Surgical Site Infection

DEFINITIONS

Cellulitis

Infection-related erythema of skin without drainage or fluctuance.
Abscess
Localised collections of purulent fluid within tissue.
Necrotising soft tissue infections
Widely invasive infections that rapidly cause tissue necrosis
- myonecrosis where underlying muscle involved.
- exceedingly unusual in post-op period (Barie).

Superficial / Deep / Organ Space SSI

(Center for Disease Control Definition)

D E A B M I M


EPIDEMIOLOGY

Incidence

~3% of all surgical procedures.
Up to 20% pts undergoing emergency intra-abdominal procedures.

Note on Wound Contamination
Risk stratification for SSI

1. Clean

Only integumentary and musculoskeletal soft tissues affected.
2. Clean-contaminated
Hollow viscus opened under controlled circumstances.
3. Contaminated
Bacteria introduced extensively into normally sterile tissue, brief so that infection would not be established during surgery.
- eg penetrating abdo trauma, enterotomy during adhesiolysis for mechanical bowel obstruction
4. Dirty
Surgery is performed to control established infection.

Risk factors
* = major recognised factors
Risk increases with number of risk factors irrespective of contamination
- and almost without regard of type of operation.

Personal

Age
*ASA
Smoking
*Wound contamination classification

Environmental
Inadequate skin antisepsis
Inadequate ventilation
Contaminated surgeon / equipment

Predisposing conditions
Amputations
Ascites
Chronic inflammation
Corticosteroids (controversial)
*Obesity (RR 1.78)
*Diabetes (RR 2.29)
Hypoalbuminaemia
Hypercholesterolaemia
Hypoxaemia
PVD
Post-op anaemia
Prior site irradiation
Recent operation
Remote infection
Skin carriage of staph
Skin disease around infection
Trauma (profoundly immunosuppressive)
- especially when cold, shocked, dirty
Undernutrition

Treatment factors
Drains
Emergencies
Hypothermia
Inadequate AB prophylaxis
Oxygenation
Prolonged preop hospitalisation
*Prolonged operative time
- where >75th percentile
Open surgery
- laparoscopic biliary, colon and gastric are -1 risk cf open.

National Healthcare Safety Network Risk Index
Uses wound class, ASA, length of operation >75% centile.
D E A B M I M
 

AETIOLOGY

Pathogenesis
Inoculation occurs during surgery
- either inward from skin, or outward from organ under operation.

Microbiology
Depends on type of operation, but most are:
- gram +ve cocci including Staph aureus (19%)
- coag -ve Staph epidermidis (14%)
- enterococcus species (12%)
These are skin derived mostly.
Enteric aerobic (eg E. coli 8%) or anaerobic (B. fragilis 3%) become important in pharyngoesophageal / GI surgery.
Also commonly isolated are pseudomonis (8%) and klebsiella (4%).


D E A B M I M
 

BIOLOGICAL BEHAVIOUR

Natural history
Many SSIs develop in first 5-10days.
- may develop as long as 30d post-surgery.

D E A B M I M
 

MANIFESTATIONS

Pain
Redness
Swelling / fluctuance
Ooze
Fever
Etc.
D E A B M I M


INVESTIGATIONS
Pus swabs are rather pointless due to contaminants.
- tissue spec or pus aseptically collected into a syringe are helpful.
D E A B M I M


MANAGEMENT

Prevention

Correct the correctable medical problems.
- good glycaemic control
- preop hyperglycaemia associated with increased SSI.
Allow open skin lesions to heal.
Quit smoking
Shower with antimicrobial soap the night before theatre.
- eg providone-iodine soap scrub (often omitted)
Avoid shaving the night prior.
Encourage idealisation of weight.
- if malnourished, as little as 5d of enteral nutrition reduces the SSI risk significantly.
Treat S. aureus carriage
- 2% mupirocin to the nares of carriers reduces incidence of SSI in this group.

Antibiotic prophylaxis
See card.

Operating Room Practice
Attentive to personal hygeine.
Clip hair, don't shave.
A brief rinse at the scrub bay followed by alcohol hand gel is equivalent to a long scrub routine.
- this is proven by meta-analysis of RCTs
- chlorhex better than providine as well.
20% of gloves fail at the operation
- regularly inspect.
Supplementary oygen.
Most gowns protect for 1.5-2 rs at most against strikethrough.
- may be prudent to change every 2hrs or so.
If the surgeon is a carrier of S. aureus in the nares, eliminate
- cover nose and mouth at all times
- keep unnecessary traffic to a minimum.
Avoid hypothermia(!)
- normothermia = good for wound risk; better blood flow and oxygen tension at wound.
- hypothermia vasoconstricts and impairs immune function at wound.

What is most important?
- proper antibiotics
- proper hair removal
- glucose control
- normothermia.

Managing the incision
Closure of contaminated wounds increases SSI.
- handle tissue gently.
- keep electrocautery to a minimum.

Can a contaminated incision be closed primarily?
- surgeons or pts do not like open wounds.
- evidence is mixed.
- can close muscle-splitting appendicectomy wounds.
- one large study shows large midline incisions closed primarily when contaminated failed more often with greater cost and failure rates.

Drain in the incision?
Cause more infections than they prevent in clean or clean-contaminated wounds.
- prevents epithelialisation and drain becomes a portal for introduction of bugs.
- don't use them for the incision.

Should I irrigate the wound?
Controversial. No evidence for routine washing of incisions with saline.
High pressure pulse-irrigation may be beneficial.
Topical antibiotics in the would can help but topical antiseptics probably preferred due to less resistance development.

What about future high-tech solutions?
Impregnated barriers and antibiotic-impregnated sutures in the pipeline; no conclusive cost-benefit evaluation yet.
Antimicrobial dressings of questionable benefit beyond 24 hrs when wound epithelialisation has occured.

Post-Operative Prevention

Blood transfusion
Avoid if possible - expanding body of evidence.
Even a single unit transfusion has shown a greater risk.
- increases with total transfusion volume.
- recent meta-analysis suggests triple risk of nosocomial infection from any volume of blood given (see Barie).
- cahnges in oxygen affinity, circulation time, cytokine generation probably responsible, amongst other things.
Hb concentrations >7g/dL well tolerated in most - avoid transfusion if possible.
Transfusing critically ill pts increases infections, may worsen organ dysfunction and increases mortality.

Sugers and nutrition
Hyperglycaemia impairs netrophils and phagocytosis.
- increases risk of infections and worsens sepsis outcome.
- tight control during surgery also decreases risk.
- in a large trial of critically ill post-op pts, exogenous insulin to keep glucose <11 associated with 40% mortality decrease, fewer nosocomial infections and less organ dysfunction.
Try to avoid parenteral nutrition: not greatly efficacious short term and may cause hepatic dysfunction.
- every chance use enteral feeding, including trying promotility agents like erythromycin.
- early enteral feeding within 36hrs decreases nosocomial infection by >50% in critically ill and injured pts.

Oxygenation.
Conflicting results from trials here.
- among 500pts undering elective colorectal surgery, 80% O2 during and 2-hrs post op decreased SSI by >50%.
- another trial showed a great increase post oxygenation!
Controversial until further evidence availabl.e

Treatment
Incise and drain.
- then basic wound care: topical saline soaked dressings are enough.
- no need for chemicals: can suppress fibroblast proliferation.
Take wound swabs
- increasing need in era of resistant organisms.
Treat associated conditions
- remove any necrotic material.
- control complicating factors.
Antibiotics are not required when opening and drainage achieved and no cellulitis.
- else early emperic antibiotics.

The opened wound is very large
Closure by 2o intention can be prolonged and disfiguring.
Close it again when settled.
Vacs are being used more and more commonly: no Class 1 evidence yet.

MRSA and Other Current Issues
MRSA now the leading cause of post-operative SSI in vascular patients in some leading reports
- nearing 50% of all US isolates in this context.
- associated with higher mortality, higher cost, longer stays.
Preventative bundles can include:
- nasal screening at admission, transfer or discharge
- contact isolation
- standardized hand hygiene and practices
- cultural campaign with stuff
- ongoing monitoring of process and outcome measures
Proliferation of community-acquired MRSA has significantly impacted SSI rates
Use of mupirocin for nasal decontamination if found is controversion
- no reduction in SSI in studies.

D E A B M I M


REFERENCES
Barie et al. Surg Clin N Am, 85(2005):1115-35.
Cameron