In our single-center for elective adult orthopedic surgery, we plausibly estimate one-third of all formal acute SSIs being acquired (originating) during surgery. When excluding with (diabetic) foot surgery, this postoperative proportion becomes one-sixth and the intraoperative part 83%. We thus confirm the expert opinion stating that the majority of SSI are presumably acquired intraoperatively; but maybe to a lesser extent than we previously anticipated.
Existing literature concentrates on intraoperative risks and perioperative interventions. Even recent publications and preventive trials prefer to target pre- and risks. Clinically, the antibiotic prophylaxis [11,23], the self-reported compliance of surgeons (only pre- and intraoperatively) [24], the increasing number of surgical skill labs, or the preoperative decolonization [10,25,26], all concentrate on the peri-, or preoperative period. Indeed, prospective trials on decolonization do not extend beyond the postoperative phase, although postoperative new wound problems would occur frequently in orthopedic surgery [27]. Similarly, bundled interventions often neglect the postoperative period [28]. Regarding the postoperative period, guidelines [18-20] and expert opinions are often limited to improve hand hygiene [9,29] and to enhance “asepsis” during dressing change [15]. Exceptionally, some research groups report a protective benefit of negative-pressure technology for preventive purposes in selected patients [14]. International guidelines state by resigning that "There is a lack of high-quality studies comparing various strategies of postoperative wound management…. This is an area for further research" [19].
Pre- or perioperative prevention might not be enough, especially when “intraoperative” measures [1] have been maxed out, and the clinicians would only expect a slight improvement of prevention just by adding supplementary efforts into the “intraoperative“ axis. We think that our relatively high proportion of postoperative SSIs needs further attention. However, and theoretically, the various aspects of postoperative prevention are more resource--consuming than only focusing on perioperative measures. They will involve a higher number of HCWs and medical disciplines, and a longer intervention compared to the relatively short duration of the surgical intervention. Moreover, postoperative interventions take place in environments that cannot compete with the “aseptic” operation theatre in terms of infection control.
The literature on postoperative orthopedic SSIs is sparse. Originally and according to early expert opinion, this literature mostly consisted of hematogenous PJIs representing up to one-quarter of all arthroplasty SSIs in selected studies [30]. Hematogenous hip PJI might occur earlier than hematogenous knee PJIs [31]. Other (smaller) implants, such as plates or nails, are quasi exempt from a hematogenous seeding from a remote infection source. In recent literature, the risk of hematogenous seeding is much lower than the risk of “intraoperative” PJIs becoming bacteremic. A retrospective cohort identified that only 1 of 79 severe remote infections would become bacteremic and seed to total joint arthroplasties [31]. A Swedish study assessed the incidence for hematogenous PJI in case of Gram-positive bacteremia. This risk was only 6% [32]. Sometimes the origin of hematogenous PJIS may remain occult. If the origin cannot be elucidated through history, a blind (radiological) and laboratory search is ineffective and costly [33].
The majority of the postoperative SSI acquisitions occur locally and are due to two major groups of postsurgical problems: wound breakdowns (ischemia [6], trauma, hematoma, dehiscence, seroma [27]) and patient’s malcompliance in its various forms (e.g. unrest against medical order [34], smoking [35], or malnutrition [36,37]). A wound breakdown is frequent. According to a prospective observation during one year in Geneva, only 40% of freshly operated orthopedic wounds yielded no clinical problems [27]. In that study, among 1,073 adult orthopedic patients, 630 operated episodes (59%) showed relevant postoperative wound complications, leading to a significantly longer hospital stay compared to patients without complications. The most frequent wound complications were serous discharge with dehiscence (41%) and hematoma (35%) [27]. Wound dehiscence due to malcompliance remains a major problem. To cite an own example, a study randomizing operation techniques for septic bursitis, identified the lack of compliance (together with psychiatric conditions) as an important reason for failure (and infection recurrence) [34]. Other breakdowns occur after postsurgical debridement or because of wound necrosis in ischemic patients such as in diabetic foot syndromes. Finally, there are PIN-tract [1] infections in orthopedic surgery of the lower extremities, especially on the toes. PINs and wires cross the skin and are kept in place during weeks. In these situations, the reason for SSI could be a wrong surgical indication for elective surgery, or, most likely, unintended shortcomings in hand hygiene [29] during professional (or less experienced) wound care [15].
Our observation study has strengths and limitations. The main strength is a prospective work-up of each attributed SSI case by judging upon playabilities of the individual patient's histories. The main limitations are multiple: i) The subjective evolution of the origins. ii) Mild superficial and late (low-grade) SSIs, which were not revised, might have gone unnoticed. iii) This is a pilot observation. We do not report intervention. Infection Control encompasses many other HAI groups and antibiotic resistances [38], which we could not evaluate in this study with very few multidrug-resistant pathogens. Indeed, operated patients reveal the same proportion of other HAIs as non-operated patients, with 5% SSIs in addition [3]. v) Lastly, our observation only concerns adult orthopedic surgery. In other disciplines with less superficial wounds, e.g. abdominal laparoscopy, the epidemiology of postoperative SSIs might be completely different.