and survival of patients with ALS,^3-5 although it is not a cure. Some researchers aim to find compounds that might be beneficial in ALS. Based on the characteristic neuroinflammation of ALS, some agents, such as aspirin or nonsteroidal anti-inflammatory drugs (NSAID), have been used to try to decrease the risk for ALS. However, the results from 5 prospective cohort studies did not support an overall effect of NSAIDs or aspirin on ALS risk.⁶ On the other hand, neuronal apoptosis may be one cause of neurodegenerative diseases including Alzheimer disease, Parkinson disease, and ALS.⁷ Angiotensin-converting enzyme inhibitors (ACEIs) are widely prescribed for the treatment of hypertension and congestive heart disease. Studies have shown that ACEIs may be neuroprotective and might decrease the risk for developing neurodegenerative diseases.^8-11 In addition, Iwasaki et al¹² demonstrated that temocapril, an ACEI, had neurotrophic properties in damaged motor neurons of rats, indicating that ACEIs may be neuroprotective in motor neuron disease. To our knowledge, data on ACEI treatment in human motor neuron disease are absent. Therefore, we conducted a total population–based case-control study to investigate the association of ACEIs and ALS risk in Taiwan.

In 1995, the National Health Insurance (NHI) program, a government-run insurance program with a single-payer insurance system, was established in Taiwan. By December 2010, more than 23 million people were enrolled nationwide, with a coverage rate of 99.6%. The registration of all cases of severely disabling diseases (SDDs), such as chronic renal failure, myasthenia gravis, cancer, and ALS, are required by the NHI bureau before SDD certification can be granted. In 2008, 37 099 medical doctors and 553 neurology specialists were registered in Taiwan. In addition, 790 621 people had SDD certificates in 2008, which constituted 3.4% of the total population.

This was a total population–based case-control study. The NHI Research Database, containing outpatient, ambulatory, hospital inpatient care, and dental records, was used in this study. Amyotrophic lateral sclerosis cases were identified according to the International Classification of Diseases, Ninth Revision (ICD-9) code 335.20. The study period that included patients with incident ALS (2002-2008) was dependent on all SDD and NHI Research Database cases published since 1996. The medical claim data from 1996 to 2001 were used to verify that our patients with ALS were new cases. The diagnosis of ALS was based on El Escorial criteria and was made by a supervising clinical neurologist, and the medical records of patients diagnosed as having ALS were sent to the NHI bureau. A group of neurologists in the NHI bureau also used El Escorial criteria to verify the diagnosis of ALS in the medical records.^13-15 All patients with ALS fulfilled El Escorial criteria in this study. Only patients with ALS and SDD certification were included. Patients with SDD certificates are eligible for exemption from insurance premiums and copayments. The approval of SDD certificates requires strict evaluation by the Department of Health, Executive Yuan, in Taiwan. In this study, all ALS cases were verified by linking encrypted identification numbers with SDD certificates. Overall, 729 new ALS cases from January 1, 2002, to December 31, 2008, were included in this study. These cases were compared with 14 580 sex-, age-, residence-, and insurance premium–matched control individuals. Residence was categorized as rural or urban. The insurance premium was a proxy indicator of economic status and was classified into 1 of 4 categories: fixed premium and dependent, less than Taiwanese new dollar (NT$) 20 000 monthly (<US $657.28), NT $20 000 to NT $39 999 monthly (US $657.28-$1314.52), and NT $40 000 or more monthly (≥US $1314.56). The fixed premium group consisted of people receiving social welfare support and included low-income people and veterans. The dependent insurance premium group consisted of people with family members who did not have a job or income. The institutional review board of Kaohsiung Medical University approved the study protocol without the need for patient consent because data for analysis were obtained from the NHI Research Database without personal identification.

The χ² test or t test was used to examine differences in demographic and clinical characteristics between patients with newly diagnosed ALS and control individuals. Use of ACEIs was analyzed using a conditional logistic regression model that controlled for other antihypertensives, aspirin, steroids, NSAIDs, Charlson Comorbidity Index score, length of hospital stay, and number of outpatient visits. To evaluate whether the drug was an independent factor for ALS, we excluded the past 1 year of drug use before first diagnosis of ALS. The study period for the Charlson Comorbidity Index score, length of hospital stay, and number of outpatient visits was 1 year prior to first diagnosis with ALS. The defined daily dose (DDD) recommended by the World Health Organization is a unit for assessing the standard dose of drug; that is, the dose for a 70-kg adult in a day would be called 1 DDD. For example, 1 DDD for captopril (an ACEI) is 50 mg and for enalapril is 10 mg. Cumulative DDD (cDDD), which indicates the exposed duration of drug use, was estimated as the sum of dispensed DDD of drug and compared with the risk for ALS. Drug use was classified into 1 of 3 categories: 0, quartiles 1 through 3, and beyond quartile 3. The cutoff point for the third quartile was the third quartile of cDDD in our study group. A similar definition of drug dose use was also included in another study.¹⁶ The model was tested first using the total group of patients and then using subgroups according to sex and age (15-54 years and >55 years). Analyses were carried out using SAS version 9.3 (SAS Institute Inc).

To validate the robustness of the main study findings, we performed several sensitivity analyses. First, we defined the use of drug as any 1 use vs no use.

Second, instead of overall ACEI use in the main analysis, we analyzed the association between individual ACEI and ALS risk. Because of the small sample size, the individual ACEI analysis defined the use of individual ACEI as any 1 use vs no use.

Third, we used propensity score matching to further control for sampling bias. The propensity scores are patients’ probability of receiving ACEIs and were constructed using a multivariate logistic regression model. The potential confounding factors that were used in the calculation of propensity scores included the year of index date, sex, age, residence, insurance premium, hyperlipidemia status¹⁷ (ICD-9 code 272), hypertension (ICD-9 code 401), ischemic heart disease (ICD-9 codes 410-414), chronic kidney disease (ICD-9 code 585), diabetes mellitus (ICD-9 code 250), stroke (ICD-9 codes 433-438), congestive heart failure (ICD-9 code 428), and atrial fibrillation (ICD-9 code 427.31). The index date was defined as the first prescription date of ACEIs and a randomly selected date among non-ACEI patients. Amyotrophic lateral sclerosis cases and control individuals were matched 1:20 by the estimated propensity score as close as possible.

Finally, because the ALS diagnosis date could not reflect the beginning of disease, calculating the cumulative period until the diagnosis may have included the latent period, which might have biased our estimates. Therefore, we subtracted 1 year from the date of ALS diagnosis for cases and their matched control individuals. The actual latent period of ALS is unclear. To address the influence of different lengths of the latent period, we assessed the ACEI exposure using 2 years and comparing the results with the 1-year latent period, as well as no latent period.

Results for the demographic and clinic characteristics of patients are summarized in Table 1. Because of the exact-matching design, age, sex, residence, and insurance premium distributions of the ALS and non-ALS groups were equivalent. The mean (SD) age of the patients with ALS was 57.43 (13.09) years, which was identical to that in control individuals. Among the 729 new patients with ALS, 451 were men and 278 were women (male to female ratio of 1.6:1). One-fifth of patients lived in rural areas. Using the proxy index of economic status (insurance premium), nearly half of the cases were listed as receiving social support or as a dependent member of a family. Approximately 15% of patients with ALS reported ACEI use between 2 to 5 years prior to first diagnosis of ALS. In the same matched study period, approximately 18% of patients without ALS reported ACEI use. The NSAID use of patients with ALS was significantly higher than that of non-ALS cases (P = .03). The use of other antihypertensives, aspirin, and steroids was similar in patients with ALS and without ALS. The mean Charlson Comorbidity Index score for patients with ALS was higher than that of non-ALS cases, and the services of hospitalization and number of outpatient visits were higher in patients with ALS than non-ALS cases.

The results for the associations of interest are summarized in Table 2. The unadjusted conditional logistic regression model showed significant factors associated with ALS incidence including use of low-dose ACEIs, low-dose aspirin, high-dose steroids, and high-dose NSAIDs; high Charlson Comorbidity Index score; high number of days spent in the hospital; and high number of outpatient visits. In the fully adjusted model, the final independent predictors were similar to the unadjusted results with the exception that steroid use had a nonsignificant effect on the risk for ALS. When compared with the no-ACEIs group, the adjusted odds ratios were 0.83 (95% CI, 0.65-1.07; P = .15) for the group that was prescribed ACEIs less than 449.5 cDDD and 0.43 (95% CI, 0.26-0.72; P = .001) for the group that was prescribed ACEIs greater than 449.5 cDDD.

The effects of ACEI use on ALS risk, calculated by analyzing ACEI use as a binary variable (ACEI use vs no use), was similar to that in the main findings in this study. When compared with the no-ACEIs group, the adjusted odds ratio was 0.74 (95% CI, 0.26-0.72; P < .001) for the group that was prescribed ACEIs. Moreover, analysis conducted according to subgroup revealed that the association was predominant in men older than 55 years (Table 3).

The adjusted association between individual compounds of ACEIs and ALS risk is shown in Table 4. Captopril and enalapril were negatively associated with ALS risk. The other individual ACEIs in our database had no statistical significance.

We also performed sensitivity analyses with different latent periods and using the propensity score–matching method. The distributions of potential confounding factors were similar between the ALS and non-ALS groups following propensity score matching (P > .05; Table 5). The results based on this sensitivity analysis also showed inverse association between ACEI use and ALS risk that was similar to our main analysis (Table 6).

The findings in this total population–based case-control study revealed that long-term exposure to ACEIs was inversely associated with the risk for developing ALS. To our knowledge, the present study is the first to screen the association between ACEIs and ALS risk in a population-based study. Using data from the total population of Taiwan, our findings indicated that the incidence of ALS was inversely associated with ACEI use when we controlled for the use of other antihypertensives, aspirin, steroids, and NSAIDs; Charlson Comorbidity Index score; length of hospital stay; and number of outpatient visits.

Angiotensin-converting enzyme inhibitors have been found to be beneficial in some neurodegenerative diseases such as Alzheimer dementia and Parkinson disease. An observational case-control study showed that the use of ACEIs was associated with a reduced rate of cognitive function in patients with dementia over 6 months of treatment.⁹ A randomized, prospective, parallel-group trial with 1-year exposure to ACEIs (peridopril, 2 mg/d, or captopril, 37.5 mg/d) also showed that ACEIs could slow the rate of cognitive decline in patients with mild to moderate Alzheimer dementia in comparison with other antihypertensive drugs.⁸ A double-blind placebo-controlled crossover study investigated the effect of an ACEI (perindopril, 4 mg/d) on Parkinson disease. After a 4-week treatment period with perindopril, patients with Parkinson disease had better motor function and reduced dyskinesia.¹¹ In our research, there was still no human study discussing the association of ACEIs and ALS.

Postmortem examinations in ALS brains reveal a loss of motor neurons in the brainstem and ventral horn of the spinal cord, accompanied with neuroinflammation consisting of astrocytic activation and microglial proliferation.¹⁸ In rodent ALS models, microgliosis occurs in presymptomatic and symptomatic SOD1^G93A mice.¹⁹ Microgliosis also occurs at the onset and early stage of the disease in SOD1^G37R mice.²⁰ Increased brain ACE activity is associated with the activation of microglia and astrocytes.²¹ Perindopril, an ACEI, was demonstrated to suppress the activation of hippocampal microglia and astrocytes in a mouse model of AD.²² Perindopril also attenuated cognitive impairment and brain injury resulting from glial activation and oxidative stress in this study.

In addition to the inflammatory response, oxidative stress is involved in the pathogenesis of ALS.^23-25 Superoxide dismutase 1 mutations are responsible for approximately 20% of familial ALS cases. Superoxide dismutase 1 is an enzyme consisting of 153 amino acids and is involved in free-radical scavenging, with more than 150 different mutations reported to be pathogenic.²⁶ It has been reported that enalapril and moexipril promote neuronal survival by scavenging free radicals.²⁷ Some studies have shown that ACEIs could act directly to scavenge the hydroxyl radical.²⁸ Increased generation of reactive oxygen species is a significant pathological feature in patients with ALS^29,30; therefore, ACEIs could therefore suppress this.^31,32

Vitamin E (α-tocopherol) is an antioxidant that may decrease the risk for ALS. The effect of ACEIs on α-tocopherol has been demonstrated in hypertensive rats. Treatment of hypertensive rats with lisinopril, one of the ACEIs, ameliorated the decreasing of α-tocopherol and returned the concentration to near-normal levels.^33,34 A large meta-analysis showed that ALS rates declined with increasing years of using vitamin E supplements.³⁵ The relative risk was 0.64 among vitamin E users for 5 years or longer. A randomized, double-blind trial demonstrated that the risk for developing ALS decreased (relative risk, 0.56) for men with high α-tocopherol serum compared with low α-tocopherol levels. Additionally, a modest, nonsignificant protective effect from supplementation was seen in patients with low α-tocopherol serum levels. Angiotensin-converting enzyme inhibitors may return the depression of α-tocopherol serum levels, leading to a protective effect on ALS risk.³⁶

Glutamate is one of many neurotransmitters. Overactivation of glutamate receptors may cause injury to neurons in many neurodegenerative diseases such as Huntington disease, dementia, and ALS.^37,38 The only US Food and Drug Administration–approved drug for ALS, riluzole, works by reducing the levels of glutamate. Some evidence suggests that ACEIs may be neuroprotective by opening the mitochondrial adenosine triphosphate-sensitive potassium channels and therefore inhibiting the degradation of bradykinin. The ACEIs ramipril and perindopril are protective against glutamate-induced neurotoxicity in neonatal rat primary cortical cell cultures.³⁹ An additional animal study also reported that temocapril may exert neurotrophic actions on spinal motor neurons.¹² In the temocapril-treated ventral spinal cord, choline acetyltransferase activity was increased 2 to 3 times more than the activity in control samples. The possible therapeutic effect of ACEIs in the treatment of damaged motor neurons needs to be further investigated.

Although ALS risk was not associated with aspirin or NSAID use in several observational studies,^6,40 aspirin was inversely associated with ALS incidence in this study. The observed differences between the inverse association in this study and previous studies that observed no association could be caused by the differences of controlling confounders. In this study, we included steroid use, Charlson Comorbidity Index score, length of hospital stay, and number outpatient visits in a matched design; therefore, the higher statistical power let the significant association between aspirin be found. The results of the epidemiological studies were not uniform, and questions remain regarding the mechanism through which aspirin may affect ALS occurrence. Thus, focused clinical trials and further epidemiological studies and laboratory investigations are necessary to define the potential of aspirin, as well as ACEIs, to prevent ALS.

In addition to overall ACEI use, use of captopril and enalapril (individual compounds of ACEIs) also had potential protective effects on ALS risk. The other individual ACEI in our database had no statistical significance. This may have been a result of different activities and relatively small numbers.

The major strength of this study was its large sample size. To our knowledge, this is the first national population study to explore the association between ACEIs and ALS. This study was conducted using the total population of Taiwanese citizens seen in general medical practice; therefore, the findings can be applied to the general population.

However, this study had several limitations. In epidemiological studies, the time of first symptom onset is most relevant. However, our study database did not contain information regarding first symptom onset. A crucial confounding factor that was also not addressed in this study was vitamin E, which was not included in our study database. Therefore, we recommend estimating the association between ACEI use and ALS incidence by controlling vitamin E serum level and/or supplement dose. The database also did not contain certain crucial predictors such as the symptoms of ALS. Smoking and alcohol consumption are other important confounding factors that could not be addressed in this study⁴¹ because our study database did not contain information on cigarette smoking and alcohol consumption.

In summary, the results of this study demonstrated a 57% risk reduction in the chance for developing ALS in people who used ACEIs greater than 449.5 cDDD in 4 years. Our study suggests that ACEI use has a potential role in the prevention of ALS. This was an observational population-based study; hence, more animal and clinical studies are required to assess the possibility of using ACEIs for treating ALS.

Corresponding Author: Charles Tzu-Chi Lee, PhD, No. 100, Shihcyuan 1st Road, Sanmin District, Kaohsiung City 80708, Taiwan (ROC) (charles@kmu.edu.tw).

Accepted for Publication: September 17, 2014.

Published Online: November 10, 2014. doi:10.1001/jamaneurol.2014.3367.

Author Contributions: Dr Tzu-Chi Lee had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Lin, Tsai, Wu, Tzu-Chi Lee.

Acquisition, analysis, or interpretation of data: Kuang-Wu Lee, Tzu-Chi Lee.

Drafting of the manuscript: Lin, Tzu-Chi Lee.

Critical revision of the manuscript for important intellectual content: Tsai, Kuang-Wu Lee, Wu.

Statistical analysis: Lin, Tzu-Chi Lee.

Administrative, technical, or material support: Wu.

Study supervision: Tsai, Kuang-Wu Lee, Wu, Tzu-Chi Lee.

Conflict of Interest Disclosures: None reported.

Funding/Support: This research was supported by the Ching-Ling Foundation of Taipei Veterans General Hospital.

Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.