Study investigators (hospital-based infectious disease specialists, internal medicine specialists with an infectious disease sub-specialty, or medical microbiologists) identified subjects eligible for inclusion. To be included, patients must have had a microbiologically confirmed MRSA cSSTI (e.g., deep or extensive cellulitis, infected wound or ulcer, major abscess or other soft tissue infection requiring substantial surgical intervention), and received ≥3 days of IV anti-MRSA antibiotics including, but not limited to: clindamycin, daptomycin, fusidic acid, linezolid, rifampicin, teicoplanin, tigecycline, TMP/SMX, or vancomycin, during their hospital stay. Patients were excluded if they had been treated for the same cSSTI within three months of hospitalization, or had suspected or proven diabetic foot infections, osteomyelitis, infective endocarditis, meningitis, joint infections, necrotizing soft tissue infections, gangrene, prosthetic joint infection, or prosthetic implant/device. Additionally, patients were excluded if they were less than 18 years old, pregnant or lactating, had significant concomitant infection at other sites, were immunosuppressed (e.g., diagnosed hematologic malignancy, neutropenia, or rheumatoid arthritis; receiving chronic steroids or cancer chemotherapy), or enrolled in another cSSTI-related clinical trial.

The key study outcomes were patient treatment patterns, clinical outcomes, and opportunities for early switch or early discharge.

Treatment patterns analyzed included time to MRSA-active treatment from cSSTI index date and number of MRSA-active treatment received. Patients’ first, last, and discharge (if applicable) MRSA-active treatment were recorded. Additionally, length of total antibiotic therapy, (time between initiation of MRSA-active therapy and last day of MRSA-active therapy or discharge); length of IV antibiotic therapy (subset of MRSA-active therapy days for which IV antibiotics were prescribed) were assessed. Length of stay (number of days from hospital admission for patients admitted for cSSTI, or cSSTI diagnosis date to hospital discharge) was also calculated for all patients.

Clinical outcomes included clinical response at end of each MRSA-active therapy, end of full MRSA-active treatment, and at hospital discharge. Development of IV-line infections, super infections, and severe sepsis were documented. Patients were also followed to record cSSTI relapse due to MRSA within 30 days of discharge or unplanned re-hospitalization for cSSTI or other health issues.

Literature review and expert consensus were used to create ES and ED eligibility criteria applicable to real-world clinical settings.12–18 For ES eligibility, at a minimum, patients had to meet all of the following key criteria prior to IV antibiotic discontinuation: stable clinical infection, afebrile/temperature <38°C for 24h, normalized white blood cell (WBC) count (4×109/L<WBC<12×109/L), no unexplained tachycardia, systolic blood pressure ≥100mmHg able to tolerate PO fluids/diet and take PO medication with no gastrointestinal absorption problems. For ED eligibility, at a minimum, patients had to meet the above key criteria for ES prior to discharge and have no other reason to remain in the hospital except for infection management.

To estimate resource use savings, potential reduction in IV days was calculated for ES patients and bed days saved were calculated for ED patients, using hypothetical and actual treatment patterns. Hypothetical length of IV therapy was calculated as the length of time between date of initial MRSA-active IV therapy and date when patient satisfied key ES criteria. Potential reduction in IV days was the difference in days between actual length of MRSA targeted therapy from hypothetical length of MRSA targeted therapy. Hypothetical LOS was calculated as number of days from cSSTI index date to date that patients satisfied all key criteria. Bed days saved was the difference in days between actual and hypothetical length of stay for ED eligible patients. The economic impact of mean cost per bed days saved in ED eligible patients was calculated by multiplying the bed days saved by a Brazil-specific unit cost of $31 international dollars (Intl.) (45 Brazilian real [BRL]) per bed day based on World Health Organization (WHO)-reported unit costs for providing a Brazil hospital bed day in the year 2007; this was reported in international dollars (local currency units) and adjusted for inflation (percent change in consumer price index) to the year 2015.19,20 Costs included by the WHO represent the “hotel” component of hospital costs, such as personnel, capital, and food, and do not include drug and diagnostic test costs.

Descriptive analyses were conducted for all outcomes; frequencies and percentages for categorical variables, and means, medians, and interquartile ranges for continuous variables. Results were analyzed using SAS software version 9.4.

Data was collected from medical charts of 199 patients in Brazil with MRSA cSSTIs. Almost half of patients were Caucasian (45%), followed by equal proportions of black/African and Latino patients (19%, respectively). Most patients were male (64%), and the average age at hospital admission was 52 years (Table 2). More than three-quarters of patients (78%) reported comorbidities, with diabetes (49%), coronary artery disease (28%) and peripheral vascular disease (19%) being most frequently reported. About 80% of patients reported treatment of MRSA cSSTI as the primary reason for hospitalization, with 29% of patients presenting with deep or extensive cellulitis, 28% with a surgical site infection, and 18% with an infected ulcer. Over half (59%) of all cSSTIs were community acquired, and a smaller proportion were hospital acquired (18%) or healthcare-associated (16%) infections. Only 6% of patients had been infected with MRSA a year prior to hospitalization.

All but one of the 79 hospital sites had an antibiotic subcommittee or steering committee. Fewer hospitals had an IV to PO antibiotic switch protocol (20%) or discharge protocol (11%) for patients admitted for cSSTI MRSA. However, one-fourth of hospitals reported having OPAT available.

All but three patients (99%) received at least one MRSA-active therapy. Of patients treated with MRSA-active therapy, 161 patients (82%) did not require subsequent antibiotic therapy, 33 patients (17%) required two courses, and two patients (1%) required three courses (Table 3). The mean time of administration of MRSA-active therapy was 2.1 days (SD 6.3) following MRSA cSSTI diagnosis. Only four patients (2%) were switched from IV-to-PO MRSA-active therapy while hospitalized. Average LOS from cSSTI diagnosis through discharge was 20.7 (SD 18.6) days. About one-quarter of cSSTIs (26%) warranted surgical procedures, incision and drainage of the infection (54%) and debridement (40%) were performed most frequently. Severe sepsis or septic shock occurred among approximately one-quarter (26%) of those hospitalized, while super infection (5%) and IV-line infection (3%) were rarely found.

The most frequently prescribed antibiotics did not change between initial line and final line treatment, with vancomycin IV (initial 58%, final 53%), clindamycin IV (initial 20%, final 14%), and daptomycin IV (initial 7%, final 9%) used most frequently. However, analysis of final inpatient MRSA-active antibiotic regimens showed that throughout the course of treatment, the proportion of inpatients receiving vancomycin IV or clindamycin IV decreased, while the proportion of patients receiving daptomycin IV or linezolid IV increased. At discharge, 31 patients (16%) were prescribed a MRSA-active antibiotic, and four-fifths of these were oral targeted therapy. Clindamycin PO (42%), linezolid PO (16%), and TMP-SMX PO (10%) were the most commonly prescribed oral antibiotics at discharge while daptomycin IV (10%) and vancomycin IV (7%) were the most common IV antibiotics. Antibiotic susceptibilities in this sample are presented in Table 4 and show susceptibility rates <40% for the commonly used oral agents TMP-SMX and clindamycin. Mean length of MRSA-active antibiotic treatment regardless of route (i.e., IV, intramuscular, or PO) was 14.7 (SD 10.1) days, while MRSA-active IV antibiotic treatment lasted a mean of 14.6 (SD 10.1) days.

At discharge, over half of patients (55%) showed resolution of signs and symptoms or improvement to such an extent that further therapy was not necessary (Table 5). In total, 69 patients (35%) were discharged with an improvement in their signs and symptoms, and 14 (7%) were discharged after failure to cure or improve cSSTI signs and symptoms. Relapse and re-hospitalization for a cSSTI due to MRSA within 30 days post-discharge occurred in only two (1%) patients while re-hospitalization for any reason occurred in 11 (6%) patients.

A total of 62 (31%) patients met all six key ES criteria prior to actual IV discontinuation. Among the ES-eligible patients, the actual mean length of IV therapy was 15.5 (SD 11.1) days but hypothetically could have been 8.7 (SD 8.9) days when removing days after patients met all ES criteria, suggesting a potential reduction in IV therapy duration of 6.8 (SD 7.8) days (Fig. 1).

Fig. 1.

Brazil Comparison of Actual and Hypothetical IV-line and Bed Days in ES- and ED-eligible Patients.

Abbreviations: ED, early discharge; ES, early switch; IV, intravenous.

(0.11MB).

Approximately one-third (32.7%) of patients met all key ED criteria prior to hospital discharge. The actual mean LOS for patients meeting key ED criteria was 22.2 (SD 23.0) days; the hypothetical LOS for these patients was 16.9 (SD 21.5) days, suggesting a potential reduction in LOS of 5.3 (SD 7.0) days. Assuming an average cost of $31 Intl, (45 real) per bed day, the total savings would be $164 Intl. (239 real) in bed-day cost savings per ED-eligible patient.