All Purulence Is Local

Epidemiology and Management of Skin and Soft Tissue Infections in Three Urban Emergency Departments

Chris Merritt; John P Haran; Jacob Mintzer; Joseph Stricker; Roland C Merchant


BMC Emerg Med. 2013;13(26) 

In This Article


The Study Population

The ICD-9 code search yielded 1,158 separate ED visits for SSTI, of which 1,094 (94.5%) were initial visits for SSTIs. The remaining 64 ED visits constituted either return visits for the same infection or ICD-9 mis-coding. Of the 1,094 ED visits, 160 (14.6%) represented patients with known healthcare exposure, leaving 936 patients – the study population – in whom the SSTI was likely community-acquired.

Table 1 summarizes demographic and clinical characteristics of the study population, stratified by age group. As compared to adult community-acquired SSTI patients, pediatric patients were more likely to be female, non-white, and insured. In addition, pediatric SSTI patients were more likely to have a diagnosis other than abscess or cellulitis (primarily impetigo or paronychia, data not shown). As compared to adults, more pediatric abscesses occurred on the buttock (28.8% vs. 15.4%; p < 0.05) and fewer on the face (6.9% vs. 15.8%; p < 0.05).

ED Management of Suspected Community-acquired SSTIs

Among suspected community-acquired SSTIs, of the ED patients diagnosed with abscesses, pediatric and adult patients were equally likely to undergo I&D in the ED (58.9% and 65.6%; p < 0.29), but microbiologic culture was ordered more often in the pediatric patients (65.8% vs. 47.6%; p < 0.005).

The majority of patients with suspected community-acquired SSTIs were evaluated in the ED and discharged. Pediatric patients with abscesses were more likely than adults with abscesses to be admitted to hospital (34.3% vs. 14.5%; p < 0.001).

Antibiotic Use

Antibiotics (whether intravenous (IV) or oral, used in the ED or prescribed at discharge, or any combination of these) were prescribed to 86.1% of the 936 ED patients with suspected community-acquired SSTIs (94% of those with cellulitis vs. 78.4% of those with abscess; p < 0.0001). For patients with cellulitis, 93.9% of adult and 94.1% of pediatric patients were prescribed antibiotics (p < 0.97); for those with an abscess, 76.9% of adult and 84.9% of pediatric patients were prescribed antibiotics (p < 0.14); and for all other suspected community-acquired SSTIs, 73.6% of adult and 85.3% of pediatric patients were prescribed antibiotics (p < 0.20).

Overall, 38.2% of SSTI patients (88.6% of admitted patients and 15.7% of discharged patients) received IV antibiotics in the ED, more frequently in adults than in children (40.4% vs. 29.8%; p < 0.009). The most commonly prescribed IV antibiotics for adults were vancomycin (24.9%), ampicillin/sulbactam (11.4%), and, cefazolin (7.9%), and for children were clindamycin (15.7%), cefazolin (5.8%), and ampicillin/sulbactam (4.7%). Adult patients were more likely than pediatric patients to receive IV vancomycin (24.9 vs. 1.6%; p < 0.001) and ampicillin/sulbactam (11.4% vs. 4.7%; p < 0.007), but were less likely than pediatric patients to receive IV clindamycin (4.6% vs. 15.7%; p < 0.001).

Among discharged patients, trimethoprim-sulfamethoxazole (TMP-SMX), clindamycin, and cephalexin were prescribed commonly. 25.3% of all SSTI patients received an oral antibiotic in the ED, and 80.8% of patients discharged received an antibiotic prescription. Adult SSTI patients were more likely than pediatric SSTI patients to be prescribed oral TMP-SMX (59.1% vs. 37.2%; p < 0.001) at ED discharge. Of those prescribed TMP-SMX, adults were more likely than children to be also prescribed oral cephalexin (53.8% vs. 34.5%; p < 0.001).

Microbiologic Data

Table 2 summarizes the results of microbiologic cultures by age group among ED patients with cultured infections. S. aureus comprised the majority of culture isolates, and was more common in pediatric patients, while mixed flora was more common among adults than among children. Of S. aureus isolated from CA-SSTIs, 60.4% was categorized as MRSA, with similar proportions in adults and children (60.9% vs. 59.2%; p < 0.84).

Table 3 depicts the antibiotic susceptibilities among S. aureus isolated from presumed community-acquired purulent infections, comparing the resistance in these ED-acquired SSTI cultures against the resistance reported for MRSA and MSSA on the antibiogram distributed by the hospitals' microbiology laboratory for S. aureus from all sources in 2010.

Discordance of Antibiotic Therapy With Culture Results

Table 4 compares antibiotic treatment with culture results among SSTIs for which both antibiotics were prescribed and cultures obtained. Clinicians using single antibiotics (anti-MRSA [Table 4A] or anti-MSSA [Table 4B]) used monotherapy that accurately targeted the resultant cultured pathogen accurately 39.3% of the time (ranging from 35–52% depending on age and treatment strategy). Cultured SSTIs from 100% of pediatric patients and 67.8% of adult patients treated with multi-drug "double coverage" (Table 4C) grew only Staphylococcus.

Table 5 displays the univariable logistic regression analyses investigating demographic and clinical correlates of coverage of the resultant pathogen with the chosen antibiotic regimen, "double coverage" antibiotic usage, and discordance of empiric MRSA therapy (use of anti-MRSA antibiotics in the absence of MRSA, or vice versa).

Patients who underwent I&D were more likely to receive antibiotics that covered the resultant pathogen than those who did not undergo the same procedure in the ED. Patients admitted to the hospital were more likely to receive antibiotics in the ED to which the resultant pathogen was susceptible than those discharged home.

Age group was strongly associated with treatment with two or more antistaphylococcal antibiotics, with adult patients more likely than pediatric patients to receive such multiple antibiotic coverage. Black patients were less likely than non-black patients to receive multi-drug coverage. However, when age and race were considered jointly as correlates, only adult age remained associated with greater "double coverage" usage.

There were no demographic or clinical factors identified in association with discordance between presence or absence of empiric anti-MRSA antibiotic therapy and the presence or absence of MRSA among those undergoing culture and receiving antibiotics.