Number Needed to Treat: Analyzing of the Effectiveness of Thoracoabdominal Aortic Repair

• Published in: European journal of vascular and endovascular surgery Vol. 28; no. 2; pp. 154 - 157
• Main Authors: Miller, C.C, Porat, E.E, Estrera, A.L, Vinnerkvist, A.N, Huynh, T.T.T, Safi, H.J
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 01.08.2004
• Subjects: Adolescent; Adult; Aged; Aged, 80 and over; Aortic aneurysm; Aortic Aneurysm, Abdominal - mortality; Aortic Aneurysm, Abdominal - surgery; Aortic Aneurysm, Thoracic - mortality; Aortic Aneurysm, Thoracic - surgery; Child; Cohort Studies; Descending; Female; Health Services Needs and Demand - statistics & numerical data; Humans; Male; Middle Aged; Number needed to treat; Outcome Assessment (Health Care); Survival Analysis; Thoracic aortic aneurysm; Thoracoabdominal aortic aneurysm; Treatment Outcome; Descending; Thoracoabdominal aortic aneurysm; Number needed to treat; Aortic aneurysm; Thoracic aortic aneurysm
• ISSN: 1078-5884; 1532-2165
• DOI: 10.1016/j.ejvs.2004.03.004

Background. Number needed to treat (NNT) is a method used to calculate the number of patients who need to be treated to prevent one adverse outcome. To analyze the effectiveness of thoracoabdominal and descending thoracic aortic aneurysm repair, we computed the NNT required to prevent one death. Methods. Between Jan 1991 and Feb 2003, we repaired 1004 aneurysms of the descending thoracic and thoracoabdominal aorta. We followed the patients from surgery until death. Five-year actuarial survival in our population was computed by the Kaplan–Meier method. Natural history data for comparison were taken from the population-based work of Bickerstaff et al., 1982. NNT was calculated as the reciprocal of the risk difference at 5 years. 95% confidence intervals were computed by the method of Daly. Results. Five-year mortality in the population-based cohort was 87 vs. 39% in our treated population, for a risk difference of 48%. 1/0.48=2, indicating that two patients need to be treated to prevent one death at 5 years (95% CI 1.8–2.5, p<0.0001). Conclusion. An NNT of two demonstrates the effectiveness of surgical repair of descending thoracic and thoracoabdominal aortic aneurysms when compared to the natural history. By comparison, carotid endarterectomy for symptomatic lesions >70% has an NNT of 15 to prevent a single stroke or death. NNT can also be applied to aneurysm size criteria to estimate the effort required to prevent death or rupture for a given aneurysm size.