Surgical Site Infection

Meet the patient

A 64-year-old diabetic man is on day 7 after a sigmoid colectomy for cancer. The SHO calls you to the ward because the wound has "started to weep". The lower third of the incision is angry red, hot, and draining thick yellow pus that has soaked through the dressing. He has a low-grade fever and his CRP, which had been falling, has climbed back up.[1]

Two questions now decide his next hour, and they decide every SSI: how deep does this go? (the bedside exam and imaging tell you), and is this simple cellulitis or a collection needing a knife? (the depth and the pus tell you). Hold those two questions and the whole topic slots into place.[3]

What SSI is — 30 days, 90 days, and a definition that dates to 1992

Surgical site infection is any infection at or near the incision, or in any deep organ or space manipulated at operation, arising within 30 days — or within 90 days if an implant is in place. The implant list to memorise: prosthetic joint, vascular graft, mesh, cardiac device, breast prosthesis. The clock starts at the operation, not at discharge.[3]

The operational definition every surveillance programme still uses is Horan et al. 1992, reaffirmed unchanged by the CDC 2017 Prevention Guideline (Berríos-Torres). A wound is an SSI if any one of five things is true, and you should be able to recite them at viva:[3]

[3]

The classic trap: purulent drainage alone is SSI by CDC criteria — you do not need a fever, a raised CRP, or a positive culture to call it. If the wound is weeping pus within 30 days, it is an SSI until proven otherwise, and the surgeon who waits for the "full picture" has already delayed source control.[3]

Three depths — "skin, fascia, or cavity"

Depth is the single most important bedside distinction, because depth decides the management. Everything turns on one question: is the infection confined to the body wall, or has it reached a deeper cavity or organ? Body-wall infection you can open and dress; cavity infection needs imaging, drainage, and often re-operation.[1]

[1]

The one-line discriminator to carry into the viva: superficial you can see and dress; deep you must open; organ-space you must drain or re-operate. The depth, not the organism, sets the first move.[1]

The wound-contamination ladder — clean under 2 to dirty over 27

Before you ever cut, the predicted SSI rate is set by what you are about to contaminate. The four-class wound scheme — originally the National Research Council in 1964, refined by CDC — predicts baseline risk and tells you which antibiotic to choose. Know the percentages cold; examiners love them.[2]

[2]

The number rule to memorise: clean under 2, clean-contaminated 3 to 7, contaminated 10 to 15, dirty over 27. A clean wound that becomes infected should make you ask what broke — usually the prophylaxis timing, the technique, or the patient's defences.[2]

The NNIS risk index — A, C, W (ASA, Contaminated, Wall-clock)

Raw SSI rates lie, because dirty cases infect more than clean ones. To compare surgeons and hospitals fairly you adjust for case-mix with the NNIS or NHSN risk index — a 0 to 3 score that adds one point for each of three things. A patient scoring 3 carries up to a ten-fold higher SSI risk than one scoring 0, which is why NNIS-adjusted rates underpin public reporting in the USA and NHSN surveillance.[2]

[2]

The 75th-percentile cutoff is published per procedure in NHSN tables, so "long" for a hernia is not "long" for a Whipple. The score is the framework examiners want when they ask how you would compare two surgeons' infection rates honestly.[1]

How common, and who pays — the preventable burden

SSI is the commonest nosocomial infection in surgical patients, and most of it is preventable. In the CDC NHSN 2014 multistate point-prevalence survey, healthcare-associated infection prevalence in US acute-care hospitals was about 4 percent, and SSI carried a major share in surgical patients. Rates in low- and middle-income countries run two to three times higher than in high-income settings.[9]

The harm is not subtle. SSI doubles perioperative mortality (attributable mortality 1 to 3 percent, far higher for mediastinitis and organ or space infection), lengthens stay by 7 to 10 days, and raises readmission risk two- to ten-fold. The bill is real: the UK study by Coello et al. costed each case at an extra 6.5 to 11 days of stay and £814 to £10,523, with US estimates of 10,000 to 25,000 dollars per infection.[18]

Consultant confession: when a ward round is shortened by an SSI, the patient pays in days and disability and the hospital pays in pounds — and almost all of it was sitting in a checklist that somebody skipped. The economics are why prevention is funded, but the patient is why it matters.[18]

Who gets it — the patient risks you screen for in clinic

Some patients arrive infected before the first cut. The high-yield list, in the order you should screen for it at pre-assessment, with the mechanism each one works through:[1]

[1]

The ones you can fix before the date — glucose, smoking, weight, nutrition, and known MRSA carriage — are the ones that turn a high-risk patient into a lower-risk one. Elective surgery in a smoker with an HbA1c of 9 percent is a decision to accept a preventable infection.[13]

The procedure risks you control on the day

On the operating list, the surgeon and the team own most of the remaining risk. These are the levers, and most of them map directly onto the prevention bundle further down:[2]

• Wound class — clean to dirty, the ladder above; you cannot change it, but you plan for it.[2]
• Operation duration — risk climbs roughly linearly with time on the table.
• Approach — laparoscopic and minimally invasive surgery carries a lower SSI rate than open.[1]
• Prophylactic antibiotic timing and choice — the single most evidence-backed lever, and the one most often missed.[6]
• Perioperative hypothermia — a core temperature under 36 degrees Celsius doubles the SSI rate.[10]
• Perioperative hypoxia, hypotension, inadequate resuscitation, and blood transfusion.[11]
• Hair removal by shaving — the razor makes micro-lacerations; clip instead.[16]
• Poor glycaemic control — even non-diabetics with stress hyperglycaemia are at risk.[1]
• Foreign material — sutures, drains, mesh, and prosthesis all lower the inoculum threshold for infection.
• Surgical technique — rough tissue handling, excessive diathermy, haematoma, dead space, and hypoperfusion all invite trouble.

Why it happens — inoculum versus host defence, and the biofilm trap

An SSI is a balance tipped: the bacterial load and virulence at the site overcome the patient's local and systemic defences. The classical quantitative threshold is more than 10 to the fifth power organisms per gram of tissue — below it, healthy tissue usually contains the contamination; above it, clinical infection is likely. Commit that number; it is a favourite viva stem.[2]

The trap that explains why implant surgery is uniquely vulnerable: any foreign material — suture, mesh, prosthesis — lowers that threshold by several orders of magnitude. A stitch that would be harmless in clean tissue becomes a beachhead for bacteria.[1]

The biofilm — 1000 times the resistance

Biofilm is the reason implant infections do not clear with antibiotics alone. Within hours of implantation, bacteria adhere to the surface of sutures, mesh, prosthetic joints, vascular grafts, and cardiac devices, and secrete an extracellular glycocalyx. Inside it they go metabolically quiescent as persister cells, hidden from neutrophils and up to 1000 times more resistant to antibiotics than their free-swimming planktonic cousins.[1]

That single fact drives the whole of implant surgery: chronic biofilm infection often persists despite weeks of antibiotics and frequently needs implant removal for cure. It is why a prosthetic joint infection is a months-long problem, not a prescription.[1]

The bug map — match the organism to the site

Eighty percent of SSIs come from the patient's own flora, which is why site-specific prophylaxis works. Know the flora by site, because the prophylaxis and the empiric therapy follow it.[2]

• Skin — Staphylococcus aureus, coagulase-negative staphylococci, Corynebacterium. This is why clean surgery gets cefazolin.
• Upper GI and small bowel — streptococci, lactobacilli, Enterobacterales.
• Distal small bowel and colon — Enterobacterales (E. coli, Klebsiella, Enterobacter, Proteus), enterococci, and anaerobes (Bacteroides fragilis, Clostridium, Peptostreptococcus). This is why colorectal surgery adds metronidazole.
• Vagina — lactobacilli, group B Streptococcus, anaerobes, Enterobacterales.
• Exogenous (surgical team, theatre air, instruments) — typically S. aureus, coagulase-negative staphylococci, and in contaminated theatres Gram-negatives including Pseudomonas.
• Haematogenous — a distant focus (endocarditis, line infection, dental abscess) seeds the site; rare, but classic for early prosthetic joint infection.[2]

The host response — and why oxygen is a drug

Tissue injury fires the inflammatory cascade, and neutrophils do the killing — but only if the tissue is oxygenated. Complement activates, fibrinogen becomes fibrin, platelets aggregate, and neutrophils and macrophages arrive to phagocytose bacteria through the respiratory or oxidative burst. That burst is critically dependent on tissue oxygen tension.[1]

This is the mechanistic thread that ties the whole prevention bundle together. Subcutaneous oxygen tension correlates inversely with SSI risk, which is why warming, supplemental oxygen, fluid resuscitation, pain control (to stop catecholamine-driven vasoconstriction), and smoking cessation all work through the same final common pathway — delivering oxygen to a healing wound.[10]

Meet the cardinal five — the signs of inflammation

SSI usually declares itself between postoperative days 3 and 10, though organ or space infections can surface weeks to months later. The local signs are the ancient five, and pain is usually the first. Cluster them once and they stay:[1]

[1]

Purulent drainage — and the differential of wound fluid

Purulent is the single most specific feature, and it is defined clinically, not in the lab. Thick, opaque, yellow, green, or brown, sometimes malodorous — and unmistakably different from the clear straw-coloured serous transudate of a seroma or the bloody drainage of a haematoma.[3]

The bedside discriminator to nail at viva: serous means seroma, sanguineous means haematoma, purulent means SSI. If you cannot tell, aspirate it — but pus is pus, and by CDC criteria it makes the diagnosis on its own.[3]

The dehiscence clock — "pink fluid" on day 5 to 8 is an emergency

Not every wound that gapes is the same emergency. Superficial dehiscence — separation of the skin edges — usually heals by secondary intention or delayed primary closure, and is a nuisance, not a crisis. Fascial or deep dehiscence is a different animal entirely.[1]

The classic presentation to tattoo into memory: a patient around postoperative day 5 to 8 suddenly feels something "give" or "pop", and a serosanguineous "pink fluid" discharge soaks the dressing. That pink fluid is peritoneal fluid escaping through a failed fascia, and viscera may be visible underneath. This is a surgical emergency — cover the wound with sterile saline-soaked gauze, give broad-spectrum IV antibiotics, and return to theatre now for resuturing and management of any underlying SSI.[1]

The recurring pitfall: a registrar who reassures himself that "it is just a seroma" when the fascia has let go. Seromas do not announce themselves with a pop on day 6. Pink fluid plus a pop is fascial dehiscence until the wound is explored.[1]

The atypical presenters — when the loudest sign is silence

In the vulnerable patient, the inflammatory response is blunted and the signs whisper. Do not be reassured by the absence of fever or pus in these groups — a high index of suspicion is the only safe stance.[1]

[1]

The differential — not every red wound is SSI

A red, swollen, or draining wound has a long differential, and the discriminator is usually in the timing and the fluid. This is the table to reproduce when asked "what else could it be?"[1]

[1]

The killer in the differential — necrotising fasciitis

A low threshold for necrotising fasciitis is the single most important reflex on this list. Missed necrotising infection carries a mortality of 30 to 70 percent and progresses by the hour, not the day. The presenting triad to never ignore: pain out of proportion to the wound appearance, crepitus or dusky skin, and rapidly progressive systemic toxicity.[1]

The LRINEC score (CRP, white cell count, haemoglobin, sodium, creatinine, glucose) helps risk-stratify, but a high clinical suspicion mandates immediate surgical exploration regardless of the score. Do not wait for imaging or cultures — the debridement is both the diagnostic and the therapeutic step. A wound that hurts more than it looks is necrotising until the fascia is seen.[1]

The bedside round — inspect, probe, swab deep

Examination of a suspected SSI is a structured wound assessment, not a glance at the dressing. Remove the dressing with sterile technique under good light, and document each of these in the notes:[1]

• Erythema — extent in centimetres beyond the wound edge, and whether well-demarcated or spreading.
• Warmth, swelling, and induration — measured and described, not "mild".
• Dehiscence depth — skin only, subcutaneous tissue, fascia, or full thickness with viscera visible.
• Discharge character and volume — serous, sanguineous, serosanguineous, or purulent; malodorous or not; volume per day.
• Surrounding skin — blistering, necrosis, crepitus, lymphangitis.
• Wound edge — undermining or tunnelling depth.
• Drains — site, effluent character, patency.[1]

A wound probe gently introduced into a sinus or open cavity establishes depth and may reveal a hidden collection. Take a sterile swab from the deepest aspect, never the surface slough — the surface is colonised and will mislead you.[1]

The ASEPSIS score and the Southampton grade

For surveillance and research, two named scoring systems quantify wound severity. The ASEPSIS wound score (Wilson et al. 1990) is the most widely used and is examiner-tested — reproduce the letters:[1]

[1]

The total, scored 0 to 665, classifies healing: 0 to 10 satisfactory, 11 to 20 disturbed, over 20 severe wound infection. The Southampton system runs a simpler 0 to VI grade — 0 heals by primary intention, I to IV track inflammation, haematoma, serous and purulent discharge, V is pus under the skin, and VI is a deep cavity with or without tissue breakdown.[1]

Investigations — swab deep, image the cavity, and culture before antibiotics

The sampling principle that juniors break: culture the pus or the deep tissue, not the surface. The surface is colonised; a surface swab grows what lives there, not what infects the wound.[1]

• Pus or deep tissue (not a surface swab) for Gram stain, culture, and sensitivity.
• Anaerobic transport if the wound is foul-smelling or anaerobes are suspected — anaerobes die in air.
• MRSA screen (nose, axilla, groin, perineum) on admission for high-risk procedures.[13]

Bloods track severity and guide response. CRP typically rises from day 1 to a peak around day 3, then falls — a CRP that fails to fall or rises after day 3 is highly suspicious for SSI, and a normal CRP by day 4 to 5 has a high negative predictive value. Add a lactate if sepsis is suspected (over 2 mmol per litre), and baseline U&E, LFT, glucose or HbA1c, and albumin for glycaemic and nutritional status.[1]

Take two sets of blood cultures — peripheral, separate venepunctures, before antibiotics — in any patient with fever over 38.5 degrees Celsius, systemic sepsis, or immunocompromise. Antibiotic susceptibility results then guide de-escalation; request MRSA, ESBL, and carbapenemase status as appropriate.[1]

Imaging is chosen by the question you are asking. Bedside ultrasound finds superficial, abdominal-wall, and pelvic collections and can guide drainage — useful in children, pregnant women, and the critically ill. CT with IV contrast and water-soluble oral and rectal contrast is the gold standard for intra-abdominal and pelvic collections, anastomotic leak, and gas-forming infection — use water-soluble, never barium, if perforation or leak is suspected. MRI is preferred for spinal hardware, prosthetic joint, and extremity soft-tissue infection, with high sensitivity for osteomyelitis. Nuclear medicine (white-cell-labelled scan, FDG-PET or CT) is reserved for chronic prosthetic joint infection when conventional imaging is indeterminate.[1]

For organ or space infections, special tests close the diagnosis: a joint aspirate with synovial white cells over 3000 per microlitre and neutrophil predominance suggests prosthetic joint infection, and multiple deep tissue biopsies (3 to 6) at revision arthroplasty feed the modified Birmingham criteria. A sinogram — contrast injected into a sinus tract — may delineate a deep collection or a connection to a viscus.[1]

Resuscitation — sepsis first, source control with it

Most SSIs are not immediately life-threatening — but a deep or organ or space SSI presenting with sepsis is a time-critical emergency. Run the Surviving Sepsis Campaign 1-hour bundle, and do not let source control wait for "stability" that will not come without it.[1]

[1]

The sentence that earns marks and saves lives: source control is not a substitute for antibiotics, and antibiotics are not a substitute for source control — both are required. Do not delay source control waiting for stability in necrotising infection, fascial dehiscence with exposed viscera, or an uncontrolled leak.[1]

The management ladder — depth decides the move

Once resuscitated, management follows the depth, and the depth alone. Three steps, three depths, three different first moves.[1]

Step 1 — Superficial incisional SSI

• Open the wound — remove the sutures or staples over the infected segment and let it drain.
• Send pus for Gram stain, culture, and sensitivity.
• Irrigate with normal saline and pack lightly with saline-soaked gauze, or use a hydrofibre or hydrocolloid dressing (alginates for highly exudative wounds).
• Daily dressing change (or every 48 to 72 hours for hydrocolloids); document dimensions, exudate, and tissue type.
• Targeted antibiotics only if there is surrounding cellulitis or systemic signs: oral flucloxacillin 500 mg four times daily for 5 to 7 days for suspected MSSA; doxycycline 100 mg daily or clindamycin 300 mg four times daily if penicillin-allergic; add co-amoxiclav 625 mg three times daily or metronidazole if a GI or GU source is suspected.
• Re-evaluate at 48 hours — if not improving, reconsider the diagnosis (deeper infection? resistant organism? necrotising?), escalate to imaging, IV antibiotics, and surgical review.[1]

Step 2 — Deep incisional SSI

• Open the entire wound, not just the infected segment.
• Surgical debridement of all necrotic tissue, slough, and non-viable fascia; explore for and drain any associated collection.
• Daily surgical dressing change, or a vacuum dressing.
• IV antibiotics: flucloxacillin 1 g every 6 hours for MSSA; vancomycin 15 to 20 mg per kg every 12 hours (trough 15 to 20) for MRSA or beta-lactam allergy; add piperacillin-tazobactam or ceftriaxone plus metronidazole for a polymicrobial or GI source.
• Image to exclude extension to organ or space (CT abdomen and pelvis).
• Duration typically 7 to 14 days; de-escalate once sensitivities are back.[1]

Step 3 — Organ or space SSI

• Source control is paramount: percutaneous image-guided drainage (CT or ultrasound) for accessible collections — preferred over surgery where feasible; re-operation for collections not amenable to percutaneous drainage, for anastomotic leak with peritonitis, for retained necrotic tissue, or for failed percutaneous drainage.
• Broad-spectrum IV antibiotics as for the resuscitation bundle; de-escalate to targeted therapy on cultures.
• Manage the underlying cause — an anastomotic leak may need resection and stoma, a prosthetic joint infection may need one-stage or two-stage revision, mediastinitis may need rewiring and flap cover.
• Duration 7 to 14 days for most organ or space infections; 4 to 6 weeks for osteomyelitis, mediastinitis, and prosthetic joint infection, with infectious diseases input.[1]

Negative-pressure wound therapy — minus 125 mmHg, and when to use it

NPWT (the VAC) delivers continuous negative pressure, usually minus 125 mmHg (range 50 to 175), to an open wound. It removes exudate, increases local blood flow, reduces oedema, promotes granulation, contracts the wound edge, and cuts dressing-change frequency. Indications: open abdomen, dehisced wounds, large defects after debridement, complex perineal wounds after abdominoperineal resection, diabetic foot ulcers, and burns.[4]

The growing use is prophylactic closed-incision NPWT (ciNPWT) on high-risk closed incisions — obese patients, midline laparotomy, sternotomy in obese diabetics, and groin incisions after vascular surgery — where it measurably reduces SSI. The WHO intraoperative and postoperative guideline endorses this approach for high-risk wounds.[4]

Antibiotic duration — short is the new normal

The modern principle is the shortest effective course, de-escalated as soon as sensitivities return. Prolonged antibiotics select resistance and Clostridioides difficile without added benefit.[1]

• Superficial with cellulitis — 5 to 7 days oral.
• Deep — 7 to 14 days IV, transition to oral once afebrile and improving.
• Organ or space with source control — 4 to 7 days after source control for most intra-abdominal infections (short courses are equivalent to longer); 4 to 6 weeks IV for osteomyelitis, mediastinitis, and prosthetic joint infection.
• De-escalate as soon as sensitivities are available; engage antimicrobial stewardship.[1]

The escalation triggers — when to worry at 48 to 72 hours

Failure to improve within 48 to 72 hours of appropriate therapy is a signal, not a setback to wait out. Reconsider the diagnosis, repeat the imaging, and seek surgical review. Worsening pain, spreading erythema, new crepitus, or systemic deterioration mean suspect necrotising infection or an uncontrolled source — emergency surgical review, not another dressing change. New organ dysfunction means ICU.[1]

The named subtypes that bite

Orthopaedic prosthetic joint infection

Biofilm is the defining feature, and biofilm is why retention usually fails for chronic infection. The commonest organisms are coagulase-negative staphylococci and S. aureus, with Streptococcus, Enterococcus, Gram-negatives, and anaerobes in the mix.[1]

Diagnosis is clinical (pain, swelling, sinus tract) plus a raised CRP and ESR, a joint aspirate (synovial white cells over 3000 per microlitre with neutrophil predominance is suggestive), and multiple deep tissue biopsies at revision surgery — 3 to 6 samples, read against the modified Birmingham criteria. Management splits by timing: acute infection (under 3 weeks post-op) gets DAIR — Debridement, Antibiotics, Irrigation, and Retention of the implant — plus IV antibiotics such as rifampicin with a fluoroquinolone for staphylococci, for 6 weeks. Chronic infection gets two-stage exchange — remove the implant, place an antibiotic-loaded cement spacer, give 6 weeks of IV antibiotics, then reimplant — which remains the gold standard. One-stage exchange is an option in selected centres. If the implant cannot be removed and the patient is not fit for surgery, suppressive antibiotics continue for life.[1]

Colorectal surgery — day-5 to day-7 fever is a leak until proven otherwise

The commonest organ or space SSI after colorectal surgery is an anastomotic leak with pelvic abscess. The organisms are enteric — E. coli, Klebsiella, Enterobacter, enterococci, and Bacteroides fragilis — which is why prophylaxis is a cephalosporin (cefuroxime or ceftriaxone) plus metronidazole.[1]

The trap that costs lives: fever, tachycardia, and ileus after a colorectal anastomosis around day 5 to 7 is an anastomotic leak until proven otherwise, not "a minor ileus". Image with CT using water-soluble oral and rectal contrast. Generalised peritonitis means emergency re-operation with diversion (Hartmann's); a localised pelvic abscess may drain percutaneously.[1]

Prevention has reversed itself in the last decade: combined mechanical and oral antibiotic bowel preparation (polyethylene glycol prep plus oral neomycin and metronidazole) reduces SSI more than IV antibiotics alone, overturning the old teaching that bowel prep was unnecessary.[1]

Caesarean section — pre-incision, not cord-clamp

The SSI rate after caesarean is 3 to 15 percent, higher after emergency caesarean in labour. The organisms are endogenous vaginal flora — group B Streptococcus, anaerobes, E. coli — with S. aureus, including MRSA, from the skin.[14]

The practice change every candidate must name: pre-incision IV cephalosporin — cefazolin 2 g IV (3 g if body weight over 120 kilograms) — is given before skin incision, not after cord clamping. The Mackeen Cochrane review confirmed this reduces postpartum endometritis and wound infection without neonatal harm, and it is now the global standard. Add azithromycin 500 mg IV for emergency caesarean in labour. For skin preparation, 2 percent chlorhexidine-alcohol is preferred to povidone-iodine.[14]

Endometritis is the organ or space equivalent — fever, uterine tenderness, and foul lochia — treated with IV clindamycin plus gentamicin (or co-amoxiclav).[17]

Cardiac surgery — mediastinitis and the decolonisation win

Deep SSI after sternotomy is mediastinitis, and it carries a mortality of 10 to 40 percent. The organisms are S. aureus (especially in nasal carriers), coagulase-negative staphylococci, and Gram-negatives.[13]

The prevention win to cite by name: pre-op screening for S. aureus nasal carriage, with carriers decolonised using mupirocin 2 percent nasal ointment twice daily plus chlorhexidine body wash for 5 days. The Bode NEJM 2010 trial showed this strategy cut deep S. aureus SSI by roughly 50 percent in cardiothoracic and orthopaedic carriers (number needed to treat 8), confirmed by the Liu Cochrane review in 2017.[13]

Prophylaxis is IV cefazolin 2 g (vancomycin 1.5 g over 60 minutes if MRSA risk or beta-lactam allergy), re-dosed at 4 hours or after 1500 mL blood loss or cardiopulmonary bypass over 2 hours. Prophylactic ciNPWT is increasingly used for high-risk sternotomy — obese, diabetic, bilateral internal mammary harvest.[15]

Vascular graft infection

Early graft infection is S. aureus and Gram-negatives; late infection is indolent coagulase-negative staphylococci, months to years later. The spectre is graft-enteric erosion — an aorto-enteric fistula presenting with a herald upper GI bleed. Diagnose with CT angiography, and a tagged white-cell scan if indeterminate.[1]

Management is graft excision plus either extra-anatomic bypass (axillo-bifemoral) or in-situ replacement with rifampicin-soaked or silver-coated Dacron, a biological (cadaveric) graft, or autogenous vein. If the graft is retained, suppressive antibiotics continue for life.[1]

Implant and mesh infections — biofilm decides the strategy

Biofilm is the central problem, so the timing of infection dictates whether you can keep the implant. Acute infection (under 3 weeks) may respond to DAIR plus targeted IV antibiotics. Chronic infection usually needs implant removal, antibiotics, and delayed reimplantation. Mesh infection after hernia repair can be tried with conservative antibiotics, but persistent infection means mesh removal — often partial.[1]

The prevention bundle — seven pillars, one mantra

SSI is preventable, and the bundle is the proof. Every pillar has a landmark trial behind it, and together they are the single highest-yield topic in surgical infection. Learn the bundle as one breath and the evidence behind each pillar.[1]

[1]

Classen 1992 — the 60-minute gold standard

One NEJM paper underpins every "within 60 minutes" rule on the planet. Classen et al. plotted SSI rate against the timing of prophylactic antibiotics in over 1700 clean and clean-contaminated operations, and the curve was unmistakable.[6]

[6]

The re-dosing rules follow from pharmacokinetics: re-dose at 4 hours or after 1500 mL blood loss, because tissue levels fall below therapeutic as the operation lengthens or blood is lost. Vancomycin and the fluoroquinolones infuse slowly, so their window is 120 minutes, not 60.[7]

Kurz, Greif, and Belda — warm and oxygenated

Two more NEJM or JAMA papers complete the physiology story: keep the patient warm, and keep the wound oxygenated. Both work through the same mechanism — tissue oxygen tension drives the neutrophil oxidative burst.[10]

[10] [11]

The oxygen controversy is named calmly: PROXI (2009) found no benefit, so the evidence is not unanimous. Current guidance keeps 80 percent FiO2 for patients with normal oxygenation because the downside is trivial and the upside, when present, is real.[12]

Bode, Liu, and Huang — decolonisation in the right patient

For S. aureus, the nose is the reservoir, and clearing it prevents the SSI. The evidence comes in three shapes — a targeted RCT, a Cochrane review, and a universal-decolonisation ICU trial.[13]

[13]

The Liu 2017 Cochrane review confirmed that nasal decontamination with mupirocin in S. aureus carriers reduces SSI.[15] The Troeman 2023 cohort quantified the stakes: preoperative S. aureus colonisation roughly quadruples the risk of postoperative S. aureus infection, which is exactly why screening the high-risk patient pays.[20]

The Huang 2013 NEJM trial took the logic one step further — in the ICU, universal decolonisation (mupirocin and chlorhexidine for every patient, without screening) reduced MRSA clinical isolates and bacteraemia more than targeted screening and isolation. The lesson for the ward is selective screening; the lesson for the ICU is sometimes to treat everyone.[19]

Haynes 2009 — the checklist that travels

The WHO Surgical Safety Checklist is the cheapest, most portable intervention on the list, and it works everywhere it is used. In a global cohort of over 7000 patients, Haynes et al. showed it cut inpatient complications from 11 percent to 7 percent and mortality from 1.5 percent to 0.8 percent — in hospitals rich and poor alike.[8]

[8]

Tanner 2021 — clip, do not shave

The razor makes micro-lacerations that seed the wound; the clipper does not. The Tanner 2021 Cochrane review settled this cleanly: there is no difference in SSI between clipping and no hair removal at all, but shaving is associated with a higher SSI rate than clipping. If hair must be removed, use clippers, not razors.[16]

Hadiati 2020 — chlorhexidine-alcohol for caesarean skin prep

For skin preparation, chlorhexidine-alcohol wins on rapid, persistent activity. The Hadiati 2020 Cochrane review found no clear difference between antiseptic agents overall for caesarean skin prep, but chlorhexidine-alcohol is generally preferred for its rapid onset and persistence — and it is the choice that maps onto the broader CDC and WHO preference.[17]

The named guidelines — who says what

Four documents set the global standard; know them by author and year. The CDC 2017 Guideline for the Prevention of SSI (Berríos-Torres et al.) superseded the 1999 CDC or HICPAC guideline (Mangram) and gives the high-priority recommendations on antibiotics, glycaemic control, normothermia, oxygenation, and hair removal that the bundle is built from.[1]

The WHO Global Guidelines for SSI Prevention (Allegranzi et al., 2016) came in two parts — 13 pre-operative recommendations and 16 intra-operative and post-operative recommendations — and was the first guideline to address low- and middle-income countries explicitly and to make recommendations on glycaemic control, anaesthetic agents, and oxygenation. The ASHP, IDSA, SIS, and SHEA surgical prophylaxis guideline (Bratzler et al. 2013) is the procedure-specific antibiotic reference — agent, dose, timing, duration, and re-dosing.[4][5]

Controversies — name them calmly

Three live debates recur in exams, and the safe answer is to name the uncertainty and quote the current guidance.[4]

• Optimal oxygen concentration — PROXI (2009) was neutral, but WHO and CDC still recommend 80 percent FiO2 intraoperatively and in early recovery for patients with normal oxygenation.[4]
• Mechanical bowel preparation — once thought useless, now vindicated: combined mechanical and oral antibiotic bowel preparation reduces SSI more than IV antibiotics alone in elective colorectal surgery.[1]
• Routine vancomycin versus cefazolin prophylaxis — cefazolin remains first-line; vancomycin is added for known MRSA carriers, beta-lactam allergy, or high local MRSA prevalence; routine dual coverage is not recommended because of nephrotoxicity and selection pressure.[7]
• Antibiotic-impregnated suture (triclosan-coated Vicryl Plus) — a modest SSI reduction in meta-analyses; WHO suggests use.[4]
• Duration of prophylaxis — a single pre-incision dose is as effective as 24-hour coverage for most procedures; prolonged prophylaxis selects resistance and C. difficile.[1]

Special populations — how the calculus shifts

Diabetes mellitus

HbA1c over 8 percent (64 mmol per mol) triples SSI risk, and perioperative glucose over 11 mmol per litre adds more. Aim for HbA1c under 69 mmol per mol (8.5 percent) before elective surgery, and consider a variable-rate IV insulin infusion for poorly controlled diabetics. Target blood glucose 6 to 10 mmol per litre perioperatively (4 to 12 acceptable); avoid both hyper- and hypoglycaemia. Continue a long-acting basal insulin (such as glargine) on the morning of surgery; omit the short-acting meal-time insulin when fasting.[1]

Obesity

BMI over 30 raises SSI risk, and over 40 raises it sharply. The mechanisms are tissue hypoxia, longer operations, under-dosed prophylaxis, and difficulty obliterating dead space. The fix is to increase the antibiotic dose — cefazolin 3 g if weight over 120 kilograms, vancomycin dosed by actual body weight — and to consider prophylactic ciNPWT for high-risk closed incisions such as midline laparotomy or panniculectomy. Close in layers to obliterate dead space.[7]

Immunosuppressed patients

Atypical and opportunistic organisms — Candida, Aspergillus, mycobacteria (M. abscessus, M. chelonae), Pseudomonas — and a blunted inflammatory response that hides them. Keep a low threshold for imaging and biopsy, involve infectious diseases and microbiology early, and cover broadly empirically (meropenem plus vancomycin, plus or minus an antifungal) until the organism is identified.[1]

Elderly

A blunted inflammatory response means atypical presentation — confusion, hypothermia, hypoglycaemia, falls — and higher mortality from comorbidity. Prioritise delirium prevention: optimise pain control, sleep, hydration, and mobility, and minimise sedatives.[1]

Pregnancy and caesarean

The caesarean SSI rate is 3 to 15 percent, higher after emergency caesarean in labour. Physiological changes increase volume of distribution and renal clearance, altering drug dosing. Pre-incision (not cord-clamp) cephalosporin — cefazolin 2 g IV (3 g if over 120 kilograms) — is safe in pregnancy and lactation and is the standard of care. Endometritis is the organ or space equivalent, treated with IV clindamycin plus gentamicin.[14]

Paediatrics

Weight-based dosing is the rule — cefazolin 30 mg per kg IV (maximum 2 g), vancomycin 15 mg per kg (maximum 1 g). Relative fluid requirements are higher, so watch the balance. Use ultrasound preferentially for imaging to avoid radiation, and involve the parents in care and consent.[1]

MRSA carriers

Screen pre-op (nasal swab) for high-risk procedures — cardiac, orthopaedic implant, neurosurgery, prolonged ICU, transfer from another hospital. Decolonise with mupirocin 2 percent nasal ointment twice daily plus chlorhexidine 4 percent body wash daily for 5 days pre-op. For prophylaxis, use vancomycin 1 to 1.5 g IV over 60 minutes or teicoplanin 400 mg IV instead of, or in addition to, cefazolin if MRSA colonisation is known. Universal decolonisation strategies reduce MRSA clinical isolates and bacteraemia in the ICU (Huang 2013).[13]

Regional deltas — the same bundle, different constraints

The diagnostic and management framework — CDC depth classes, the prevention bundle, source control — is globally consistent. What changes is the resource envelope around it.[4]

[4]

How patients come to harm — the preventable list

These are the failures that turn a clean operation into a long admission, and most are preventable.[1]

• A missed necrotising fasciitis read as cellulitis and treated with antibiotics while the fascia dies — the preventable death, with a mortality of 30 to 70 percent.[1]
• Prophylaxis given after incision — the single most common breach, and the one Classen 1992 quantified.[6]
• A fascial dehiscence on day 6 written off as a seroma when the pink fluid and the pop were the giveaway.[1]
• A day-7 fever after colorectal anastomosis treated as "minor ileus" instead of a leak until proven otherwise.[1]
• An under-dosed prophylaxis in obesity — cefazolin 2 g for a 140-kilogram patient, when 3 g was needed.[7]
• A surface swab growing colonisers, while the deep pus that would have guided therapy was never sent.[1]
• A hypothermic patient left cold because "the theatre was busy", when warming would have halved the SSI rate.[10]
• An S. aureus carrier sent to sternotomy without screening or decolonisation, when mupirocin and chlorhexidine would have halved the deep SSI risk.[13]

The mantra, and the mnemonics

[1] [1]

The mantra: SSI is preventable — antibiotics within 60 minutes, normothermia, glucose control, clip not shave, oxygen, chlorhexidine, checklist. Say it as one breath and you have the prevention marks.[1][8]

Ward-round test — four stems, thirty seconds each

Stem 1 — the day-7 colectomy wound (answer)

The 64-year-old diabetic from the opening vignette, day 7 after sigmoid colectomy, with an erythematous weeping lower wound, low-grade fever, and a rising CRP. The SHO asks whether to start oral flucloxacillin. What is the right first move? Model: First, confirm the depth and exclude a collection — this is the step the SHO is about to skip. Inspect and probe the wound, swab the deep pus, and image with CT if there is any concern about an organ or space component (a leak is the killer in this patient, presenting exactly at day 5 to 7). If it is genuinely a superficial incisional SSI, open the wound, pack it, and give oral flucloxacillin 500 mg four times daily for 5 to 7 days only if there is surrounding cellulitis or systemic signs. Re-evaluate at 48 hours. Starting antibiotics without opening the wound and without excluding a leak is the trainee error.[1]

Stem 2 — the pink fluid on day 6 (answer)

A patient six days after a laparotomy feels something "pop" while coughing, and the dressing soaks through with pink fluid. What happened, and what do you do in the next 15 minutes? Model: This is fascial dehiscence until proven otherwise — the pop and the serosanguineous pink fluid are pathognomonic, and viscera may be visible underneath. Cover the wound with sterile saline-soaked gauze, give broad-spectrum IV antibiotics, resuscitate, and return to theatre now for resuturing and management of any underlying SSI. Do not reassure yourself it is a seroma — seromas do not announce themselves with a pop on day 6.[1]

Stem 3 — the wound that hurts more than it looks (answer)

A post-operative wound looks only mildly erythematous, but the patient is in severe pain, the skin over it is dusky, and you feel crepitus. The CRP is high. What is the diagnosis and the next step? Model: This is necrotising fasciitis until proven otherwise — pain out of proportion, crepitus, dusky skin, and systemic toxicity are the classic tetrad, and the LRINEC score will be raised. Do not wait for imaging or cultures. The next step is emergency surgical exploration and debridement — the operation is both diagnostic and therapeutic, and mortality climbs by the hour. Broad-spectrum IV antibiotics run alongside, but the knife is the treatment.[1]

Stem 4 — the MRSA-positive cardiac patient (answer)

A patient listed for elective coronary artery bypass grafting screens positive for MRSA nasal carriage at pre-assessment. What do you do before the operation, and what prophylaxis do you give on the day? Model: Decolonise before surgery — mupirocin 2 percent nasal ointment twice daily plus chlorhexidine 4 percent body wash daily for 5 days pre-op (Bode 2010, which halved deep S. aureus SSI in cardiothoracic carriers). On the day, give prophylaxis active against MRSA — vancomycin 1.5 g IV over 60 minutes (or teicoplanin 400 mg IV) — instead of or in addition to cefazolin, re-dosed at 4 hours or after 1500 mL blood loss or cardiopulmonary bypass over 2 hours. Consider prophylactic ciNPWT if the sternotomy is high-risk (obese, diabetic, bilateral internal mammary harvest).[13]

References

1. [1]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.PMID 28467526
2. [2]Mangram AJ, Horan TC, Pearson ML, et al. Guideline for Prevention of Surgical Site Infection, 1999. Centers for Disease Control and Prevention (CDC) Hospital Infection Control Practices Advisory Committee Am J Infect Control, 1999.PMID 10196487
3. [3]Horan TC, Gaynes RP, Martone WJ, et al. CDC definitions of nosocomial surgical site infections, 1992: a modification of CDC definitions of surgical wound infections Infect Control Hosp Epidemiol, 1992.PMID 1334988
4. [4]Allegranzi B, Zayed B, Bischoff P, et al. New WHO recommendations on intraoperative and postoperative measures for surgical site infection prevention: an evidence-based global perspective Lancet Infect Dis, 2016.PMID 27816414
5. [5]Allegranzi B, Bischoff P, de Jonge S, et al. New WHO recommendations on preoperative measures for surgical site infection prevention: an evidence-based global perspective Lancet Infect Dis, 2016.PMID 27816413
6. [6]Classen DC, Evans RS, Pestotnik SL, et al. The timing of prophylactic administration of antibiotics and the risk of surgical-wound infection N Engl J Med, 1992.PMID 1728731
7. [7]Bratzler DW, Dellinger EP, Olsen KM, et al. Clinical practice guidelines for antimicrobial prophylaxis in surgery Am J Health Syst Pharm, 2013.PMID 23327981
8. [8]Haynes AB, Weiser TG, Berry WR, et al. A surgical safety checklist to reduce morbidity and mortality in a global population N Engl J Med, 2009.PMID 19144931
9. [9]Magill SS, Edwards JR, Bamberg W, et al. Multistate point-prevalence survey of health care-associated infections N Engl J Med, 2014.PMID 24670166
10. [10]Kurz A, Sessler DI, Lenhardt R. Perioperative normothermia to reduce the incidence of surgical-wound infection and shorten hospitalization. Study of Wound Infection and Temperature Group N Engl J Med, 1996.PMID 8606715
11. [11]Greif R, Akça O, Horn EP, et al. Supplemental perioperative oxygen to reduce the incidence of surgical-wound infection N Engl J Med, 2000.PMID 10639541
12. [12]Belda FJ, Aguilera L, García de la Asunción J, et al. Supplemental perioperative oxygen and the risk of surgical wound infection: a randomized controlled trial JAMA, 2005.PMID 16249417
13. [13]Bode LG, Kluytmans JA, Wertheim HF, et al. Preventing surgical-site infections in nasal carriers of Staphylococcus aureus N Engl J Med, 2010.PMID 20054045
14. [14]Mackeen AD, Packard RE, Ota E, et al. Timing of intravenous prophylactic antibiotics for preventing postpartum infectious morbidity in women undergoing cesarean delivery Cochrane Database Syst Rev, 2014.PMID 25479008
15. [15]Liu Z, Norman G, Iheozor-Ejiofor Z, et al. Nasal decontamination for the prevention of surgical site infection in Staphylococcus aureus carriers Cochrane Database Syst Rev, 2017.PMID 28516472
16. [16]Tanner J, Melen K Preoperative hair removal to reduce surgical site infection Cochrane Database Syst Rev, 2021.PMID 34437723
17. [17]Hadiati DR, Hakimi M, Nurdiati DS. Skin preparation for preventing infection following caesarean section Cochrane Database Syst Rev, 2020.PMID 32580252
18. [18]Coello R, Charlett A, Wilson J, et al. Adverse impact of surgical site infections in English hospitals J Hosp Infect, 2005.PMID 15866006
19. [19]Huang SS, Septimus E, Kleinman K, et al. Targeted versus universal decolonization to prevent ICU infection N Engl J Med, 2013.PMID 23718152
20. [20]Troeman DPR, Hazard D, Timbermont L, et al. Postoperative Staphylococcus aureus Infections in Patients With and Without Preoperative Colonization JAMA Netw Open, 2023.PMID 37906196