Open Versus Laparoscopic Treatment for Pan-Peritonitis Secondary to Perforated Appendicitis in Children: A Prospective Analysis

Introduction

Since the first report on laparoscopic appendectomy in 1983, ¹ there have been a large number of studies comparing laparoscopic with open appendectomy in patients with appendicitis, both complicated and uncomplicated.^2–4 However, these reports also include localized peritonitis, appendiceal mass, and pan-peritonitis (PP); so to the best of our knowledge, there is no paper specifically referring to PP in children.

This study aimed at assessing the feasibility and safety of laparoscopic treatment (LT) for the treatment of PP due to perforated appendicitis (PA) in children and at comparing it with open treatment (OT).

Materials and Methods

With institutional review board approval (Juntendo University School of Medicine Ethics Committee: No. 17-016), we prospectively collected data and reviewed all patients who underwent appendectomy for PA with PP at Juntendo University Hospital from 2004 to 2008.

For this study, we defined PP as generalized preoperative guarding/rigidity of the entire abdomen with presence of a purulent peritoneal collection throughout the abdominal cavity. In LT, presence of a purulent peritoneal collection in the entire intraabdominal cavity was easily confirmed under direct vision. In OT, a tube was inserted into a right subphrenic space, a left subphrenic space, and the pouch of Douglas, respectively, through the wound, to confirm presence of a purulent intraabdominal collection. Cases that did not meet these criteria were excluded; in particular, PA, localized peritonitis, and appendiceal mass. Patients with PP were divided into two groups according to treatment, laparoscopic (LT), and open (OT).

All surgeries were performed under the direct supervision of three board-certified pediatric surgeons who each chose their approach (OT or LT) according to their preferences. The OT group included open surgical cases as well as cases converted from LT. OT was performed under general anesthesia through a transverse right lower quadrant skin crease incision according to conventional techniques described in standard textbooks on general surgery. LT was performed using a 3-trocar technique with a combination of 5 mm and 10 mm cannulas. Ultrasonographic shears (ETHICON. Inc) were used for tissue dissection, and Endoloop (ETHICON. Inc) was used for appendix stump closure. Specimens were retrieved inside disposable specimen retrieval bags through the 10 mm site. Saline irrigation was ceased when fluid collected from the pouch of Douglas and both subphrenic spaces was clear. A J-VAC drainage tube was inserted through the right lower quadrant trocar site and placed within the pouch of Douglas. In all patients, initial doses of piperacillin and tazobactam were intravenously given within 30 minutes of the induction of anesthesia, then continued twice daily at a reduced dosage. During the operation, when the diagnosis turned into PA, clindamycin phosphate, 25 mg/kg per day, was started to be administered thrice daily for 2 days. Intravenous antibiotics were ceased once patients were afebrile for >24 hours and white blood cell count (WBC) normalized. Once appropriate, oral metronidazole and cefotiam were commenced and continued until C-reactive protein (CRP) normalized. For analgesia, acetaminophen suppositories were used up to a maximum of thrice per day. Drainage tubes were removed after any discharge became clear. Any patient with clinical features suggestive of a postoperative intraabdominal collection had abdominal ultrasonography and was commenced on antibiotics or prepared for surgical drainage as required.

We collected data on mean operative time; the incidence of operative complications; duration of antibiotic and analgesic usage; time required for WBC, CRP levels, and fever to normalize; time taken for oral feeding to commence; and duration of hospitalization. Data were analyzed using standard statistical methods. Demographic data were compared using the Student's t-test. The chi-squared test or Fisher's exact test were used for analyzing the incidence of complications. For all statistics, 0.05 was used to determine significance.

Results

Generalized guarding/rigidity was observed in the entire abdomen in all patients in both groups. Purulent collection extending throughout the intraabdominal cavity was confirmed in all patients in both LT and OT groups. Thus, all 23 patients fulfilled our criteria for “pan-peritonitis.” Treatment was OT in 11 and LT in 12. Only one patient required conversion from LT to OT, and this patient was included in the OT group.

Mean age at surgery (OT: 7.7 years, LT: 7.6; P = 0.54) and sex ratios (OT: 3F, LT: 7F; P = 0.14) were comparable between the two groups. Preoperative mean WBC (OT: 14,600, LT: 14,900; P = 0.89) and CRP (OT: 16.5, LT: 17.1; P = 0.83) were also comparable. Appendectomy was successfully performed in all 23 patients without intraoperative complications. Mean operative time was 107 minutes for OT and 119 minutes for LT (P = 0.56). Mean volume of saline used for peritoneal lavage was 2950 mL in OT and 2750 mL in LT (P = 0.45). Average duration of analgesic usage was significantly shorter in LT (0.5 days) than in OT (1.7 days), P = 0.006. Postoperatively, the incidence of wound infection was significantly lower in LT (0/12) than in OT (4/11: 36.4%) (P = 0.037). Adhesive bowel obstruction occurred in 1 LT and 3 OT cases (P = 0.26), but none required re-operation. Two cases of intraabdominal collection developed in each group during hospitalization (OT: 2/11 or 18.2%, LT: 2/12 or 16.7%; P = 0.67), but all successfully responded to antibiotics without surgical drainage being required. All collections had disappeared by the time of discharge from hospital, and there have been no recurrences (Table 1).

No significant differences were identified between the groups for time taken for patients to become afebrile (OT: 5.2, LT: 4.8 days; P = 0.75), for WBC and CRP to normalize (OT: 7.7, LT: 5.7 days; P = 0.39; OT: 11.2, LT: 12.2 days; P = 0.60 respectively), for intravenous antibiotics to be ceased (OT: 8.9, LT: 7.6 days; P = 0.44), and for oral feeding to be commenced (OT: 5.4, LT: 4.8 days; P = 0.65). Mean hospitalization was significantly shorter in LT (8.7) than in OT (13.3 days) (P = 0.03) (Table 1).

Discussion

Complicated appendicitis is generally defined as gangrenous, PA, and appendiceal abscess ⁵ and is associated with both local and diffuse peritonitis. Adults are often able to give an accurate history of the prodrome of symptoms experienced, whereas children are less reliable historians given their age and younger children, in particular, can often present with advanced appendicitis involving perforation, abscess formation, and possible PP. ⁶ To date, PP has not been specifically defined in children; and one point of this study is that we define PP both clinically and intraoperatively to provide a standard to base further research on. Surprisingly, data comparing LT with OT for the specific treatment of PP cannot be found in English literature; and to the best of our knowledge, this is the first report comparing LT with OT for the treatment of PP secondary to appendicitis.

There is always a high risk for postoperative intraabdominal collection secondary to complicated appendicitis, and there are reports on the higher incidence of collections after LT in both adults and children in the literature.^3,7 One traditional explanation for the high rate of postoperative intraabdominal collections associated with LT is that infected matter can readily spread throughout the abdominal cavity during pneumoperitoneum.^8,9 However, in patients with PP, infected contents have already spread throughout the abdomen. Thus, we believe, even if pneumoperitoneum could have made infected contents spread further, it would not affect the incidence of intraabdominal collection. Further, “pan-peritonitis” could have been misdiagnosed or over diagnosed in OT cases if the entire abdomen was not examined for infected material. In the present study, we first focused only on cases with PP who fulfilled our criteria for “pan-peritonitis” (generalized preoperative guarding/rigidity of the entire abdomen together with presence of a purulent peritoneal collection in the entire intraabdominal cavity) and found the incidence of postoperative intraabdominal collections to be similar in both LT and OT (LT: 2/12 (16.7%); OT: 2/11 (18.2%); P = 0.67).

We cannot fully exclude selection bias in the present study, as the board-certified pediatric surgeon on duty chose which procedure to perform according to personal preference rather than preoperative clinical signs and symptoms, operative findings, or the surgeon's technical skills; and LT were performed by trainee surgeons with learning curves, which could affect outcome, especially complication rates and operating times compared with OT. We feel that both could be the cause of bias.

From this study, we can conclude that LT would appear to be superior for the treatment of PP secondary to appendicitis in children and would even seriously consider it as the procedure of choice, because the incidence of postoperative wound infection is significantly lower; the duration of analgesic usage is significantly shorter; and mean hospitalization is significantly shorter without any significant difference in operating time or time taken to postoperatively recover.