Literature Review of Criteria for Defining Recipient-Site Infection after Oral Oncologic Surgery with Simultaneous Reconstruction

Patients and Methods

A Medline search was performed via PUBMED using the following combination of key terms that were tagged in the title, abstract, or both: “surgical site infection—head neck,” “surgical site infection—oral cancer,” “antibiotic prophylaxis—head neck,” and “surgical site infection—oral carcinoma.” Search results were filtered for the literature between 2005 and 2017.

The articles used had to meet the following inclusion criteria: human studies; studies including cases in which oral cancer resection with primary reconstruction was performed; explicit description of the criteria applied for the definition of SSI; explicit description of the number of reconstructed cases; and studies published in the English language. The following exclusion criteria were applied: animal studies; no mention of the criteria for defining SSI; no mention of disease location (because the percentage of patients with oral cancer was unknown); studies including only operations on the larynx, thyroid, skull base, and major salivary gland; studies in which the surgical sites were clean in most cases; studies in which the number of reconstruction cases was small (<30% of all cases); and studies described by the same authors of which the purpose was different but the enrolled population was similar. We reviewed all clinical trials and included both prospective and retrospective studies. Review articles were excluded.

The titles of the articles were screened for relevancy according to the above-mentioned criteria. If these criteria were met, the abstracts were screened according to the inclusion criteria. If the abstracts did not provide sufficient information, the full-text article was downloaded and reviewed. A few articles that might have met the inclusion criteria, but could not be downloaded as the full-text version, were excluded in this review. If information on recipient- or donor-site infection was not mentioned, distinction between recipient- and donor-site infections is not shown in the tables.

Results

The search resulted in 1,089 hits on PUBMED. After reading the downloaded full texts, 24 articles satisfied the inclusion criteria. The numbers of articles that applied the criteria of Johnson, the CDC, and others were nine, nine, and six, respectively.

Table 1 shows a summary of studies that applied the Johnson criteria. In all of the studies, only grades 4 (purulent drainage) and 5 (orocutaneous fistula) were defined as SSI. In some studies, there was no mention of whether flap failure was included in SSI, whereas three studies clearly specified that they regarded spontaneous or incisional purulent drainage caused by flap failure as SSI [7–9]. In one study, SSI was said to be present only when the microbial cultures were positive [10]. In another study, incisional dehiscence or ischemic necrosis of incision edges requiring local care was not considered an infection [7]. Six of nine articles that applied the Johnson criteria were from Asian countries. Some reports were from the same department, including four from Taiwan [11–14] and two from France [8,9]. The other three reports from Asian countries were from Japan [7,10,15]. Although information on recipient- or donor-site infection was not mentioned in some studies, the highest incidence of post-operative SSI was 45% [9], and the lowest incidence of recipient-site infections was 13% [15].

Table 2 shows a summary of the studies that applied the CDC criteria. In these studies, the diagnosis of SSI was based on clinical findings, such as purulent drainage, an incision that either dehisced spontaneously or was opened by the surgeon, and abscess formation, which are similar to the Johnson criteria. Although some papers did not mention whether superficial infection was regarded as SSI, one report clearly specified that only deep incisional infections based on the CDC criteria were considered SSI [16]. The application criteria in the study by Pool et al. [17] were similar to the diagnostic criteria for deep incisional SSI. In only one study, by Wagner et al. [18], were SSIs classified as superficial or deep. Notably, in three studies, orocutaneous fistulas that did not meet the CDC criteria were not categorized as an SSI [3,17,19]. In contrast, one study that applied the CDC criteria said that a fistula, which was defined as the presence of orocutaneous or pharyngocutaneous leakage, was considered an SSI, regardless of the origin [20]. The highest incidence of post-operative SSI was 38.3% [21], and the lowest incidence of recipient-site infection was 11.1% [22].

Table 3 shows a summary of studies that applied other criteria. Two studies used both the Johnson and the CDC criteria. In these studies, SSIs were classified as incisional (deep or superficial) or space SSIs, and fistula was included in SSI [23,24]. A prospective study by Mücke et al. [25] defined SSI only as purulent discharge from the incision. Frederick et al. [26] restricted SSI to an infection requiring surgical intervention. Avery et al. [27] applied the Szilagyi classification for vascular surgery. The highest incidence of post-operative SSI was 62.1% [28], and the lowest was 4%.

Discussion

There have been numerous reports on SSI after head and neck cancer surgery. However, the results, such as the incidence of SSI presented by different authors, appear to lack uniformity and are conflicting [2]. In addition to the heterogeneity of the population included in each article, we focused on the lack of uniformity of the criteria for defining SSI. Therefore, this literature review was conducted to determine the differences in criteria. The most popular criteria for SSI are those from the CDC. The Johnson criteria also are often applied. Our review showed that, in studies that applied the Johnson criteria, only purulent drainage and orocutaneous fistula (regardless of etiology) were categorized as SSI. In contrast, two studies that applied the CDC criteria did not regard orocutaneous fistula that did not meet the CDC criteria as SSI [17,19]. Moreover, in two studies that applied the CDC criteria, only deep incisional infections were regarded as SSI [16,17].

Head and neck fistula tends follow extensive resection for cancer, with a reported incidence ranging from 3%–65% [29]. In a recent study in which almost all of the cases were oral lesions (e.g., cancer, osteoradionecrosis, or ameloblastoma), the incidence of orocutaneous fistula after reconstructive head and neck surgery was 10% [30]. Additionally, the incidence of infection (no mention of the definition criteria), dehiscence, and flap failure was 5%, 7%, and 1%, respectively [30]. In their study, a significant association was identified only between fistula formation and previous chemoradiotherapy. In a study by Guidera et al. [31], the incidence of orocutaneous fistula after oral cavity cancer resection and free flap reconstruction was 7% (4/54), of which three cases (75%) were associated with partial or total flap failure. There was no mention of the association between infection and orocutaneous fistula. Importantly, Johnson et al. [6] described “the cart before the horse” dilemma, where incision separation caused by tissue ischemia (e.g., suture-line ischemia) or other technical difficulties precedes infection. In our experience, almost all formation of an orocutaneous fistula after oral cancer resection with simultaneous flap reconstruction is followed by ischemic complications of a transferred flap. Orocutaneous fistula results in infection. In contrast, recipient-site infection causes flap failure in only 1% of free-flap reconstructions for intra-oral cancer [32]. Taken together, these findings suggest that inclusion of orocutaneous fistula without considering the etiology of SSI is inadequate for evaluating recipient-site infection in oral oncologic surgery with flap reconstruction. Therefore, orocutaneous fistula should be categorized as post-operative SSI with consideration of the etiology. This is because such fistulas generally occur because of flap complications or ischemia of the incisional edges in patients who undergo oral oncologic surgery with simultaneous reconstruction. Only orocutaneous fistula caused by apparent infection such as abscess formation should be categorized as post-operative SSI.

The subcutaneous tissue and fascia of the neck are thinner than those of the limbs and trunk so that recipient-site infections after cervical neck dissection are difficult to classify into superficial and deep. Candau-Alvarez et al. [16] focused only on deep incisional infection because infection after cervical neck dissection does not expand into other organs or spaces. Although a few studies distinguished superficial and deep incisional infections, they included donor-site infections. In donor-site infection, distinguishing the superficial and deep layers can be easier than in the cervical neck. Another element making the diagnosis of post-operative superficial infection in recipient sites difficult is flap transfer. Flaps larger than the defect should be transferred because the volume of transferred flaps is sequentially reduced [33]. A bulky flap occasionally causes ischemia of the suture line and spontaneous dehiscence, followed by serous discharge. This wound dehiscence generally heals without antibiotic therapy but tends to be misdiagnosed as superficial infection in the recipient site by inexperienced physicians (e.g., early-year residents). Generally, the causes of recipient-site infection after oral cancer surgery requiring neck dissection are the formation of a hematoma, salivary leakage, and dead space resulting from a complex defect (i.e., problematic recipient-site infection after oral cancer surgery with reconstruction always occurs under the subcutaneous layer). With regard to recipient-site infection, differentiation of superficial and deep incisional infections appears to be unnecessary because of the difficulty of the clinical diagnosis in some circumstances.

Our literature review has some major limitations. This study aimed to complete the important studies with the recently publicized literature, with no claim to completeness on the whole issue. The study populations in articles that were included in this review were heterogeneous. Types of diseases that were included were not only oral malignancy, but also benign lesions or osteoradionecrosis. Moreover, hypopharyngeal or laryngeal carcinomas were included in some articles that were considered in this review. We did not conduct a meta-analysis because of the heterogeneity of the literature results. Regardless of these limitations, our review highlights an overlooked, but critical issue. We can hope that uniform, novel criteria for defining recipient-site infection in oral oncologic surgery with reconstruction will be established in the future based on the proposal described in this paper.

Conclusions

This literature review reveals the lack of uniformity of the criteria for defining recipient-site infection after oral oncologic surgery with simultaneous reconstruction. This non-uniformity of criteria is attributable mainly to differences in interpretation of orocutaneous fistula and superficial incisional infection. Categorization of orocutaneous fistula as SSI, regardless of etiology, does not contribute to correct assessment of post-operative SSI. Moreover, for evaluating recipient-site infection in oral cancer surgery, differentiation of superficial and deep incisional infections appears to be unnecessary.

Author Disclosure Statement

The authors have no conflicts of interest related to this manuscript.