ORIGINAL INVESTIGATION Impact of Methicillin Resistance on the Outcome of Patients With Bacteremia Caused by Staphylococcus aureus Stephan Harbarth, MD; Olivier Rutschmann, MD; Philippe Sudre, MD, MS; Didier Pittet, MD, MS Background: Uncertainties remain about the contri- Main Outcome Measure: The in-hospital mortality bution of methicillin resistance to morbidity and mor- after staphylococcal bacteremia. tality associated with bacteremia caused by Staphylococ- cus aureus. Results: In the population-based study, the relative haz- ard of death among patients with MRSA BSI (n=39, 14 Objective: To assess the impact of methicillin resistance deaths, 36% fatality rate) compared with patients with on patient outcome after staphylococcal bacteremia. MSSA BSI (n=145, 40 deaths, 28% fatality rate) was 1.1 (95% confidence interval, 0.5-2.1), after adjusting for age Methods: We investigated a cohort of 145 patients with and length of stay from admission to the onset of blood- methicillin-sensitive S aureus bloodstream infection stream infection. Following pairwise matching (n=38), (MSSA BSI) and 39 patients with methicillin-resistant the in-hospital mortality was 34% in both groups (odds S aureus bloodstream infection (MRSA BSI) and further ratio, 1.0; 95% confidence interval, 0.4-2.5). Infection was performed a pairwise-matched (1:1) case-control study. the probable or definite cause of death in 54% of pa- All patients in the University Hospital of Geneva, Geneva, tients with MRSA BSI and 69% of patients with MSSA BSI Switzerland, with clinically significant staphylococcal bac- who died. teremia between January 1, 1994, and December 31, 1995, were included in the study. For the case-control study, Conclusion: Methicillin resistance in patients with S au- cases were defined as patients with MRSA BSI; control reus bacteremia had no significant impact on patient out- patients with MSSA BSI were selected in a stepwise man- come as measured by in-hospital mortality after adjust- ner according to the following matching variables: age, ment was made for major confounders. sex, number of comorbidities, severity of underlying ill- ness, and prior length of stay in the hospital. Matching was successful for 97% of the cohort. Arch Intern Med. 1998;158:182-189 ETHICILLIN-resistant caused by methicillin-sensitive S aureus Staphylococcus aureus (MSSA).14-17 Methicillin-resistant S (MRSA) has become a aureus BSI, in particular, can be associ- worldwide problem, ated with a crude case-fatality rate rang- adding to the overall ing from 10% to almost 60%.3,4,10,18 Nev- Mburden of nosocomial infections.1,2 Blood- ertheless, the real effect on mortality of stream infections (BSIs) due to MRSA may methicillin resistance in staphylococcal account for up to 50% of all staphylococ- bacteremia is still unknown. cal BSI.3-6 Factors predisposing to MRSA Romero-Vivas et al10 showed that BSI include previous antibiotic treat- nosocomial bacteremia due to MRSA was ment, prolonged hospital stay, intravas- associated with a 3-fold higher mortality cular catheters, severe underlying condi- than MSSA BSI after adjustment for sev- tions, and MRSA nasal carriage.6-11 eral risk factors. The authors determined Despite ongoing controversy about that this difference was unaffected by age, the relative virulence of MRSA in vitro the severity of the underlying condition, and in animal models,12,13 most studies the length of hospital stay, preinfection an- From the Infection Control Program, Division of Infectious have concluded that infections due to tibiotic treatment, and previous surgery, Diseases, Department of MRSA are probably similar in virulence, but they hypothesized that it might have Internal Medicine, University as measured by the duration of fever, been related to an inaccurate adjustment Hospital of Geneva, Geneva, infectious complications, length of hos- for the severity of illness and underlying Switzerland. pital stay, and mortality, as infections diseases. ARCH INTERN MED/ VOL 158, JAN 26, 1998 182 ©1998 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 PATIENTS AND METHODS S aureus within 24 hours and the clinical course was con- sistent with staphylococcal infection. Organisms of S aureus isolated from the bloodstream were considered to SETTING be contaminants if culture of only 1 blood specimen yielded the organism, if the clinicians or the infectious This study was undertaken in the University Hospital of diseases consultant judged the organism to be a contami- Geneva, Geneva, Switzerland, a 1500-bed health care cen- nant, and if antibiotic therapy directed against the organ- ter providing primary and tertiary care for Geneva and the ism was not administered. Bloodstream infection was con- surrounding areas (500 000 inhabitants). About 40 000 pa- sidered community-acquired if the blood specimen for the tients are admitted annually for a mean length of stay of first positive culture was obtained within 72 hours of 10 days. admission or in the presence of S aureus infection at At this center, the incidence rates of MRSA coloniza- another body site at the time of hospital admission.22 An tion or infection increased significantly between 1989 and infection that was neither present nor incubating at the 1995 from 0.05 cases per 100 admissions in 1989 to 0.81 time of admission was considered nosocomial.23 cases per 100 admissions in 1995 (R2=0.94, P,.001).19 Af- ter the implementation of infection control measures in Foci of BSI 1993, the annual attack rate (±SD) of MRSA colonization or infection remained stable: 0.55±0.05 cases per 100 ad- In the case of secondary bacteremia, a primary focus of in- missions (range, 0.51-0.62). fection was determined using the following definitions: The term “pneumonia” was retained for patients with clinical STUDY OBJECTIVE AND DESIGN signs of lower respiratory tract infection associated with ra- diographic evidence of pulmonary infiltrates not attribut- The objective of the study was to evaluate the effects of able to other causes. An intravenous catheter was consid- methicillin resistance on the morbidity and mortality as- ered as the source of BSI if the catheter had been in place sociated with S aureus bacteremia. All patients with clini- for at least 72 hours, culture of a quantitative catheter speci- cally substantial episodes of S aureus bacteremia between men yielded more than 100 colonies of S aureus,24 or cul- January 1, 1994, and December 31, 1995, were included ture of a specimen of purulent drainage from the insertion in a retrospective population-based cohort study. The main site grew S aureus. Endocarditis was considered in pa- outcome measure was the in-hospital mortality after staphy- tients with S aureus bacteremia and 1 or more of the fol- lococcal BSI. In addition, a pairwise-matched (1:1) case- lowing characteristics: surgical or autopsy findings con- control study concerning 38 patients with MSSA and MRSA sistent with endocarditis, echocardiographic evidence of bacteremia was performed to confirm findings from the valvular vegetation, and the presence of septic emboli. Uri- population-based cohort study. nary tract infection was considered if the patient had uri- nary symptoms and S aureus (.105 colony-forming units PATIENT POPULATION per milliliter) was identified as the sole pathogen from urine. Soft tissue infection was considered in the case of a pa- All episodes of MSSA or MRSA bacteremia from January tient who had a pure culture of S aureus from a tissue or 1, 1994, through December 31, 1995, were identified us- drainage specimen from the affected site and signs of in- ing 3 sources of information: (1) database of the comput- fection. “Surgical wound infections” were diagnosed fol- erized clinical microbiology laboratory, (2) prospective sur- lowing the standard definitions from the Centers for Dis- veillance records of patients with nosocomial BSIs from the ease Control and Prevention.25 “Primary bacteremia” defined infection control program, and (3) follow-up data of cases conditions in which no primary focus could be deter- by infectious diseases consultants. When a patient had more mined.26 than 1 episode of S aureus bacteremia or more than 1 hos- pital admission, only the first episode was considered. Pa- In-hospital and Associated Mortality tients younger than 16 years at the time of the onset of in- fection were excluded. In-hospital mortality reflects the mortality of the underly- ing illness and the mortality attributable to BSI.26 The MICROBIOLOGICAL METHODS mortality associated with methicillin resistance in staphy- lococcal bacteremia was defined as the difference between Staphylococcus aureus was identified using standard labo- the mortality among case-patients with MRSA BSI and the ratory procedures.19,20 Methicillin resistance was deter- mortality among control patients with MSSA BSI. mined according to methods recommended by the Na- tional Committee for Clinical Laboratory Standards for disk Follow-up and Clinical Outcome diffusion testing and the use of an oxacillin (0.85%) agar screening plate.21 Patients were observed from the day of hospitalization un- til hospital discharge or death. Death was attributed defi- DEFINITIONS nitely to staphylococcal BSI in the presence of at least 1 of the following criteria: (1) blood cultures positive for S au- Bloodstream Infection reus at the time of death; (2) a persistent focus of staphy- lococcal infection associated with clinical signs of sepsis Patients were considered to have staphylococcal BSI if cultures of 2 or more blood specimens were positive for Continued on next page ARCH INTERN MED/ VOL 158, JAN 26, 1998 183 ©1998 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 such as fever, leukocytosis, or hypotension; (3) death system similar to described methods.34,35 Control patients within 14 days of the documentation of BSI without with staphylococcal bacteremia during a previous hospi- another explanation; or (4) autopsy findings indicating tal stay were excluded.