Increase in gram-negative organisms indicates need to reconsider prophylaxis

Learning from your experience is good; learning from the experiences of others is even better. So that their experiences might enable others to better identify and address modifiable risk factors for surgical site infections (SSIs), three tertiary care children’s hospitals conducted a retrospective cohort study of more than 1,350 spinal instrumentation procedures for scoliosis (Fig. 1). Michael G. Vitale, MD, MPH, chief of the pediatric spine service at Children’s Hospital of New York, presented the results during the 2011 annual meeting of the Pediatric Orthopaedic Society of North America (POSNA).

Spinal fusion in pediatric patients is associated with a significant risk of SSI, and recent studies have shown that patients with nonidiopathic scoliosis have higher rates of SSI than those with idiopathic scoliosis.

The retrospective chart review included all scoliosis patients who underwent posterior spinal instrumentations at the three hospitals between January 2006 and December 2008, a total of 905 patients and 1,352 procedures. Demographic and surgical data were collected for each patient. The definition of SSIs was based on the Centers for Disease Control and Prevention’s (CDC) National Healthcare Safety Network case definition, which covers 1 year of postoperative follow-up and is based on positive deep cultures or a diagnosed SSI by a surgeon or attending physician.

Patients were primarily female (61 percent), and the etiology of the scoliosis was primarily idiopathic (Table 1). In all, 78 SSIs were identified, providing an overall infection rate of 5.8 percent.

“We broke down infection rates by scoliosis etiology, surgical procedure (primary and revision/converted fusion and expandable rod growing constructs), and causative pathogen,” explained Dr. Vitale. “We specifically looked at the issue of gram-negative pathogens, which we thought might be on the rise.”

Infection and etiology
Scoliosis etiology was strongly associated with the risk of infection; the SSI rate in idiopathic scoliosis was just 2.6 percent, while in nonidiopathic scoliosis, it was 7.6 percent (p < 0.001). With further analysis, the researchers found that the rate of SSI varied among the different etiologies for nonidiopathic scoliosis and ranged from 3.9 percent among patients with congenital scoliosis to 9.2 percent among those with neuromuscular scoliosis.

Overall, 45.9 percent of SSIs with positive cultures contained at least one gram-negative pathogen. “This strongly suggests a role for routine, targeted prophylaxis for gram-negative pathogens, based on local susceptibility patterns,” said Dr. Vitale.

Infection and procedure
“Of 417 primary fusion procedures in patients with idiopathic scoliosis, the infection rate was just 1.2 percent, which is pretty reasonable,” said Dr. Vitale. In contrast, among the 198 patients with neuromuscular scoliosis who underwent primary fusion, the infection rate was 13 percent. “This is an area that needs some attention,” he noted.

Growing construct revision procedures had a significantly higher rate of SSI than lengthening procedures, and nonidiopathic etiologies had a significantly higher infection rate than idiopathic scoliosis.

Pathogens
In contrast to previous reports, the current review documented a high prevalence of gram-negative organisms in these infections. “Of all the cultured organisms associated with SSI, both gram-positive and gram-negative pathogens had similar cefazolin susceptibility,” said Dr. Vitale.

“In this multicenter cohort, gram-negative organisms were associated with roughly half of all SSIs,” he noted. “All three centers (one West Coast, two Northeast) had almost the exact percentage of gram-negative infections, so this seems to be a general problem. But many hospitals are just using cefazolin, which doesn’t cover the gram-negative pathogens, or many gram-positive organisms.

“We need to think about whether preoperative cefazolin is enough, without any other agent,” he continued. “We need to think about best practices and put together a working group to develop a consensus-based protocol.”

Dr. Vitale urged the adoption of better processes, infrastructure, and systems.

“Because we had a high infection rate, we went to the hospital administration and presented our findings and recommendations. The administration ordered that our recommendations be adopted, and we now have nearly 100 percent compliance,” he said.

CDC’s definition of SSIs (PDF)

Dr. Vitale’s coauthors for “Surgical site infection following spinal instrumentation for scoliosis: Lessons learned from an analysis of 1,352 procedures at three centers” include W.G. Stuart Mackenzie, BS, MA; Brendan A. Williams; Hiroko Matsumoto, MA; Jacqueline Corona, MD; Christopher Lee, BS; Stephanie R. Cody, BS; Lisa Covington, RN, MPH; Lisa Saiman, MD, MPH; Jack M. Flynn, MD; David L. Skaggs, MD; David P. Roye, MD.

Disclosure information: Dr. Vitale—Biomet, Stryker, Synthes, Chest Wall and Spinal Deformity Study Group; Scoliosis Research Society (SRS); POSNA. Dr. Flynn—Biomet; Wolters Kluwer Health, Lippincott Williams & Wilkins; Orthopedics Today, POSNA, SRS. Dr. Skaggs—Medtronic; Stryker; AO North America; Biomet; POSNA; Growing Spine Study Group. Dr. Roye—Synthes, Journal of Bone and Joint Surgery–American. Dr. Corona; Messers. MacKenzie, Williams, Matsumoto, Lee; Ms. Cody—no conflicts. Dr. Saiman, Ms. Covington—no disclosure.

Mary Ann Porucznik is managing editor of AAOS Now. She can be reached at porucznik@aaos.org

Bottom Line

• A retrospective chart review of pediatric spine patients conducted at three children’s hospitals found a strong association between scoliosis etiology and SSI, with the highest rate of infection in patients with neuromuscular scoliosis.
• Growing construct revisions procedures had significantly higher SSI rates than lengthening procedures.
• A high prevalence of gram-negative organisms in these infections suggests a role for routine, targeted prophylaxis for these pathogens.