Author + information
- Leslie T. Cooper, MD, FACC* (, )
- Tak Sun Tse, MD,
- Michael A. Mikhail, MD,
- Robert D. McBane, MD,
- Anthony W. Stanson, MD and
- Karla V. Ballman, PhD
- ↵*Gonda Vascular Center, Department of Biostatistics, Mayo Clinic and Foundation, Division of Cardiovascular Medicine/Mayo East 16B, Rochester, Minnesota 55905
To the Editor:
Thromboangiitis obliterans (TAO), or Buerger's disease, is an unusual tobacco-associated vasculopathy that is characterized by segmental thrombotic occlusions of small- and medium-sized arteries of the extremities. The long-term risks of amputation and death in TAO have not been described in detail. The incidence in North America is 8 to 11.6 per 100,000 persons per year, estimated from referral populations (1). Olin et al. (2) reported a study of 112 patients with TAO that was gathered from 1970 to 1987. Most, but not all, of the patients in that study were diagnosed by histopathology (n = 17) or arteriography (n = 88). During a mean follow-up of 91.6 months, 27% underwent one or more amputations. Despite the high morbidity of this illness, life expectancy among patients with Buerger's disease is reported to be comparable with that of age-matched controls (3).
A vascular medicine-trained physician abstracted the medical records of all patients with TAO registered at Mayo Clinic from January 1, 1976, to December 31, 1999 (n = 344). The inclusion criteria were: age <50 years, tobacco use, and objective evidence of medium or small arterial disease by either angiography (n = 107) or tissue pathology (n = 4). Exclusion criteria included any concomitant cause for arterial occlusive disease, including proximal atherosclerosis, trauma, entrapment syndromes, diabetes, or connective tissue or thrombotic diseases. Two hundred thirty-three patients were excluded.
All 111 patients had tests to exclude other causes of arterial occlusions as follows: erythrocyte sedimentation rate (n = 111; 100%), antinuclear antibody (n = 111; 100%), rheumatoid factor (n = 75; 68%), extractable nuclear antigen (n = 34; 31%), thrombosis panel (protein C, protein S, antithrombin III and, recently, factor V Leiden and prothrombin 20210A gene studies; n = 35; 32%), and anticardiolipin antibodies (n = 22; 20%).
A survey requesting detailed information on amputations, nicotine use, and family medical history was mailed to all surviving members of the study population (n = 88). A modified Fagerstrom Test for Nicotine Dependence questionnaire (FTND) was used to ascertain the degree of nicotine dependence (4). The FTND is a six-item self-reported instrument designed to evaluate the level of nicotine dependence. For former smokers, a modified version of the FTND with past tense verbs was used to allow collection of retrospective information. Fifty of 88 patients (57%) completed the survey.
In the statistical analysis, continuous variables were compared using a Wilcoxon rank-sum test. Categorical variables were compared with a chi-square test. Kaplan-Meier curves were generated to estimate rates of major and minor amputation and death, and a two-sample log-rank test was used to compare the survival experiences between groups (5). The Kaplan-Meier estimate of survival for patients with TAO was compared with expected survival in the U.S. population (on the basis of age- and gender-specific mortality rates) using a one-sample log-rank test (6). A Cox proportional hazards model was used to identify patient characteristics (including gender, race, treatment, and time-dependent continued tobacco exposure after diagnosis) that were significantly associated with the aforementioned events. Because patients typically experience multiple events (i.e., ulceration, amputation, and death), a multiple events extended Cox model also was used (7).
The mean age of the entire TAO cohort was 33 ± 8 years, and 40% were women. The percentage of female subjects increased from 32% for the 5-year period from 1976 to 1980 to 50% after 1996 (p = 0.09 for trend, by Armitage test). The mean follow-up was 15.6 ± 10.1 years (187 months; range, 6 months to 59 years; median, 14.8 years). Risk factors for atherosclerosis included hypertension (8.1%), hyperlipidemia (46.5%), and a family history of cardiovascular disease (56.6%). In addition, 63.1% of patients with TAO reported current use of tobacco at the time of diagnosis. Those subjects who completed the survey had a mean Fagerstrom Nicotine Dependency Score of 4.8, indicating mild to moderate addiction (4).
Treatment variables after diagnosis included the following: use of antiplatelet agents (46.9%), dihydropyridine calcium channel blockers (24.3%), alpha-blockers (19.8%), anticoagulants (18.9%), other vasodilators (5.4%), statins (5%), transdermal nitrates (4%), and angiotensin-converting enzyme inhibitors (0.9%). Forty-two patients (39%) had sympathectomy, and 13 (12%) had bypass surgery at the time of or subsequent to diagnosis.
The time to first amputation (any and major) ascertained from survey follow-up and chart review is illustrated in Figure 1.Forty patients underwent first amputations during a median follow-up of 14.8 years (mean follow-up, 15.6 years). The risk was 25% (0.16 to 0.33) at 5 years, 38% (0.27 to 0.47) at 10 years, and 46% (0.33 to 0.56) at 20 years. Twenty patients underwent first major amputations, defined as above-the-knee, below-the-knee, or hand amputation during this time period. The risk of major amputation was 11% (0.05 to 0.17) at 5 years, 21% (0.12 to 0.30) at 10 years, and 23% (0.13 to 0.31) at 20 years.
The effect of tobacco cessation on amputation risk and survival was assessed by estimating the rates for the interval of continued tobacco exposure and also for the interval of tobacco exposure cessation. Specifically, an individual who has stopped smoking would be in the interval of continued tobacco use from the time of diagnosis until he or she ceased smoking, at which point he or she would be placed in the interval of tobacco exposure cessation. In an analysis in which each amputation in persons with multiple amputations was considered a separate event, the amputation rate was 84.3 (confidence interval 68.4 to 102.7 amputations/person year × 1000) in the interval of continued tobacco use versus 30.6 (confidence interval 17.5 to 49.6 amputations/person year × 1000) in the interval of discontinued tobacco use.
Survival in the TAO cohort was compared with the expected survival in the U.S. population on the basis of age- and gender-specific mortality rates (Fig. 2).Fifteen of 111 subjects died in a median of 14.8 years of follow-up. The average age at death was 52.2 ± 8.9 years. The survival of the TAO cohort was significantly lower than that in the matched U.S. population (p < 0.001 by log-rank test). In univariate analysis, only older age at diagnosis and decade of diagnosis were associated with an increased risk of death. The risk of death was nearly identical in those subjects who continued to use tobacco and those who quit. In multivariate models, no variables were associated with risk of death.
The present TAO series is the first to report time-dependent event rates for major or any amputation and death. Our analyses have resulted in several novel observations: The risk of amputation continues in ongoing smokers up to a median of 14.8 years (mean, 15.6 years) after initial diagnosis. The risk of amputation in previous smokers is eliminated by 8 years after smoking cessation. In contrast to previous studies, we report an excessive late mortality in patients with TAO compared with the U.S. population (2,3).
Our data are limited in several ways. Not all patients had systematic evaluation for thrombophilia and autoimmune disease. The results of this study may not apply to patients with TAO in the community. The cause of death is not known for most patients. The ascertainment of nicotine cessation by chart review and survey may not accurately reflect actual smoking cessation. Those with mild addiction might be more likely than those with severe addiction to complete the survey. Finally, a substantial fraction of patients with TAO who reported smoking cessation may have continued to smoke.
Please note: the first two authors contributed equally to this work. Funded by a Mayo Foundation Protocol Grant.
- American College of Cardiology Foundation
- ↵Joyce JW. Buerger's disease (thromboangiitis obliterans). Rheum Dis Clin North Am 1990;16:463–70.
- Harrington D.P.,
- Fleming T.R.
- Therneau T.M.,
- Grambsch P.M.