Conference Coverage

Predicting functional outcome after pediatric osteomyelitis


 

AT ESPID 2017

– Ninety percent of children with acute hematogenous osteomyelitis will do fine after their initial course of antibiotics and don’t require long-term follow-up; and the other 10% can be identified within the first few days of hospitalization, Lawson A. Copley, MD, said at the annual meeting of the European Society for Paediatric Infectious Diseases.

The tool that enables physicians to distinguish the 10% of children at high risk for severe orthopedic sequelae is a validated severity of illness score that can be determined within the first several days of hospitalization. The 0-10 score, developed by Dr. Copley and his coinvestigators (J Pediatr Orthop. 2016 Oct 12. doi: 10.1097/BPO.0000000000000879), awards points for the patient’s initial C-reactive protein level, the C-reactive protein levels on hospital days 2-3 and 4-5, the number of febrile days on antibiotic therapy, the band percentage of WBC, ICU admission, and disseminated disease such as endocarditis, septic pulmonary embolism, and deep venous thrombosis.

Dr. Lawson A. Copley, University of Texas Southwestern

Dr. Lawson A. Copley

He and his colleagues developed the severity of illness score because children with osteomyelitis have wide variance in their disease severity, treatment requirements, and long-term risks, explained Dr. Copley, professor of orthopedics and pediatrics at the University of Texas, Dallas.

There is a dearth of long-term follow-up studies of pediatric osteomyelitis. To address this unmet need, he and his coinvestigators have enrolled 198 children with acute hematogenous osteomyelitis in an ongoing prospective study. All were treated with antibiotics until clinical and laboratory resolution of the infection and achievement of a normal erythrocyte sedimentation rate. All patients are being followed in a specialized multidisciplinary clinic at Texas Scottish Rite Hospital for Children directed by Dr. Copley. To date, 118 patients have been seen for their 2-year follow-up visit, which includes radiographs of the previous infection site, an orthopedic exam, and completion of the Pediatric Quality of Life Inventory and the Pediatric Outcomes Data Collection Instrument.

At follow-up, the children fell into three broad categories. Ten percent had severe radiographic and/or clinical sequelae such as limb length discrepancy, visible deformity, limited range of motion, osteonecrosis, physeal arrest, or joint destruction. Roughly 40% had complete resolution with normal function and no growth disturbance or other sequelae. And 50% had clinical resolution with a completely normal physical exam and excellent outcome measures, but minimal radiographic sequelae, mainly consisting of central physeal tenting.

“We think that they’re probably a low-risk group,” he said of that last group.

Children with severe sequelae had greater severity of illness at presentation and a more complicated course of initial therapy than those with complete resolution at 2 years of follow-up. Their mean severity of illness score was 4.9, compared with 1.8 in the 40% of children with complete resolution and 3.4 in those with mild radiographic sequelae.

In a univariate logistic regression analysis, each point increase in initial disease severity score was associated with a 20% bump in the risk of developing severe sequelae, with a predictive area under the curve of 0.67. A multivariate logistic regression analysis identified other independent predictors of severe sequelae: age below 6 years, being culture positive for methicillin-resistant Streptococcus aureus, and osteomyelitis contiguous with septic arthritis or abscess, which ultimately led to osteonecrosis and destruction. Incorporating these additional risk factors along with the initial severity of illness score improved the predictive area under the curve to 0.85.

About one-half of patients seen in the pediatric osteomyelitis clinic were bacteremic on admission, and of those, roughly half continued to be bacteremic despite antibiotic therapy. However, there was no difference in the prevalence of bacteremia between the groups with mild versus severe illness.

Asked how introduction of the severity-of-illness score has affected his surgical approach, Dr. Copley said he has become selectively more surgically aggressive.

“A lot of our children have abscesses that are pretty substantial,” he noted. “We’ve learned the hard way. I’ve been doing this for about 14 years now, and initially I used to do a lot of simple debridement of the infection. Now we’re much more extensive in our approach, so we do fewer surgeries, but those surgeries are more extensive.”

Dr. Copley reported having no financial conflicts regarding his study.

Next Article: