Population-based estimates of mortality associated with diabetes: use of a death certificate check box in North Dakota.

OBJECTIVES: Overall and cause-specific mortality among persons with diabetes in North Dakota was estimated and compared with estimates from previous population-based studies. METHODS: Data were derived from North Dakota death certificate data, which included unique information on decedents' diabetes status and Behavioral Risk Factor Surveillance System estimates of the diabetic and nondiabetic adult populations of North Dakota. RESULTS: The risk of death among adults with diabetes was 2.6 (2.2, 2.9) times that of adults without diabetes. Relative risks of death among adults with diabetes were at least twice as high for heart disease, cerebrovascular disease, accidents and adverse events, and kidney disease and 70% to 80% higher for pneumonia and influenza, malignant neoplasms, arterial disease, and other causes. Risks remained substantial in the oldest age group. These findings are comparable to results of other population-based studies. CONCLUSIONS: Diabetes status information enhanced the usefulness of death certificate data in examining mortality associated with diabetes and confirms that the effect of diabetes on death is substantial.

Serratia marcescens surgical wound infection following breast reconstruction.

Surgical wound infections due to gram-negative bacilli have been rarely reported following breast implant surgery. From April to November 1989, four patients from one plastic surgeon's practice developed Serratia marcescens surgical wound infection (SWI) following breast reconstruction procedures with implantation of six expandable mammary implants. All six implants were removed for unabated S. marcescens SWI. Symptoms developed 13-161 days (median, 66 days) after surgery. When compared with nonexpandable silicone breast implants used during the period November 1, 1988, to October 31, 1989, expandable implants were associated with a greater risk of S. marcescens SWI (4/10 versus 0/11 patients, p = 0.04). Epidemiologic studies revealed that infection was associated with saline expansion of the implants performed in the surgeon's office. S. marcescens was cultured from a bag of commercial saline used on at least two of the four patients with SWI; the isolate from the saline and the three available patient isolates had identical serotype (O-undetermined:H4) and antimicrobial susceptibility patterns. Review of office procedures revealed that hands were not routinely washed before and aseptic technique was not used during the expansion procedure. Cultures of unopened bags of saline and an unused expandable implant were sterile. We hypothesize that multiple use of saline bags and nonsterile expansion technique extrinsically contaminated saline solutions and resulted in implant and/or surgical site infection. This investigation underscores the importance of avoiding multiple use of solutions intended for single use and of using aseptic technique when manipulating prosthetic devices.