Risk factors for intra-abdominal abscess following laparoscopic appendectomy for acute appendicitis: a retrospective cohort study on 2076 patients

Intra-abdominal abscesses (IAA) may develop after laparoscopic appendectomies (LA) for acute appendicitis. The identification of risk factors for postoperative IAA could lead to a decrease in the readmission rate and surgery redoes after LA for acute appendicitis. The present study retrospectively analyzed patients undergone LA for acute appendicitis during the period 2001–2017. Clinical, intraoperative, and postoperative outcomes were described. Comparison between groups was made via univariate and multivariate analyses. The charts of 2076 patients undergone LA were reviewed. Thirty-seven patients (1.8%) developed a postoperative IAA. Male gender (p < 0.05), ASA score ≥ 2 (p < 0.05), a gangrenous or perforated appendicitis (p < 0.0001), abscess or pelvic peritonitis (p < 0.0001), clipping the mesoappendix (p < 0.0001), appendix division by mechanical stapler (p < 0.05), prolonged antibiotic therapy (p < 0.05), and piperacillin/tazocin regimen (p < 0.0001) were significantly more frequent in the group of patients with IAA. In terms of multivariate analysis, only pelvic peritonitis (p = 0.010), perforated appendicitis (p = 0.0002), and clipping the mesoappendix (p = 0.0002) were independent predictive factors for postoperative IAA. Patients with peritonitis or a perforated appendicitis, and those who had their mesoappendix clipped showed a higher likelihood of developing an IAA. At risk patients should be provided with careful follow-up for the early detection and management of this complication.


Background
Post-appendectomy intra-abdominal abscesses (IAA) are estimated to complicate up to 4.2% of acute non-perforated appendicitis cases and 6.7% to 28% of acute perforated appendicitis cases [1]. Laparoscopic appendectomy (LA) has been demonstrated to be safe and effective for acute appendicitis but, compared to open appendectomy (OA), an increased rate of IAAs has been reported [2,3]. IAA is a significant cause of morbidity that increases the length of the hospital stay and needs to be treated by prolonged antibiotic therapy, percutaneous drainage or even reoperation.
The etiology of this complication remains unclear. On the one hand, some authors speculate that the occurrence of postoperative IAA might be influenced by several factors, such as advanced appendicular disease, appendicular perforation, the type of technique used for the removal of the appendix and the lack of an adequate perioperative antibiotics' regimen [4]. On the other hand, there are suggestions that a large-volume intraoperative peritoneal lavage might increase the rate of post-appendectomy IAA [5][6][7][8]. Unfortunately, most of the studies investigating the risk factors for IAA are relatively small-sized, retrospective, and often report conflicting results. The aim of the present study was to discover what factors might contribute to the onset of IAA in patients undergoing LA.

Data collection
We reviewed the data of a prospectively collected database of patients who underwent laparoscopic appendectomy for acute appendicitis during the period 2001-2017 at the Department of Emergency Surgery of Pisa University Hospital. The diagnosis of acute appendicitis was based on clinical presentation, laboratory parameters, and imaging studies. Patients treated with laparotomy and those requiring a conversion to open were excluded. An ultrasound scan was performed for all patients with suspected acute appendicitis. Computed tomography was required in those cases for which a discordance was noted between the clinical presentation and the ultrasound. For all patients, antibiotics were administered about 30 min before the surgical operation or at the moment of the diagnosis in case that surgery would have been delayed after more than six hours, according to the international guidelines on antibiotic prophylaxis in general surgery. The surgical operations were directly performed by experienced surgeons or by a senior resident in co-presence with the consultant. This study is compliant with the STROCSS criteria [9] and is in accordance with the ethical standards of the institutional research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The informed consent for this study was waived because of the retrospective nature.

Surgery
We adopted the policy of performing a laparoscopy for all comers. Laparoscopic surgery was accomplished with the patient supine in the Trendelenburg position, and the table rotated 10°-15° to the left. Pneumoperitoneum was established with CO2 at 12 to 14 mmHg by placing a Veress needle in the umbilical area or by using a Hasson open technique in case of previous surgical operations. A 5-mm trocar was inserted in the periumbilical region for the introduction of a 5-mm-30° laparoscope. Then, another 5-mm trocar was inserted in the suprapubic area, and a 12-mm trocar was inserted in the left iliac fossa for the removal of the appendix. A complete exploration of the whole abdominal cavity, including the identification of signs of acute appendicitis, was firstly performed. The mesoappendix was transected with electrocautery or using clips (HEM-O-LOK® Ligation System, Teleflex, Inc., Morrisville, NC, USA). The base of the appendix was ligated with endoloops and divided or sectioned with a 45-mm laparoscopic articulating stapler in the case of a large or gangrenous/perforated appendicular base. The appendix was then removed with a bag. After the complete evacuation of abdominal collections, a peritoneal lavage with warm saline solution, followed by suctioning, was performed in all patients until a satisfactory peritoneal clearance was achieved. One or two drains were placed in the pelvis and/or in the right paracolic gutter at discretion of the operating surgeon. Drains were removed maximum 48 h after surgery.

Post-operative management
Postoperative antibiotic treatment (amoxicillin/clavulanic acid or ciprofloxacin) was considered "prolonged" in case it was continued up to five days after surgery. In cases of severe and diffuse contamination of the peritoneal cavity, a piperacillin/tazobactam regimen of antibiotic therapy was given.

Follow-up and IAA diagnosis
After discharge, a follow-up program was accorded with patients in case of appearance of fever or abdominal pain, referring to outpatient clinic or emergency department. IAA was suspected in the presence of fever, leukocytosis, and abdominal pain, and an abdominal imaging investigation confirmed the diagnosis (US and CT scan). Treatment consisted of broad-spectrum antibiotics for small-sized abscesses (< 3 cm) and percutaneous drainage for larger abscesses (> 3 cm). Abscesses not amenable to percutaneous drainage or associated with peritonitis were best managed by surgical exploration.

Data analysis and statistics
The following parameters were the target of the analysis: gender, age, American Society of Anesthesiologists (ASA) score, BMI, time interval between symptoms onset and surgery, intraoperative findings (abscess, gangrenous appendicitis, perforated appendix, pelvic peritonitis), method of management of the mesoappendix, type of technique adopted for the division of the inflamed appendix, use of peritoneal drains, postoperative use of antibiotics, and postoperative use of piperacillin/tazobactam. Pelvic peritonitis was defined as the presence of suppurative fluid with fibrin and local contamination in the pelvis at the moment of the operation. The Lilliefors test was used to assess the normality of the distributions. A Mann-Whitney test was carried out to analyze quantitative variables. In addition, the Chisquared test was performed to study categorical variables. Furthermore, multivariate logistic regression was used, and the Hosmer-Lemeshow test was carried out to evaluate the goodness of fit of the multivariate model. A p value < 0.05 was considered statistically significant. The statistical analysis was performed using the R software package (The R Foundation for Statistical Computing, Vienna, Austria), version 3.5.0. In 1,438 cases (69.3%), the mesoappendix was coagulated and dissected by bipolar current forceps. In the remaining 638 patients (30.7%), the mesoappendix vessel was clipped and divided via scissors.

Results
Endostapling was the preferred method of dealing with the basis of the inflamed appendix in 467 (22.5%) patients, whereas division of the basis of the appendix between endoloop ligatures was the most used method. Peritoneal drains were placed in 1533 cases (73.8%).

Discussion
The present study shows that pelvic peritonitis, perforated appendicitis, and clipping the mesoappendix represent independent predictive factors for postoperative IAA following laparoscopic appendectomy. As a matter of facts, these results not only confirm data reported in the literature but also highlight the use of clips for the mesoappendix as a further potential risk factor for postoperative IAAs. So far, they should be avoided for the closure of the appendicular artery.
The severity of the appendix inflammation is believed to be the main risk factor associated with IAA. The IAA rate after laparoscopic appendectomy reported in non-perforated appendicitis is 1-4%, but the incidence of postoperative IAA increases by up to 26% in patients with gangrenous perforated appendicitis [10][11][12][13]. We found perforated  (Table 2). According to Schlottmann et al. the higher is the grade of intraperitoneal bacterial contamination, the higher is the risk of postoperative IAA [14], and few doubts exist about the fact that perforation increases the grade of contamination of the peritoneum surrounding the appendix. Indeed, this study strongly supports the hypothesis that a perforated appendix should be regarded as a predictive factor for postoperative IAA (OR 6.4; p = 0.0002) ( Table 3). In addition, pelvic peritonitis was significantly more frequent in patients who developed postoperative IAAs (17.5% vs. 43.2%; p < 0.0001) ( Table 2) and was recognized as a predictive factor for postoperative IAA via multivariate analysis (OR 3; p = 0.010) ( Table 3).
According to the WISS study, acute appendicitis is still the most frequent cause of intra-abdominal sepsis [15]. A delayed diagnosis, mostly occurring in patients who are unreliable or have an atypical clinical onset, may lead to severe, life-threatening complications such as gangrene, perforation, appendiceal mass, and peritonitis [16]. Overall, 18% of patients enrolled had a peritonitis localized in the pelvis or in the paracolic gutter. This result is in line with the rate reported in the literature (17.3%) [13].
The clipping of the mesoappendix is associated with a higher incidence of postoperative IAAs. In this study, the use of clips was left to the operating surgeon's discretion, but such use was always limited to the management of the mesoappendix. Compared to bipolar coagulation, we recorded a significantly higher occurrence of IAA when the mesoappendix was secured with polymeric clips (29.7% vs. 62.1%; p < 0.0001). Polymeric clips have been increasingly used for the closure of the appendiceal stump [17,18], but the correlation between the use of polymeric clips and the increased rate of IAA, if it exists, remains unexplained. Some studies have indeed investigated the possible correlation between the method of mesoappendix dissection and the onset of IAA, with no significant conclusion. Wright et al., in a review of 565 patients undergoing laparoscopic appendectomy for uncomplicated appendicitis, analyzed 149 patients (26%) who had a transection of the mesoappendix and appendix with a single staple line, 259 (46%) who had multiple staple lines, and 157 (28%) who had their mesoappendix dissected with the aid of ultrasonic shears and their appendix divided by a single staple line. They found that the incidence of complications was low, with hematoma/abscess, transfusion, and reoperation occurring in only 1.4%, 0.4%, and 0.9% of cases, respectively. There were no significant differences among groups [19]. In a retrospective study of about 1178 patients who received laparoscopic appendectomy, Lee et al. compared 460 (39%) patients who had their mesoappendix managed with endoclips, 346 (29%) who had their mesoappendix managed by an ultrasonic device, and 346 (32%) who had their mesoappendix managed by monopolar electrocautery. They found no significant differences in terms of postoperative complications, including wound infection, abscess, paralytic ileus, and hemorrhage [20]. We cannot give a reasonable explanation regarding the correlation between the use of polymeric clips for the mesoappendix and increased rate of IAA reported in the present study. We could speculate that a non-absorbable foreign material in a contaminated field might be a "culture medium" for bacteria, supporting the onset of postoperative IAAs. However, Soll et al. in their retrospective and comparative study on 813 patients receiving laparoscopic appendectomy either with endoclips or endoloop for uncomplicated acute appendicitis found that postoperative intra-abdominal abscesses occurred significantly more frequent in the endoloop group compared with the endoclips group (4 vs. 1%; p = 0.012) [4]. It must be said that our study considered complicated acute appendicitis, which represented a 21.6% of the overall cohort. So far, our experience demonstrates that the use of non-absorbable clips should be avoided, a fortiori in complicated acute appendicitis. However, in order to give a definitive conclusion, a focused randomized study is recommended. It emerges from this study that a prolonged administration of postoperative antibiotics had no significant effect on the occurrence of postoperative IAA. From this point of view, we confirm the data reported by Kimbrell et al. on a cohort of 52 young patients undergoing appendectomies for complicated appendicitis. They found that the occurrence of an abdominal abscess was not significantly lower if the postoperative antibiotics regimen was prolonged for more than 24 h compared to when antibiotics were withdrawn within 24 h after surgery. Thus, they concluded that postoperative antibiotics might not provide an appreciable clinical benefit in terms of preventing intra-abdominal abscesses [21]. The postoperative use of antibiotics for complicated appendicitis has proven to be beneficial [22], but the optimal duration of treatment has not yet been established. In a recent retrospective study on 6,412 patients with complicated acute appendicitis, Anderson et al. suggested that not all patients with complicated appendicitis should be discharged with antibiotics after surgery. Accordingly, only patients exhibiting SSI before discharge or those whose clinical progress requires a more extended in-hospital stay might benefit from extra-time antibiotic treatment. [23].
The present study suffers several limitations. Primarily, this is a retrospective study. Secondly, this series represents a vastly complex, heterogeneous patient population scattered over a considerable period. Thirdly, the lack of including other risks factor can affect the statistics of the multivariate analysis. Furthermore, the number of patients presenting a postoperative IAA constitutes a small cohort of cases. However, since the study was carried out at a tertiary referral center, the high volume of patients undergoing laparoscopy for acute appendicitis makes the statistical analysis credible and valuable. Lastly, we acknowledge that the patients were not routinely followed once symptoms resolved after the surgical intervention, making it possible that patients could have been lost during follow up.

Conclusions
Patients with perforated appendicitis, pelvic peritonitis and a clipped mesoappendix have a higher chance of developing postoperative IAA. The type of technique adopted for the division of the appendix (endostapler vs. endoloop) does not seem to be a risk factor for the development of postoperative IAA. The lack of postoperative antibiotic treatment was not identified as an independent risk factor for abscess formation.