Surgical antimicrobial prophylaxis

Personal study digest for a new ID fellow. Recommendations are from the cited guideline; the "What's changed since 2013" section is reviewer synthesis of newer evidence, each claim cited. Not a substitute for the full guideline.

Sources digested: ASHP/IDSA/SIS/SHEA Clinical Practice Guidelines for Antimicrobial Prophylaxis in Surgery (2013); CDC SSI Prevention Guideline (2017); WHO Global SSI Guidelines 2nd ed. (2018)

In one line

Cefazolin 2 g IV within 60 min of incision, single dose, stop at skin closure — this one sentence covers most clean and clean-contaminated procedures.

When does a fellow think about surgical prophylaxis?

Any consult asking "what antibiotic should we give before surgery?" or "how long should we continue prophylaxis?"
Post-op SSI where you need to know what prophylaxis should have been given.
ASP review: surgeons continuing "prophylactic" antibiotics for days post-op, especially with drains/catheters still in place.
Pre-op clinic: S. aureus decolonization decisions (mupirocin ± CHG) before cardiothoracic or orthopedic implant surgery.

Core principles

Drug of choice: Cefazolin for most procedures. Adequate gram-positive and many gram-negative coverage; long half-life (1.2–2.2 h) allows q4h redosing.
Timing: Within 60 min of incision. Vancomycin and fluoroquinolones need 120 min (infusion time).
Dose: Cefazolin 2 g (3 g if ≥120 kg). Vancomycin 15 mg/kg actual body weight. Gentamicin 5 mg/kg single dose.
Redosing: Redose if procedure exceeds 2 half-lives of the agent from the pre-op dose, or blood loss >1500 mL. Cefazolin: redose at 4 h.
Duration: Single dose for most procedures. Do NOT continue beyond skin closure, even with drains or catheters in place (CDC 2017: Category IA).
Vancomycin is NOT routine: Reserve for documented MRSA colonization, outbreak/cluster, or true β-lactam anaphylaxis. Always give with cefazolin (not instead of) when used for MRSA coverage — vancomycin alone has inadequate gram-negative activity.
Decolonization: Intranasal mupirocin 2% ± CHG body wash for known S. aureus nasal carriers undergoing cardiothoracic or orthopedic implant surgery (WHO 2018: strong).

Empiric prophylaxis by procedure

Procedure	Recommended agent(s)	β-Lactam allergy alternative	Notes
Cardiac (CABG, valve, pacemaker/ICD, VAD)	Cefazolin 2 g IV	Clindamycin or vancomycin	Duration ≤48 h (48 h is an exception to the 24-h rule). Mupirocin decolonization for S. aureus carriers.
Thoracic (non-cardiac: lobectomy, VATS)	Cefazolin or ampicillin-sulbactam	Clindamycin or vancomycin	Single dose.
Orthopedic — implant (arthroplasty, ORIF, spine with instrumentation)	Cefazolin 2 g IV	Clindamycin or vancomycin	Single dose. Mupirocin for S. aureus carriers. No prophylaxis needed for clean procedures without implant.
Colorectal	Cefazolin + metronidazole IV plus MBP + oral neomycin-erythromycin (or neomycin-metronidazole)	Clindamycin + aminoglycoside/aztreonam/FQ; or metronidazole + aminoglycoside/FQ	Most complex regimen. Oral abx given as 3 doses over ~10 h the afternoon/evening before OR, after MBP. Combination oral + IV is superior to IV alone (Cochrane RR 0.55).
Appendectomy (non-perforated)	Cefoxitin, or cefazolin + metronidazole	Clindamycin + aminoglycoside/aztreonam/FQ	Single dose. If perforated → treatment, not prophylaxis.
Biliary (open) / cholecystectomy (high risk)	Cefazolin, cefoxitin, cefotetan, or ampicillin-sulbactam	Clindamycin + aminoglycoside/aztreonam/FQ; or metronidazole + aminoglycoside/FQ	Low-risk lap chole: no prophylaxis needed.
Upper GI / bariatric	Cefazolin 2 g IV	Clindamycin or vancomycin	Single dose.
Neurosurgery (craniotomy, CSF shunt)	Cefazolin 2 g IV	Clindamycin or vancomycin	Single dose. Antimicrobial-impregnated shunts not routinely recommended.
Head & neck — clean-contaminated (cancer, with oral mucosa incision)	Cefazolin + metronidazole, or ampicillin-sulbactam	Clindamycin ± aminoglycoside	≤24 h. Clean H&N (thyroidectomy): no prophylaxis.
Cesarean delivery	Cefazolin 2 g IV before incision	Clindamycin + gentamicin	Single dose. Give before incision (not at cord clamping — older practice).
Hysterectomy (vaginal or abdominal)	Cefazolin, cefoxitin, cefotetan, or ampicillin-sulbactam	Clindamycin or metronidazole + aminoglycoside/FQ	Single dose.
Urologic (clean-contaminated with urine entry)	Cefazolin or fluoroquinolone	Aminoglycoside ± clindamycin, or aztreonam	Prophylaxis indicated if entering urinary tract or if implant. Clean without prosthesis: none.
Vascular (with prosthetic graft)	Cefazolin 2 g IV	Clindamycin or vancomycin	Single dose. Includes aortic, lower extremity bypass.

Duration & stopping

Default: Single pre-operative dose; stop when the incision is closed.
Maximum 24 h for most procedures. The one exception is cardiac surgery, where ≤48 h is accepted.
Do not extend for drains, catheters, or open wounds — no evidence supports it, and prolonged prophylaxis selects resistant organisms (CDC 2017: Category IA).
Redose intraoperatively if procedure >2 half-lives from the pre-op dose (cefazolin: redose at 4 h) or blood loss >1500 mL.

Key decisions a fellow owns

Calling the stop. The most common ASP intervention: politely stopping prophylaxis at 24 h when the surgical team wants to "keep it going until the drain comes out." The evidence is clear — there is no benefit.
Allergy navigation. True IgE-mediated penicillin allergy is rare (~2% of labeled patients). For most "penicillin allergy" patients, cefazolin is safe. Reserve vancomycin/clindamycin for documented anaphylaxis or severe reaction.
MRSA prophylaxis decision. Vancomycin is not routine. Add it (to cefazolin, not instead of) only for known MRSA colonization or institutional cluster.
Colorectal prep recommendation. Combination MBP + oral antibiotics + IV prophylaxis is the evidence-based standard. Push back on IV-only prophylaxis for elective colorectal cases.
Decolonization. Recommend mupirocin ± CHG for S. aureus carriers before cardiothoracic and orthopedic implant surgery.

Special populations

Obesity (≥120 kg): Cefazolin 3 g (though the evidence for this over 2 g is debated — see "What's changed").
Renal impairment: No dose adjustment for single-dose prophylaxis. Vancomycin infusion time is the same regardless of renal function (dose adjustment applies only to repeated dosing).
True β-lactam anaphylaxis: Clindamycin or vancomycin (gram-positive), plus aminoglycoside, aztreonam, or fluoroquinolone (gram-negative) as procedure requires.
Pediatric: Same agents, weight-based dosing (cefazolin 30 mg/kg, max 2 g).

What's changed since 2013

Reviewer synthesis — not the guideline. Each claim cited.

No newer ASHP/IDSA/SIS/SHEA surgical prophylaxis guideline — the 2013 document remains the definitive U.S. multi-society guideline. Still current as the anchor reference.
CDC 2017 and WHO 2018 SSI prevention guidelines reinforce and extend the 2013 framework. CDC (JAMA Surg 2017; doi:10.1001/jamasurg.2017.0904) upgraded "no post-closure prophylaxis, even with drains" to Category IA with high-quality evidence. WHO 2018 (2nd ed.) added a strong recommendation for mupirocin decolonization in S. aureus carriers undergoing cardiothoracic and orthopedic surgery, and formally recommended MBP + oral antibiotics for colorectal.
ACS/SIS 2016 Update (Ban et al., J Am Coll Surg 2017; PMID 27915053) — 22 consensus recommendations largely concordant with the 2013 guideline.
Colorectal MBP + oral antibiotics — evidence massively strengthened. The 2013 guideline already recommended combination oral + IV for colorectal. Since then, multiple meta-analyses have confirmed the benefit: a 2025 meta-analysis of 9 RCTs in CRC patients found RR 0.55 for SSI and OR 0.45 for anastomotic leak with MBP+OAB vs MBP alone (Gosavi et al., Colorectal Dis 2025; doi:10.1111/codi.70263); a 2026 meta-analysis of 12 RCTs (n=4073) confirmed OR 0.53 (Guadalajara et al., Medicina 2026; doi:10.3390/medicina62061161). The open question is whether oral antibiotics alone (without MBP) are sufficient — a Canadian RCT is enrolling (Ghuman et al., Colorectal Dis 2024; doi:10.1111/codi.17037).
Vancomycin addition for routine arthroplasty — negative trial. The VANCO trial (Peel et al., NEJM 2023; PMID 37851875): 4113 arthroplasty patients without known MRSA colonization randomized to vancomycin 1.5 g + cefazolin vs placebo + cefazolin. SSI 4.5% vs 3.5% (RR 1.28, p=0.11). No benefit — confirms the 2013 position that vancomycin should be reserved for MRSA-colonized patients. Practice-confirming RCT
Cefazolin 3 g for ≥120 kg — evidence questioned. A 2022 systematic review found no clinical outcome data supporting >2 g prophylactic cefazolin in obese patients; PK studies show no incremental adipose tissue concentration with the 3-g dose (Schreiter et al., Obes Surg 2022; doi:10.1007/s11695-022-06196-5). The 3-g dose remains in the 2013 guideline, and CDC 2017 made no recommendation on weight-based dosing ("insufficient RCT evidence"). Open/unresolved — continue to follow institutional policy.
FDA 2016 fluoroquinolone boxed warning. The 2013 guideline lists FQ alternatives (ciprofloxacin, levofloxacin) for β-lactam-allergic patients. Since 2016, FQs carry a boxed warning for tendinitis, tendon rupture, peripheral neuropathy, and CNS effects. FQs remain an option but should be considered only when no safer alternative exists.

Anki cards minted this run

Cefazolin timing and dosing for surgical prophylaxis — the universal default
Post-closure prophylaxis duration — stop at incision closure, even with drains
Colorectal prophylaxis regimen — combination MBP + oral + IV
VANCO trial result — vancomycin addition not beneficial in routine arthroplasty

Sources

[1] Bratzler DW, Dellinger EP, Olsen KM, et al. Clinical practice guidelines for antimicrobial prophylaxis in surgery. Am J Health-Syst Pharm. 2013;70:195-283. doi:10.2146/ajhp120568
[2] Berríos-Torres SI, Umscheid CA, Bratzler DW, et al. Centers for Disease Control and Prevention guideline for the prevention of surgical site infection, 2017. JAMA Surg. 2017;152(8):784-791. doi:10.1001/jamasurg.2017.0904
[3] World Health Organization. Global guidelines for the prevention of surgical site infection, 2nd ed. Geneva: WHO; 2018.
[4] Ban KA, Minei JP, Laronga C, et al. ACS/SIS surgical site infection guidelines — 2016 update. J Am Coll Surg. 2017;224(1):59-74. PMID 27915053
[5] Gosavi R, Tan R, Teoh W, et al. MBP plus oral antibiotics reduces SSI and anastomotic leak in elective CRC surgery: systematic review and meta-analysis of RCTs. Colorectal Dis. 2025;27(10):e70263. doi:10.1111/codi.70263
[6] Guadalajara H, Putan A, García Arranz M, et al. Adding preoperative oral antibiotics to MBP reduces SSIs: meta-analysis of RCTs. Medicina. 2026;62(6):1161. doi:10.3390/medicina62061161
[7] Peel TN, et al. Trial of vancomycin and cefazolin as surgical prophylaxis in arthroplasty. N Engl J Med. 2023;389:1488-1498. doi:10.1056/NEJMoa2301401
[8] Schreiter J, et al. Prophylactic cefazolin dosing in obesity — a systematic review. Obes Surg. 2022;32:3836-3846. doi:10.1007/s11695-022-06196-5