Objective: To examine the state of research on population-based studies of the incidence of limb amputation and birth prevalence of limb deficiency.
Data sources: A total of 18 publication databases were searched, including MEDLINE, CINAHL, and the Cochrane Library.
Study selection: The search was performed by using a hierarchical process. Articles were reviewed for inclusion by 3 reviewers. Inclusion criteria included defined catchment area, calculation of population-based incidence rates, defined etiology of limb loss, and English language. Review articles, animal studies, case reports, cohort studies, letters, and editorials were excluded.
Data extraction: Figures on the estimated incidence of amputation and birth prevalence of congenital limb deficiency were gleaned from selected reports and assembled into a table format by etiology.
Data synthesis: The studies varied in scope, quality, and methodology, making comparisons between studies difficult. Incidence rates of acquired amputation varied greatly between and within nations. Rates of all-cause acquired amputation ranged from 1.2 first major amputations per 10,000 women in Japan to 4.4 per 10,000 men in the Navajo Nation in the United States between 1992 and 1997. Consistent among all nations, the risk of amputation was greatest among persons with diabetes mellitus.
Conclusions: Surveillance of congenital limb deficiency exists in much of the developed world. Existing studies of acquired amputation suffer from a host of methodologic problems. Future efforts should be directed toward the application of standardized measures and methods to enable trends to be evaluated over time and comparisons to be made within and between countries.