Pharmacologic interventions for thoracic aneurysms remain poorly characterized. The results of a pilot study by our group suggested improved outcomes among patients with thoracic aortic aneurysm who were taking statins. In the present study, we undertook a comprehensive analysis of a larger cohort of patients from the Database of the Aortic Institute at Yale-New Haven Hospital. A total of 1,560 patients met the inclusion criteria. The adverse events (i.e., death, dissection, or rupture) and surgery rates for patients with (n = 369, 24%) and without (n = 1,191, 76%) statin therapy were compared. We evaluated 3 anatomic components of the aorta: root, ascending and arch, and descending and thoracoabdominal aortic aneurysms. A smaller proportion of the statin group had adverse events: overall, 7% versus 15%; ascending and arch, 6% versus 15%; and descending and thoracoabdominal aortic aneurysms, 8% versus 20%. Also, a smaller proportion of statin patients required surgery: overall, 48% versus 60%; ascending and arch, 51% versus 62%; and descending and thoracoabdominal aortic aneurysms, 36% versus 59% (p <0.001 to 0.01). The protective effect of statins was seen in all segments, except the aortic root. Log-rank evaluation of the interval to an adverse event or surgery was longer among statin-treated patients (p <0.001). Logistic regression analysis found statin use, angiotensin receptor blocker use, and chronic obstructive pulmonary disease were associated with decreased adverse events, and statin use, angiotensin receptor blocker use, β-blocker therapy, and age were associated with a decreased odds of requiring surgery. Multiple logistic regression analysis found only statins were associated with a decreased odds of an adverse event and that statins, coronary artery disease, and chronic obstructive pulmonary disease were associated with a decreased odds of undergoing surgery. In conclusion, these findings provide a medicinal option for the arsenal of treatment options for patients with aneurysms of the thoracic aorta.
Increasing evidence has suggested that 3-hydroxy-3-methylglutaryl-coenzyme A reductase inhibitors, statins, influence the natural history of abdominal aortic aneurysms (AAAs). This has been thought to reflect statin’s altering the inflammatory milieu of the aneurysmal tissue, rather than any direct effect of lowering the lipid levels. Proteolytic enzymes, especially matrix metalloproteinases, have been implicated in the pathophysiology of aortic aneurysms. Evidence in vitro and in vivo has associated statin therapy with decreased metalloproteinase expression. Reason exists to suspect that thoracic aortic aneurysms (TAAs) might not share the same benefit from statin therapy as AAAs. For example, although AAAs have been closely associated with atherosclerosis, ascending aortic aneurysms have not. The segments of the aorta derive from different embryonic origins. The aortic root arises from the lateral plate, the ascending and arch from the neural crest, and the descending and abdominal aortas are of mesodermal origin. A beneficial effect of statin therapy on AAAs has been previously reported ; however, the effect of statin therapy on TAAs has been less characterized. In a pilot study with a smaller cohort and shorter follow-up, our group demonstrated improved outcomes in patients with TAAs who were receiving statin therapy. In the present study, we have explored further the association of statins with thoracic aneurysm progression and complications, with more than double the patient number of our original investigation and substantially longer follow-up than in our pilot study.
Methods
The present study used the Database of the Aortic Institute at Yale-New Haven Hospital (New Haven, Connecticut), which has been the basis of our numerous previous and multiple ongoing investigations of the natural history of TAAs. A total of 2,185 patients were included in the database from January 1985 through July 2011, encompassing 4,593 patient-years of follow-up. Our inclusion criteria for the present study were as follows: age >16 years, aneurysms of the aortic root or ascending, descending, or thoracoabdominal aorta, and documented information on statin use. A total of 1,560 patients met these criteria.
We examined the clinical variables, including aneurysm location, medication usage, and co-morbid conditions. These characteristics were entered at the initial presentation and then adjusted during follow-up. In the final analysis, the patients were recorded as taking a given medication if they had ever been treated with that medication. The end points of our data analysis were (1) adverse event, defined as the occurrence of dissection, rupture, or disease-related death (occurring within 30 days of surgery, dissection, or rupture), and (2) the need for surgery. All patients who had undergone surgical repair had undergone open procedures. The indications for surgery included evidence of rupture or dissection, symptoms caused by the aneurysm, rapid aneurysm growth, or aneurysm size of 5 to 5.5 cm in the ascending aorta or 6 to 6.5 cm in the descending aorta in patients with and without Marfan syndrome, respectively.
The baseline characteristics between the statin and nonstatin groups were compared using Student’s t test. One-way analysis of variance was used to compare the distribution of aneurysms and the severity of chronic obstructive pulmonary disease (COPD) and renal disease. The outcomes were evaluated in 4 groups: group 1, all thoracic aneurysms combined; group 2, aneurysms of the aortic root; group 3, ascending and arch aneurysms; and group 4, descending and thoracoabdominal aortic aneurysms. The proportion of patients reaching the end points for each group were evaluated using chi-square proportions testing. A survival analysis was performed grouping all aneurysms distal to the root. Log-rank analysis of the Kaplan-Meier survival curves compared the effect of pharmacologic therapy. Survival was measured as the interval from enrollment in the study to the measured outcome. If no end point was reached by the last follow-up visit, that patient was censored at that date.
Univariate and multivariate logistic regression analysis were used to evaluate the influence of risk factors for adverse events or surgery. Risk factors previously identified to be related to the natural history of TAAs and AAAs—including hypertension, coronary artery disease, congestive heart failure, COPD, obesity, and Marfan syndrome—were entered into our initial analysis. The baseline characteristics differing between the 2 groups were included in this analysis. The growth rate was calculated as the linear change in the aortic diameter over time. In some patients, the growth rate was not calculated because only 1 imaging study had been performed before their final outcome. Statistical analysis was performed using Statistical Package for Social Sciences (IBM SPSS, Armonk, New York).
Results
A total of 1,560 patients met the inclusion criteria. Of these, 1,191 did not received statin therapy (76%), and 369 did (24%). The baseline clinical characteristics, medications, and co-morbidities are listed in Table 1 . The patients receiving statin therapy were slightly older. An equivalent proportion of patients in each group were treated with angiotensin-converting enzyme inhibitors, insulin, or nonsteroidal anti-inflammatory drugs. A larger proportion of patients in the statin group were treated with angiotensin receptor blockers (ARBs), β blockers, antiarrhythmic agents, or calcium channel blockers. An equal proportion of patients in each group had been diagnosed with congestive heart failure and peripheral vascular disease. A larger proportion of the statin group had been diagnosed with hypertension, coronary artery disease, COPD, and renal disease. The nonstatin group contained more obese patients and, although uncommon, slightly more patients with Marfan syndrome. Analysis of variance demonstrated an equal distribution of root, arch, descending, and thoracoabdominal aortic aneurysms in the statin and nonstatin groups.
| Characteristic | No Statins (n = 1,191) | Statins (n = 369) | p Value |
|---|---|---|---|
| Age (yrs) | 62 ± 17 | 65 ± 16 | <0.004 |
| Men (%) | 61 | 65 | NS |
| Initial aneurysm diameter (cm) | 4.5 ± 0.05 | 4.5 ± 0.08 | NS |
| ACE inhibitor | 277 (23) | 103 (28) | NS |
| Antiarrhythmic | 26 (2) | 20 (5) | 0.003 |
| ARB | 78 (7) | 65 (17) | <0.001 |
| β Blocker | 520 (43) | 224 (61) | <0.001 |
| Calcium channel blocker | 252 (21) | 101 (27) | 0.038 |
| Insulin | 12 (1) | 10 (3) | NS |
| NSAID | 38 (3) | 6 (2) | NS |
| Hypertension | <0.0001 | ||
| Mild | 419 (39) | 133 (49) | |
| Moderate | 218 (20) | 111 (33) | |
| Severe | 116 (11) | 44 (13) | |
| Total | 753 (63) | 288 (78) | |
| None | 332 (30) | 48 (14) | |
| Coronary artery disease | 216 (18) | 123 (33) | <0.0001 |
| Congestive heart failure | 74 (6) | 19 (5) | NS |
| AAA | 168 (12) | 42 (10) | NS |
| Peripheral vascular disease | 66 (5) | 22 (6) | NS |
| Chronic kidney disease | NS | ||
| Mild | 72 (6.8) | 19 (6.0) | |
| Moderate | 14 (1.3) | 1 (0.1) | |
| Severe | 9 (0.8) | 7 (2) | |
| Total | 95 (9) | 27 (8.8) | |
| None | 960 (91) | 279 (88) | |
| COPD | 0.03 | ||
| Mild | 136 (13) | 57 (18) | |
| Moderate | 78 (7) | 33 (10) | |
| Severe | 36 (3) | 11 (4) | |
| Total | 250 (23) | 100 (31) | |
| None | 818 (77) | 224 (69) | |
| Obese | 43 (3) | 1 (0.25) | <0.0039 |
| Marfan syndrome | 57 (5) | 2 (0.5) | 0.0009 |
| Aneurysm location | NS | ||
| Root | 158 (13) | 46 (12) | |
| Arch | 121 (10) | 40 (11) | |
| Ascending | 595 (50) | 192 (52) | |
| Descending | 196 (16) | 66 (18) | |
| TAAA | 121 (10) | 25 (7) |
Figure 1 demonstrates the proportion of patients experiencing an adverse event or requiring surgery within 10 years after their initial presentation. For all aneurysms combined, a larger proportion of nonstatin patients (15%) than statin patients (7%) experienced adverse events (p = 0.017). This protective association was seen with the ascending and arch aneurysms (15% vs 6%; p <0.001) and descending aneurysms and thoracoabdominal aortic aneurysms (20% vs 8%; p = 0.012) but not with aneurysms of the aortic root. Similarly, for all aneurysms combined, a larger percentage of nonstatin patients than statin patients required surgery (60% vs 48%; p = 0.001). The protective effect was not seen for root aneurysms but was evidenced in ascending and arch aneurysms (62% vs 51%; p = 0.002) and in descending aneurysms and thoracoabdominal aortic aneurysms (59% vs 36%; p <0.001).
Kaplan-Meir freedom from adverse events and surgery for ascending, arch, descending, and thoracoabdominal aortic aneurysms was studied for the statin and nonstatin groups. Log-rank analysis demonstrated an increased duration to an adverse event for ≤10 years of follow-up in the statin group ( Figure 2 ; p = 0.000). The statin patients had a slightly longer time to surgery for ≤8 years, after which, a nearly equivalent proportion of patients in each group required surgery ( Figure 2 ; p = 0.05). The statin and nonstatin patients had no difference in baseline aneurysm size ( Table 2 ). The linear growth rate was calculated for 817 patients. No difference was noted in the aneurysm growth rates of the patients with and without statin treatment.
| Anatomic Location | Statin Use | Aortic Diameter | Growth Rate | ||
|---|---|---|---|---|---|
| Patients (n) | Mean (cm) | Patients (n) | Mean (cm/yr) | ||
| All | No | 1,184 | 4.52 ± 1.67 | 603 | 1.72 ± 8.25 |
| Yes | 368 | 4.56 ± 1.46 | 214 | 0.88 ± 4.48 | |
| Root | No | 158 | 4.06 ± 1.36 | 84 | 1.07 ± 5.53 |
| Yes | 46 | 4.40 ± 1.22 | 23 | 0.58 ± 1.37 | |
| Ascending plus aortic arch | No | 589 | 4.64 ± 1.59 | 710 | 0.65 ± 2.25 |
| Yes | 191 | 4.52 ± 1.32 | 231 | 0.34 ± 1.06 | |
| Descending plus thoracoabdominal | No | 316 | 4.47 ± 1.92 | 316 | 0.92 ± 2.27 |
| Yes | 91 | 4.47 ± 1.85 | 91 | 0.45 ± 1.24 | |
A significantly larger proportion of statin patients were treated with ARBs ( Table 3 ). Because ARBs are associated with improved AAA outcomes, we evaluated how ARB therapy might have confounded our findings. ARB treatment was not related to the proportion of patients experiencing adverse events ( Table 4 ). Overall, ARB therapy was associated with a smaller proportion of patients requiring surgery (p <0.018). Log-rank analysis of the Kaplan-Meier curves demonstrated this small protective effect was continued for 4 years of observation ( Figure 3 ; p = 0.04).
| ARB use | No Statins | Statins |
|---|---|---|
| No | 1,107 (94) | 302 (82) |
| Yes | 77 (6) | 66 (18) |
| Location | ARB | Adverse Event | p Value | Surgery | p Value |
|---|---|---|---|---|---|
| All | No | 201 (14) | NS | 824 (58) | 0.018 |
| Yes | 13 (9) | 69 (48) | |||
| Root | No | 18 (10) | NS | 101 (54) | NS |
| Yes | 2 (13) | 7 (44) | |||
| Ascending plus aortic arch | No | 115 (13) | NS | 513 (61) | NS |
| Yes | 7 (9) | 96 (54) | |||
| Descending plus thoracoabdominal | No | 68 (18) | NS | 376 (55) | 0.011 |
| Yes | 4 (13) | 10 (32) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
