Difficulty in diagnosing surgical site infection after arthroscopy in developing countries.

To the Editor—In the United States, the use of arthroscopy has been increasing since the 1970s. Although infectious complications after arthroscopy appear to be fairly rare (0.01%-0.48% of procedures), significant morbidity and significant costs are associated with the procedure. In Thailand, although arthroscopy has been increasingly performed during the past decade, it has not been incorporated into the surveillance systems of most Thai hospitals, and postdischarge surveillance is often suboptimal. The lack of a national benchmark also makes it difficult to compare rates of surgical site infection (SSI) after arthroscopy between hospitals. We report on the difficulty of using Centers for Disease Control and Prevention (CDC) definitions to diagnose SSI after arthroscopy and highlight some implications for surveillance in developing countries. In September 2008, an infection control unit was notified of 4 potential cases of SSI after arthroscopy at a hospital in Thailand; all cases met the definitions for postoperative infections—these definitions were introduced into hospitals participating in the CDC National Nosocomial Infections Surveillance (NNIS) system—and occurred in 2008. An outbreak investigation was initiated in September 2008. In this particular hospital, 6 surgeons performed this type of surgery. The surgical logbook was reviewed to identify all patients who had undergone arthroscopy, followed by a careful chart review, for evidence of postoperative SSI. A line list was created. Postdischarge surveillance (by mail and/or telephone) was also performed for all patients who had undergone arthroscopy during the period from January through November 2008, according to the CDC-NNIS recommendations. Infection control practices were observed in the operating room of the orthopedic ward (for suboptimal hand hygiene or lapses in sterile techniques or in the implementation of some other infection control measure). There were 293 arthroscopic procedures performed during the period from January through November 2008. A total of 6 suspected cases of SSI after arthroscopy were identified. However, after a careful review of the medical records, only 3 of these 6 cases showed evidence of SSI. Three (50%) of the 6 original suspected cases did not show evidence of infection. An additional case of SSI after arthroscopy was identified during postdischarge surveillance. Notably, the treating physicians had diagnosed all 4 cases of SSI. Each of the 4 patients required a 6-week course of systemic antibiotic prophylaxis. The median age of the patients was 26 years (range, 23-31 years), and the median duration of surgery was 5 hours (range, 4-6 hours): 2 (50%) of the 4 patients were operated on by doctor A, and 2 (50%) had material implanted in the joint space. Joint fluid specimens from each patient were obtained for culture, and these specimens revealed that 1 patient was infected with methicillin-resistant Staphylococcus aureus and that another was infected with coagulase-negative Staphylococcus species; 2 of the 4 patients' joint fluid specimens did contain a microorganism. Two (50%) of the 4 patients were readmitted to the hospital for multiple surgical joint procedures. Compared with patients who did not develop an SSI after arthroscopy, patients who did develop an SSI after arthroscopy were more likely to have a prolonged duration of surgery (median duration, 1.4 vs 5 hours; P = .04). There were no differences in other preoperative, operative, and postoperative risk factors between patients who developed an SSI after arthroscopy and patients who did not. The rates of SSI after arthroscopy and the infection control practices observed are summarized in the Table for each surgeon. After the investigation, feedback was provided to the surgeons, and educational sessions were held for healthcare personnel who worked in the operating rooms. Because of the low incidence of SSI after arthroscopy, studies that try to delineate the risk factors for SSI and/or that try to describe the effects of the implementation of preventive measures during orthopedic surgery are challenging.