Recommended Antibiotics for Sepsis
Administer broad-spectrum IV antibiotics within 60 minutes of recognizing sepsis or septic shock, using empiric therapy that covers all likely bacterial pathogens including gram-positive, gram-negative, and potentially resistant organisms. 1, 2
Timing of Administration
- IV antimicrobials must be initiated within one hour of recognition for both sepsis and septic shock - this is the single most critical intervention for reducing mortality. 1, 2, 3
- Obtain at least two sets of blood cultures (aerobic and anaerobic) before antibiotics, but never delay antimicrobials beyond 45 minutes waiting for cultures. 2, 4
Empiric Antibiotic Selection
First-Line Broad-Spectrum Regimens
For septic shock, use combination therapy with at least two antibiotics from different antimicrobial classes: 1
Recommended anti-pseudomonal β-lactams include: 5
- Piperacillin-tazobactam 4.5 g IV every 6-8 hours (preferred dosing: 3.375 g every 6 hours achieves higher time above MIC than 4.5 g every 8 hours) 1, 6
- Cefepime (4th generation cephalosporin) 5
- Meropenem 1 g IV every 8 hours (most consistently achieves adequate serum concentrations in early sepsis) 7, 5, 8
- Imipenem 5
Aminoglycoside option: 9
- Gentamicin (dose adjusted for renal function and weight)
Fluoroquinolone option: 10
- Levofloxacin (covers gram-positive and gram-negative bacteria)
Coverage Considerations
- Add vancomycin or linezolid if MRSA is suspected (healthcare-associated infection, known MRSA colonization, severe skin/soft tissue infection). 1
- Add antifungal coverage (e.g., anidulafungin, caspofungin) if risk factors for invasive candidiasis exist (immunosuppression, prolonged ICU stay, total parenteral nutrition, broad-spectrum antibiotics). 1
- Ensure anaerobic coverage for intra-abdominal infections (piperacillin-tazobactam, meropenem, or imipenem provide this; add metronidazole if using cefepime). 1, 3
Specific Pathogen Considerations
- For Pseudomonas aeruginosa bacteremia with respiratory failure and septic shock: Use extended-spectrum β-lactam PLUS aminoglycoside or fluoroquinolone. 1
- For bacteremic Streptococcus pneumoniae with septic shock: Use β-lactam PLUS macrolide combination. 1
- For suspected Staphylococcus aureus bacteremia: Ensure anti-staphylococcal coverage (vancomycin for MRSA, nafcillin/oxacillin for MSSA). 1, 9
Dosing Optimization
- Use loading doses for vancomycin (25-30 mg/kg actual body weight) to rapidly achieve therapeutic levels in septic shock due to expanded extracellular volume from fluid resuscitation. 1
- Consider extended or continuous infusions of β-lactams (after initial bolus) to maximize time above MIC, particularly for resistant organisms. 1
- Standard dosing regimens may be insufficient in early severe sepsis - meropenem achieves adequate concentrations in 75% of patients, but piperacillin-tazobactam only in 44%, ceftazidime in 28%, and cefepime in 16%. 8
De-escalation Strategy
- Reassess antimicrobial therapy daily for potential narrowing once pathogen identification and sensitivities are available. 1, 4
- Discontinue combination therapy within 3-5 days in response to clinical improvement and/or evidence of infection resolution. 1
- De-escalate to single-agent targeted therapy as soon as susceptibility profiles are known. 1
Duration of Therapy
- 7-10 days is adequate for most serious infections associated with sepsis and septic shock. 1
- Longer courses are appropriate for: 1
- Slow clinical response
- Undrainable foci of infection
- Staphylococcus aureus bacteremia
- Fungal and viral infections
- Immunodeficiency including neutropenia
- Use procalcitonin levels to support shortening duration in patients without confirmed infection. 1
Critical Pitfalls to Avoid
- Do not use combination therapy routinely for sepsis without shock - reserve for septic shock only. 1
- Do not use combination therapy routinely for neutropenic sepsis/bacteremia - this does not improve outcomes. 1
- Do not delay antibiotics to obtain cultures - mortality increases 8% for each hour delay before antibiotics are started. 3
- Recognize that 1 in 3 patients treated for suspected sepsis may not have bacterial infection, and 4 in 5 with confirmed infection receive broader antibiotics than necessary in retrospect, contributing to antibiotic-associated complications in 1 in 6 patients. 11
- Consider local resistance patterns and hospital antibiograms when selecting empiric therapy, as standard regimens may be inadequate for multidrug-resistant organisms. 3, 5