Microbiologic Predictors of Pelvic Osteomyelitis Related to Decubitus Ulcers

Abstract

Background. Management of pelvic osteomyelitis related to decubitus ulcers (PODU) remains challenging, whereas definitive therapy is based on blood, bone, or deep tissue cultures, empirical therapy prior to culture results may be indicated in patients with sepsis or cellulitis surrounding PODU. The objective of this retrospective case series is to develop an institutional protocol for empirical therapy of PODU when indicated based on local microbiology results. Methods. Hospitalized adults with PODU from 1 August 2005 to 1 August 2015 at Palmetto Health hospitals in Columbia, SC were identified. PODU was defined based on clinical, radiographic, and microbiology criteria. Descriptive statistical methods (Fisher's exact) were used for preliminary analysis. Results. Seventy-five cases with PODU were included with a mean age of 53 years and male predominance (48; 64%). The most common comorbidities were paraplegia (45, 60%), diabetes (23, 31%) and previous strokes (17, 23%). Forty-nine cases (65%) received antibiotics within a year of PODU. Prior infections or colonization with P. aeruginosa within the past year was present in 24/75 (32%) cases. Most cases had multiple sources of cultures: blood (61; 81%), bone/deep tissue (37; 49%), and/or superficial (73; 97%). Among a total of 99 clinical isolates, 56 (57%) were Gram-positive cocci (GPC) and 43 (43%) were Gram-negative bacilli (GNB). The most common organisms were Enterobacteriaceae (26; 26%), coagulase negative staphylococci (CONS) (20, 20%), Stapylococcus aureus (19, 19%), [12 (12%) methicillin- resistant S. aureus], and P. aeruginosa (9, 9%). Of the Enterobacteriaceae, 69% (18/26) were susceptible to ciprofloxacin and 88% (23/26) to ceftriaxone. All cases (9/9) of PODU due to P. aeruginosa had a prior infection/colonization with P. aeruginosa within 1 year as compared with 15/66 (23%) in those with PODU due to other organisms (P = 0.001). Conclusion. The microbiology of PODU is diverse (including GPC and GNB). Prior positive P. aeruginosa culture was a predictor of P. aeruginosa PODU. When empirical antimicrobial therapy is indicated, data support the use of intravenous vancomycin plus ceftriaxone in the absence of prior infection/colonization with P. aeruginosa, or intravenous vancomycin plus an anti-pseudomonal agent in the presence of prior P. aeruginosa within the past year.