Risk Factors for Retained Instruments and Sponges After Surgery

Abstract

This case-control study was conducted to investigate the factors that lead to retained instruments and sponges during surgery. Cases were identified from the records of a large malpractice insurer in Massachusetts. All ofthe claims or incident reports of retained surgical sponges or instruments for the years 1985 to 2001 were included. Control cases were randomly selected from patients identified in hospital records of patients who had undergone the same operative procedure. Four controls were selected for each case. In the 15 years of the study period, there were 54 confirmed instances of a foreign body retained after surgery (total number of objects, 61). Most objects (54%) were left in the abdominal or pelvic cavity, 22% were left in the vagina, 7.4% in the thorax, and 17% in the spinal canal, face, brain, or extremities. In more than two thirds of cases (69%), the object retained was a surgical sponge. More than one sponge was retained in 7% of the cases, a clamp was left in 7%, and other objects, such as a retractor or electrode, were retained in 24%. At the major reporting hospitals included in the insurance company records, the rate of retained foreign bodies ranged from 1 per 8801 to 1 per 18,760 surgeries. The time to detection of the foreign body varied from the day of surgery to 6.5 years, with a median 21 days to detection. Three (6%) cases were identified on the day of surgery, and 14 (26%) were not detected until 60 days or more after surgery. Radiography or computed tomography was the usual method of detection (69%), but detection was made by physical examination or self-examination in 25% of the cases. In 9% of patients, the object was discovered incidentally during reoperation. Sixty-nine percent of the patients required surgery for removal of the foreign body and treatment of the resulting complications. Twelve of these retained objects led to small-bowel complications, including fistulae, obstruction, or visceral perforation. There was one death among these patients. When the cases and their matched controls were evaluated by univariate analyses, emergency surgery, unexpected change in procedure, involvement of more than one surgical team, and failure to perform a count of sponges and instruments were all associated with a higher rate of retained foreign body. No association was seen for age, duration or lateness of the operation, or the performance of multiple procedures. With multivariate analysis, three variables that posed a significantly higher risk of a retained object were identified: emergency procedure (risk ratio, 8.8; P <.001), unplanned change in procedure (risk ratio, 4.1; P <.01), and body mass index of the patient (risk ratio for each one-unit increase, 1.1; P <.01). Failure to perform a count of sponges and instruments had a strong association with emergency surgery but did not retain individual significance. Forty-seven claims involved litigation and resulted in an average of $52,581 in costs for compensation and legal-defense expenses.