Clinical and laboratory parameters associated with multiple surgeries in children with acute hematogenous osteomyelitis.

BACKGROUND: In the era of methicillin-resistant Staphylococcus aureus, surgical intervention seems to be increasingly common as an adjunct to treatment for pediatric acute hematogenous osteomyelitis (AHO). The purpose of this study is to identify objective clinical and laboratory parameters that are associated with repeated surgical intervention during the acute phase of treatment. METHODS: Fifty-seven children who were consecutively evaluated and treated for AHO at a single institution during 2009 were studied retrospectively. Objective clinical and laboratory parameters related to length of hospitalization were recorded for each child. Univariate analysis was performed with ordinal logistic regression, χ, Fisher exact, and Wilcoxon rank-sum and 2-value tests to identify independent variables associated with the occurrence of surgery in children with AHO. Multivariate logistic regression was used to identify parameters associated with repeated surgical intervention. RESULTS: Sixteen children were treated with antibiotics alone and no surgery. There were 41 children who had at least 1 surgery and 12 who underwent ≥2 surgeries. Multiple logistic regression showed that a swollen extremity (P=0.002), initial C-reactive protein (CRP) value >9.9 mg/dL (P=0.02), and respiration rate >27 breaths/min (P=0.02) were significantly associated with the occurrence of at least 1 surgery. The best model to identify the occurrence of repeated surgical intervention in children with AHO included: ≥4 febrile days on antibiotics; and the CRP values at admission (>19.8 mg/dL), 48 hours after the initial surgery (>21.5 mg/dL), and 96 hours after the initial surgery (>15.3 mg/dL). CONCLUSIONS: Children with AHO who have sustained marked elevation of CRP values during the first 96 hours after surgery and who remain febrile while on antibiotics have an increased likelihood of repeated surgical intervention and should be evaluated carefully for additional surgical treatment. LEVEL OF EVIDENCE: Level III, retrospective comparative study.

The impact of evidence-based clinical practice guidelines applied by a multidisciplinary team for the care of children with osteomyelitis.

BACKGROUND: Care of children with osteomyelitis requires multidisciplinary collaboration. This study evaluates the impact of evidence-based guidelines for the treatment of pediatric osteomyelitis when utilized by a multidisciplinary team. METHODS: Guidelines for pediatric osteomyelitis were developed and were implemented by a multidisciplinary team comprised of individuals from several hospital services, including orthopaedics, pediatrics, infectious disease, nursing, and social work, who met daily to conduct rounds and make treatment decisions. With use of retrospective review and statistical analysis, we compared children with osteomyelitis who had been managed at our institution from 2002 to 2004 (prior to the implementation of the guidelines), referred to as Group I in this study, with those who were managed in 2009 according to the guidelines, referred to as Group II. RESULTS: Two hundred and ten children in Group I were compared with sixty-one children in Group II. No significant differences between the two cohorts were noted for age, sex, incidence of methicillin-resistant Staphylococcus aureus infection (18.1% in Group I compared with 26.2% in Group II), incidence of methicillin-sensitive Staphylococcus aureus infection (23.8% in Group I compared with 27.9% in Group II), bacteremia, or surgical procedures. Significant differences (p < 0.05) between cohorts were noted for each of the following: the delay in magnetic resonance imaging after admission (2.5 days in Group I compared with one day in Group II), the percentage of patients who had received clindamycin as the initial antibiotic (12.9% in Group I compared with 85.2% in Group II), the percentage of patients who had had a blood culture before antibiotic administration (79.5% in Group I compared with 91.8% in Group II), the percentage of patients who had had a culture of tissue from the infection site (62.9% in Group I compared with 78.7% in Group II), the percentage of patients in whom the infecting organism was identified on tissue or blood culture (60.0% in Group I compared with 73.8% in Group II), the number of antibiotic changes (2.0 changes in Group I compared with 1.4 changes in Group II), and the mean duration of oral antibiotic use (27.7 days in Group I compared with 43.7 days in Group II). When compared with Group I, Group II had clinically important trends of a shorter total length of hospital stay (12.8 days in Group I compared with 9.7 days in Group II; p = 0.054) and a lower hospital readmission rate (11.4% in Group I compared with 6.6% in Group II; p = 0.34). CONCLUSIONS: Evidence-based treatment guidelines applied by a multidisciplinary team resulted in a more efficient diagnostic workup, a higher rate of identifying the causative organism, and improved adherence to initial antibiotic recommendations with fewer antibiotic changes during treatment. Additionally, there were trends toward lower hospital readmission rates and a shorter length of hospitalization.