Disposable Specimen Retrieval Bags
SSIs are responsible for increased length of hospitalization and health care costs. Diabetes, malnutrition, male gender, chronic anemia, obesity, drug abuse, smoking-related diseases and previous Staphylococcus aureus infection were reported in several studies as patient-related risk factors for SSIs after cholecystectomy [19].
In order to avoid surgical site contamination from bile and stone spillage, surgeons pay attention not to open the gallbladder during its dissection from the liver bed and retrieval from the abdominal cavity. Bile in the gallbladder or bile ducts in the absence of gallstones or biliary tract disease is normally sterile. In the presence of gallstones, the prevalence of bacteria increases: the percentage of positive gallbladder bile cultures among patients with symptomatic gallstones and chronic cholecystitis ranges from 11 to 30%. Positive bile cultures are significantly more common in elderly (> 60 years) patients with symptomatic gallstones than in younger patients (45% versus 16%) [20].
From the early days of laparoscopic surgery, many manufacturers have developed different types of retrieval bags, whose use has become popular among minimal-invasive surgeons in laparoscopic appendectomy, cholecystectomy, bowel resection and annexectomy.
According to the “Guidelines for the Clinical Application of Laparoscopic Biliary Tract Surgery” of the Society of American Gastrointestinal and Endoscopic Surgeons (SAGES), the use of an endoscopic bag is left at the discretion of the operating surgeon [2]. In order to evaluate the popularity of retrieval bags in elective laparoscopic cholecystectomy, we sent a questionnaire via email to 150 consultant general surgeons working in 5 different countries (Switzerland, United Kingdom, Germany, Italy and Austria), asking whether and why they would routinely use an extraction bag during this procedure. We received an answer from 61 surgeons (40.7%). With one exception, all those interviewed (98%) confirmed the routine use of this device. Twenty-five (41%) surgeons justified it by the opinion that it contributed to prevent a wound infection at the trocar site. The other surgeons answered with “comfort” or “no reason”. Clearly, it should be taken into account that the retrieval bag gives the possibility to remove spilt stones and to aid extraction in patients affected by morbid obesity and when the gallbladder has been opened during the dissection [3].
The incidental gallbladder cancer is considered a rare eventuality happening in 0.2–3.3% of elective cholecystectomies [21]. In fact, 50% of cases are identified preoperatively and 29% intraoperatively. Only 21% of cases are recognized at the definitive histologic examination [22]. In such cases, the routine use of a retrieval bag to prevent post-site metastasis was claimed to be mandatory [3], even though several factors should be considered. In addition to the rarity of the disease, it is described that almost 90% of patients do not develop a port metastasis. In case of port-metastasis, this is localized in 53% of cases at the site of the extraction trocar, being the risk in non-extraction trocars 47% [22].According to some authors, a port-site metastasis reflects more likely the tumor wide-spreading and the aggressive biology rather than a direct contact of the gallbladder with abdominal wall [22]. Moreover, data of the central register of “incidental gallbladder carcinoma” of the German Society of Surgery suggested that the usage of retrieval bags was not associated with a decreased risk of seeding if gallbladder perforation did not occur intraoperatively [3]. Finally, in case port metastasis occur, the port-sites excision could be a valuable therapeutic option, as this surgical step is nowadays still a matter of debate [22, 23]. To date, these elements made the prophylactic use of a retrieval bag to avoid neoplastic cells seeding in all elective cholecystectomies widely debatable.
In case of acute cholecystitis, many authors recommend the extraction of the gallbladder in a retrieval bag as port site infections are frequently associated with spillage of infected bile, stones or pus [4,5,6]. Even if the use of a retrieval bag in the above-mentioned situations seems justified or reasonable, there is no strong evidence to support the use of a retrieval bag in elective laparoscopic cholecystectomy. All wound infections in the study of Harling et al. [16] were associated with skin commensals. In the study of Comajuncosas et al. [17], in all cases, except one (E. coli), organisms isolated from the wound sites of those patients that developed postoperative infections were skin commensals (Corynebacterineae, coagulase-negative Staphylococcus spp., Streptococcus pyogenes). In previous studies, similar results were obtained [24,25,26,27,28,29,30]. The absence of correlation between typical (gram negative) bile and wound infection organisms suggest that most port site infections do not depend on a direct contact of the gallbladder with the wound.
In our meta-analysis a SSI was documented in 14 of 334 (4,2%) patients operated using a retrieval bag versus 16 of 271 (5,9%) patients operated without the use of a retrieval bag. The statistical analysis revealed a RR of 0.82 (0.41–1.63 95% CI) and no statistically significant reduction in SSI when the extraction of the gallbladder from the abdominal cavity was performed with a retrieval bag.
The latter, in addition, is not risk-free. In the largest study of our meta-analysis [18], an enlargement of the port site incision was required in 9,7% (36/373) of patients. At 1 year follow-up, there was no recorded cases of port site hernia in the group without the use of a retrieval bag and two (0,9%) cases of port site hernia in the retrieval bag group. In addition, there is an anecdotal risk of abdominal organs damage during bag insertion and retrieval [7,8,9].
Retrieval bags are not cheap, ranging from € 25 to € 120, and their use must be questioned in a time of rising economic pressure on the health care providers. Interestingly, there are plenty of reports in the medical literature about “cost-effective, self-made” specimen extraction bags [31,32,33,34].
This study has several limitations. A small number of trials were eligible for the meta-analysis, resulting in a low number (605 altogether) of patients. The study of Majid et al. [18] albeit prospective, is non-randomized. Another limitation is the mean prevalence of SSIs in the included studies. It is stated that an acceptable SSIs rate ranges between 1.6 and 3.2% [35, 36], defined according to Centers for Disease Control [37] as purulent discharge from the surgical site, with or without positive culture or signs of inflammation. However, in our analysis it was 7.9% in Harling et al. [16], 9.6% in Comajuncosas et al. [17] and 2.4% in Majid et al. [18] studies. The high incidence of SSI could be explained by different definitions adopted. Nevertheless, regardless the absolute number of infections, the primary endpoint of our study was the evaluation of postoperative SSI rate, which resulted equal between groups in all studies. Considering the lack of significant difference related between groups, the cause of infection is improbable to be related to the direct contact of the bile and gallbladder with the wound.
Moreover, the included studies have different sample sizes, being the majority of patients in the study of Majid et al. [18]. In order to evaluate this possible bias, we assessed the sensitivity analysis (ranging from 0.72 to 0.96) and it was not statistically significant. Harbord test did not reveal the occurrence of small-study effect and the funnel-plot showed no noteworthy pattern, both indicating this bias unlikely.
Another limitation is the heterogeneity in the antibiotic prophylaxis regimens. In the study of Harling et al. [16], the patients were randomized to receive a single dose of Cefuroxime (750 mg, i.v.) or to have the gallbladder removed from the abdomen with a retrieval bag. Comajuncosas et al. [17] used no antimicrobial prophylaxis, Majid et al. [18] gave a single dose of 1.2 g Co-Amoxiclav at the time of induction (1,5 g Cefuroxime in case of penicillin allergy). However, the use of retrieval bag did not change the rate of SSI, nor in presence or absence of antimicrobial prophylaxis, neither compared with antibiotics administration.
We applied the GRADE approach [12] in order to evaluate the quality and, taking into account limitations mentioned above, the quality of evidence of our paper ranked from “moderate” to “low”.