SSI – It’s not what you do, but the way that you do it?

20140801-135-see-rock_festival_2014-rick_parfittAnother of my favourite guitarists succumbed to sepsis following surgery just before Xmas, with the sad passing of Rick Parfitt following shoulder surgery, spookily the day after his band Status quo performed their final electric gig (which I was at). The other was Rory Gallagher, who died a few years ago now of MRSA.  Surgical procedures are normally carried out under what should be the most controllable of conditions, yet there are variations in practice, a paucity of quality studies on even the most basic of interventions (such as pre-op bathing) and even when there is good evidence, it is ignored. However I do also wonder if we have been missing something. A paper that suggests no difference between Chlorhexidine (CHG) and Povidone Iodine (PI) for pre-surgical skin prep (both aqueous) recently piqued my interest. It was an RCT (non-blinded) undertaken in clean-contaminated upper gastrointestinal or hepatobiliary–pancreatic open surgery, however that wasn’t the aspect that I became interested in.In many countries, guidelines indicate that preoperative skin prep should be based in alcohol (adding a potential confounder to any study). This would undoubtedly assist with drying, however in this paper aqueous solutions were used and were allowed to ‘dry’ for three minutes before nurses then ‘dried the skin’ with a sterile towel. I do wonder about standard practice before the study as interestingly the pre-intervention SSI rate for the procedures under study for the three months previous was 15% (using PI), yet when implemented during the study it fell to 6.0%, perhaps as a result of using a pre-scrub prior to the skin prep (not described in the pre-study period). The rate for CHG was 5.6% for CHG, not different to the PI, but this wasn’t the part that interested me.

The authors described significant risk factors for SSI as the ‘usual suspects’ of age, diabetes mellitus, ASA grade, resected organ, duration of surgery and bilioenteric anastomosis with an additional factor of that I hadn’t seen often (with the exception of laparoscopic) of type of incision.

Age, biliary–pancreatic surgery and incision type (P =0⋅005) were the significant risk factors in multivariate analysis. The ‘inverted L’ incision had an odds ratio of 12.25 (p<0.001) over a midline incision. What influences the choice of incision? I have no idea, I’m not a surgeon, however maybe this is something that should be studied or included in surveillance when there is more than one accepted method of performing a procedure as there is some evidence that smaller incisions for the same procedures are associated with lower complication rates. The thought of wandering into surgery to lean over the shoulder of a surgeon and asking him if he has considered a midline incision as opposed to an inverted L isn’t an appealing one, however I do recall many surgeons advising me on the best way to take a blood culture, that hand hygiene isn’t evidence-based etc etc..


4 thoughts on “SSI – It’s not what you do, but the way that you do it?

  1. Martin

    This is intriguing! Your title is near identical to a letter to the editor in JHI from Dec 2016, entitled:
    ‘Preoperative skin antisepsis – it ain’t what you do but the way that you do it’ –

    In response, myself and a colleague have submitted the below which is currently in press-

    Mode of application is important and should include sufficient volume of antiseptic, contact time and appropriate mechanical action. Good clinical practice in pre-operative skin antisepsis is key.


  2. It is interesting to see that both trial arms used aqueous antiseptics, i.e. aqueous CHG versus aqueous PI, and that there was no significant difference in SSIs between the trial arms. Both CHG and PI (on their own) are well known from the microbiological antiseptic literature — for many decades — to be comparatively weak antiseptics. Unfortunately, much of the clinical trial-based literature on skin antisepsis has been skewed by unfair/inappropriate comparisons of CHG-alcohol (i.e. 2 antiseptics) versus PI alone (one antiseptic). We have written about this aspect here:

    Maiwald M, Chan ESY. Pitfalls in evidence assessment: the case of chlorhexidine and alcohol in skin antisepsis (Leading Article). J. Antimicrob. Chemother. 2014 Aug;69(8):2017-21.

    What we found is that much of the trial-based literature seems to view the alcohol component as a kind of solvent or carrier substance for the CHG, as expressed in the common term “chlorhexidine in alcohol”. This is despite about a century of research that shows it to be a highly effective antiseptic, in fact microbiologically much more effective than either CHG or PI alone. There are only very few appropriately-designed and -conducted clinical trials that compare CHG-containing with PI-containing antiseptics under equivalent conditions, e.g. both combined with alcohol. One prominent trial published in 2016 seemed to show superiority of CHG-ALC over PI-ALC (Tuuli MG et al. N Engl J Med. 2016;374:647-55. PMID: 26844840), but another similar trial published earlier (Ngai IM et al. Obstet Gynecol. 2015; 126: 1251-7. PMID: 26551196) showed no difference, and when the trials are taken together, the significance disappears.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s