Supplementary Material.

Study design and enrollment issues.

Design: Open-label randomized clinical trial.

Setting: Ten urban tertiary-care Spanish hospitals.

Period of study: From October 2009 to December 2014.

Inclusion criteria: Adults with suspected or confirmed MRSA infective endocarditis (IE; native or prosthetic valve, pacemaker/defibrillator) according to the modified Duke criteria [1], prosthetic vascular graft infection or complicated bacteremia (CB; septic thrombophlebitis, soft tissue-skin infection, pneumonia, osteomyelitis, or unknown source CB).

Exclusion criteria: Aged under 18 years, antibiotics with anti-MRSA activity received for >72h, shock or hypotension, urgent surgery needed, active intravenous drug use, vancomycin trough levels >15 µg/mL at 72h in patients with chronic renal insufficiency or hemodialysis, MRSA strains with fosfomycin MIC >64mg/L or vancomycin MIC ≥2mg/L or a known allergy to vancomycin, fosfomycin or imipenem.

Intervention: After providing informed consent, patients were randomized to receive vancomycin (30–45 mg/kg daily IV divided into 2–3 doses, trough levels ≥15 mg/L) or fosfomycin (2 g/6h IV) plus imipenem (1 g/6h IV), adjusted for renal function. In case of treatment failure or renal failure with vancomycin, the patient was switched to the other treatment arm.

Scheduled length of treatment: Two weeks in cases of CB with rapid control of the source and negative first control BC, four weeks for non-complicated native valve IE and pacemaker/defibrillator IE or non-rapidly controlled CB, and six weeks for prosthetic valve IE, complicated native IE and CB with osteomyelitis.

Definitions: Persistent bacteremia: positive blood cultures after seven days of treatment initiation; Complicated IE: valve prosthesis, periannular complications, systemic emboli, persistent bacteremia, and development of heart failure (NYHA >II), complete heart block; Cure: survival and absence of relapse without change of treatment arm at four and 12 weeks follow up.

Collected variables: Demographics, predisposing conditions and underlying diseases, clinical manifestations, type of CB or IE and source, previous treatment, length of antibiotics, surgery, and complications due to CB/IE or to treatment.

Outcomes: The primary endpoint was persistent bacteremia at seven days. Secondary endpoints were the clearance of blood cultures at 72h after the initiation of study treatment, the development of secondary effects related to the study drug, the development of C. difficile diarrhea or Gram-negative rods superinfection, relapses and mortality during treatment, at four or at 12 weeks of follow up.

Sample size: The estimated initial sample size was 25 patients in each arm of the study, assuming that at day seven, the duration of bacteremia would be at least 50% in patients receiving vancomycin and less than 15% in those treated with fosfomycin and imipenem, with a power of 80% and a level of significance of 5%.

Reasons for stopping the enrollment phase before the targeted sample size was achieved:

Initially, this RCT aimed to evaluate the efficacy and safety of fosfomycin plus imipenem vs. vancomycin in 50 cases of MRSA IE (25 per study arm). The estimated study completion date was December 2011. However, due to a drop in MRSA IE cases attending participating centers, a protocol amendment was solicited and approved in 2010 to widen the inclusion criteria so that patients with complicated MRSA CB could also be recruited. Despite this, the recruitment turned out to be much more difficult than expected, and, although four extensions were approved (2011, 2012, 2013 and 2014), only 7.46% (15/201) of the patients assessed could eventually be enrolled in the study. The funding entity therefore closed the study in 2015.

Details on the design and other issues related to FOSIMI trial can be consulted at:

References

1. Li JS, Sexton DJ, Mick N, Nettles R, Fowler VG Jr, Ryan T, Bashore T, Corey GR. Proposed modifications to the Duke criteria for the diagnosis of infective endocarditis. Clin Infect Dis. 2000;30:633-8.

Supplementary Table S1. Main baseline characteristics and outcomes of Methicillin-resistant Staphylococcus aureus episodes treated with Fosfomycin plus Imipenem and with vancomycin.

Fosfomycin plus Imipenem
(N=8) / Vancomycin
(N=7)
Type of bacteremia, n (%):
  • Infective endocarditis
  • Complicated bacteremia
/ 4 (50%)
4 (50%) / 4 (57%)
3 (43%)
Age, median (IQR)
Female sex, n (%): / 83.5 (67-86)
4 (50%) / 76.0 (71-80)
2 (29%)
Predisposing conditions and underlying diseases, n (%):
  • Diabetes mellitus
  • Chronic lung disease
  • Ischemic heart disease
  • Chronic renal failure
  • Chronic liver disease
  • History of cancer
  • HIV infection
  • Previous intravenous drug use
  • Congenital heart disease, history of infective endocarditis
  • Pacemaker/cardiac defibrillator
  • Valve prosthesis
  • Age-adjusted Charlson score, median (IQR)
/ 3 (38%)
5 (63%)
1 (12%)
1 (12%)
1 (12%)
4 (50%)
0
0
0
0
2 (25%)
8.0 (8-12) / 3 (43%)
0
3 (43%)
1 (14%)
0
1 (14%)
1 (14%)
0
1 (14%)
3 (43%)
2 (29%)
6.0 (4-8)
Presumed mode of acquisition, n (%):
  • Nosocomial
  • Non-nosocomial health care associated
  • Community acquired
/ 4 (50%)
3 (38%)
1 (12%) / 4 (57%)
3 (43%)
0
Type of endocarditis, n (%):
  • Native valve
  • Prosthetic valve
  • Pacemaker lead/intracardiac device
/ 2 (25%)
2 (25%)
0 / 1 (14%)
2 (29%)
1 (14%)
Outcomes, n (%):
  • Positive blood cultures at day 7 of study treatment
  • Switch of treatment arm at day 7
  • Secondary effects related to study drug
  • Gram negative rods superinfection
  • In-hospital mortality
  • Mortality at 12 weeks after study drug completion
  • Relapse at 12 weeks
  • Cure at end of study
/ 0
0
1 (12%)
0
4 (50%)
4 (50%)
0
4 (50%) / 1 (14%)
2 (29%)
3 (42%)
4 (57%)
0
1 (14%)
1 (14%)
3 (42%)

Supplementary Figure S1. Patients’ inclusion disposition; IE=Infective endocarditis; CB=Complicated bacteremia; FOS=Fosfomycin; IMI=Imipenem.