Emphysema: a potential risk factor for subarachnoid hemorrhage and ruptured aortic aneurysm

Abstract

Background and Purpose:

Protease/antiprotease imbalance is implicated in the pathogenesis of emphysema and may also lead to vessel-wall weakening, aneurysm development, and rupture. However, it is unclear whether emphysema is associated with cerebral and aortic aneurysm rupture.

Methods :

We performed a retrospective cohort study using outpatient and inpatient claims data from 2008 to 2014 from a nationally representative sample of Medicare beneficiaries ≥66 years of age. Our predictor variable was emphysema and our outcome was hospitalization for either aneurysmal subarachnoid hemorrhage (SAH) or a ruptured aortic aneurysm. All predictors and outcomes were defined using previously reported ICD-9-CM diagnosis code algorithms. Survival statistics and Cox regression were used to compare risk between patients with and without emphysema.

Results:

We identified 1,670,915 patients, of whom 133,972 had a diagnosis of emphysema. Over a mean follow-up period of 4.3 (±1.9) years, we identified 4,835 cases of aneurysm rupture, 433 of which occurred in patients with emphysema. The annual incidence of aneurysm rupture was 6.5 (95% confidence interval [CI], 6.4-6.8) per 10,000 in patients without emphysema and 14.6 (95% CI, 13.3-16.0) per 10,000 in patients with emphysema. After adjustment for demographics and known risk factors for aneurysmal disease, emphysema was independently associated with aneurysm rupture (hazard ratio [HR], 1.7; 95% CI, 1.5-1.9). Emphysema was associated with both aneurysmal SAH (HR, 1.5; 95% CI, 1.3-1.7) and ruptured aortic aneurysm (HR, 2.3; 95% CI, 1.9-2.8).

Conclusion:

Patients with emphysema face an increased risk of developing SAH and aortic aneurysm rupture, potentially consistent with shared pathways in pathogenesis.

Aneurysmal subarachnoid hemorrhage (SAH) and ruptured aortic aneurysm have significant morbidity and mortality. There are shared risk factors for aneurysm formation and rupture in these different vascular beds.^1,2 The causes of vessel-wall rupture are poorly understood but it may be partly triggered by an imbalance in protease/anti-protease activity that leads to a cascade of inflammation, oxidative stress, and vessel-wall destruction.² A similar protease/anti-protease imbalance is thought to occur in lung diseases such as emphysema.^3,4,5 Given the theoretically similar pathogenic pathways in emphysema and aneurysmal rupture, we evaluated the association between emphysema and ruptured cerebral or aortic aneurysm.

Methods

We performed a retrospective cohort study using a nationally representative sample of Medicare beneficiaries. The U.S. Federal Government’s Centers for Medicare and Medicaid Services (CMS) provides health insurance to a large majority of U.S. residents once they reach 65 years of age. Each claim includes the date of service and up to 25 International Classification of Diseases, 9^th Revision, Clinical Modification (ICD-9-CM) diagnosis codes which justify the claim. Multiple claims are linked using a unique, anonymous identifier code, allowing for a comprehensive and longitudinal analysis of each beneficiary’s care over time. We are not allowed to share these data directly, but investigators can apply to CMS for these data. The Weill Cornell Medicine institutional review board approved this study and waived the requirement for informed consent.

We included claims from 2008-2014 in a 5% nationally representative sample of beneficiaries. We limited our cohort to beneficiaries ≥66 years of age with traditional fee-for-service Medicare (both Parts A and B) for at least 1 year (or until death, if applicable). Our exposure was emphysema, identified using ICD-9-CM code 492, as commonly used in previous studies.⁶ Our outcome was a composite of SAH or ruptured aortic aneurysm. SAH was identified using a previously validated ICD-9-CM code algorithm that excludes traumatic SAH and has a good positive predictive value for nontraumatic SAH, which is mostly comprised of aneurysmal SAH.⁷ Ruptured aortic aneurysms were identified by ICD-9-CM codes used in prior studies.⁸ As covariates, we used ICD-9-CM codes to identify the following previously reported risk factors for aneurysmal disease: hypertension, diabetes, tobacco use, and alcohol abuse.^1,2 Survival statistics were used to calculate incidence rates of SAH and aortic aneurysm rupture. Patients were censored at the time of death, end of Medicare coverage, or December 31, 2014. Patients were also censored after their first outcome and thus emphysema cases diagnosed after a first aneurysm event were excluded. Cox proportional hazards analysis was used to compare the likelihood of SAH or aortic aneurysm rupture after adjustment for age, sex, race, hypertension, diabetes, tobacco use, and alcohol abuse. All covariates were left in the model regardless of statistical significance. In secondary analyses, SAH and ruptured aortic aneurysm were considered separately. We performed a sensitivity analysis limited to patients without any documented tobacco use disorder. All analyses were performed using Stata/MP, version 13 (StataCorp, TX). The threshold of statistical significance allowed for an alpha error of 0.05.

Results

We identified 1,670,915 Medicare beneficiaries, of whom 133,972 (8%) received a diagnosis of emphysema. Patients with emphysema were older (77 ± 8 versus 74 ± 8 years of age); less likely to be female (52% vs 58%); and more likely to have hypertension, diabetes, and both tobacco and alcohol use (Table 1). Over a mean follow-up period of 4.2 (±1.9) years, there were 4,835 cases of aneurysm rupture, 433 of which occurred in patients with emphysema. Of the 4,835 cases of aneurysm rupture, most (76%) involved SAH.

Table 1.

Characteristics of Medicare Beneficiaries, Stratified by Emphysema, 5% National Sample.

Characteristic a	Emphysema (N=133,972)	No Emphysema (N = 1,536,943)
Age, mean (SD), y	77 (8)	74 (8)
Female	69,196 (52)	861,454 (58)
White race	120,376 (89.9)	1323,135 (86.1)
Hypertension	117,638 (88.2)	1,023,624 (68.9)
Diabetes	64,941 (48.7)	482,975 (32.3)
Tobacco use	37,035 (27.8)	48,442 (3.2)
Alcohol use	36,956 (27.7)	67,547 (4.6)

^aData are presented as number (%) unless otherwise specified.

^bSelf-reported by patients or their surrogates.

The annual incidence of aneurysm rupture was 6.5 (95% CI, 6.4-6.8) per 10,000 in patients without emphysema and 14.6 (95% CI, 13.3-16.0) per 10,000 in patients with emphysema (Figure 1). After adjustment for demographics and risk factors, emphysema was independently associated with aneurysm rupture (hazard ratio [HR], 1.7; 95% CI, 1.5-1.9). In secondary analyses, emphysema was associated with both ruptured aortic aneurysm (HR, 2.3; 95% CI, 1.9-2.8) and SAH (HR, 1.5; 95% CI, 1.3-1.7). In sensitivity analyses excluding patients with any documented tobacco use disorder, we found a similar association between emphysema and aneurysm rupture (HR, 1.7; 95% CI, 1.5-1.9).

Figure 1.

Cumulative Rate of Aneurysm Rupture in Patients with Emphysema versus Patients without Emphysema.

We performed two post hoc analyses. We found a significant association between emphysema and unruptured cerebral or aortic aneurysms (HR, 1.7; 95% CI, 1.6-1.8). As expected, we found no association between renal colic and our composite outcome of SAH or aortic aneurysm rupture (HR, 1.2; 95% CI, 0.8-1.7).

Discussion

In a large group of elderly patients, we found that emphysema was associated with aneurysmal SAH and aortic aneurysm rupture. Our findings are consistent with the hypothesis that the protease/anti-protease imbalance seen in emphysema may be involved in the pathogenesis of aneurysm rupture. Emphysema is characterized by an imbalance in protease/anti-protease activity that predisposes to connective tissue destruction and activation of an inflammatory cascade in alveolar walls,² and this same mechanism may occur within the media and adventitia of the aorta and cerebral arteries. Alpha-1 antitrypsin is a potent protease inhibitor that has been implicated in the pathogenesis of emphysema.^2-5,9 Several studies have found an inverse relationship between alpha-1 antitrypsin levels and aneurysm formation and rupture.^10-13 A population-based study found an association between emphysema and SAH,¹⁴ but the investigators could not adjust for tobacco use, which has an important role in both emphysema and aneurysm formation. Our findings build on these prior studies by indicating that emphysema may be associated with aneurysm rupture independently of other established risk factors, including documented tobacco use.

The results of our study should be considered in light of its limitations. First, by using claims data, we lacked information regarding important clinical variables such as severity, duration, and treatment of emphysema such as any steroid use, as well as detailed information regarding cerebral or aortic aneurysms such as location, size, or symptoms. The diagnosis codes for aortic aneurysms have not been specifically validated, leaving open the possibility of misclassification. Second, although we accounted for tobacco use, we lacked detailed history regarding the duration and amount of cigarette use. Therefore, our findings may be partly due to residual confounding by this important shared risk factor. However, this is unlikely to substantially explain our findings because the ICD-9-CM codes for tobacco use disorder in Medicare claims are likely to at least reliably capture whether or not patients have any tobacco use since Medicare reimburses providers for tobacco cessation counseling, and we found similar results in sensitivity analyses limited to patients without any documented tobacco use. Third, we were unable to measure serum biomarkers or perform genetic analyses to test our hypothesis regarding protease/anti-protease imbalance as a potential explanation for the link between emphysema and ruptured aneurysm. Future studies are warranted to confirm this association and explore its potential mechanisms.

Conclusion

Patients with emphysema faced an increased risk of SAH and aortic aneurysm rupture. This finding suggests that emphysema and aneurysm formation may share a similar underlying pathophysiology. Further studies to explore biomarkers of vessel-wall weakening and potential molecular targeted therapies of aneurysms are encouraged.