Surgical Management of Gallstone Pancreatitis in Children

Introduction

The incidence of pancreatitis in children has significantly increased over the last decade. ¹ Although the most common causes in this patient population are trauma and metabolic diseases, ¹ biliary pancreatitis is reported in as many as 5.4–28.3%.^1,2 This is in stark contrast to adults, in whom gallstones are implicated as the cause of pancreatitis in over 60% of nonalcoholic patients. ³ With the growing incidence of obesity in children, the incidence of gallstone pancreatitis may increase as well. Management of gallstone pancreatitis in children has been discussed in small case series.^4,5

Because of recurrence rates of 25–63% for gallstone pancreatitis in adults,^6–8 cholecystectomy during the initial hospitalization is recommended.^9,10 However, this well-accepted approach is not practiced in the majority of cases.^11–13 Optimal surgical management of gallstone pancreatitis in children is not well defined but could be assumed to need to mimic practice in adults given the similar pathophysiology. We sought to review our experience with the timing of cholecystectomy in children with gallstone pancreatitis and to determine the effects on varied practice patterns, if present.

Subjects and Methods

After obtaining Institutional Board Review approval, we performed a retrospective chart review of all patients undergoing cholecystectomy for gallstone pancreatitis between July 2000 and July 2011. Patients who underwent cholecystectomy during the initial hospitalization (Group 1) were compared with patients who were discharged home prior to surgery (Group 2). Demographics, radiographic findings, interventional information, hospital courses, readmissions, and complications were evaluated.

Unpaired t test was used for continuous variables, and categorical variables were analyzed with Fisher's exact test or chi-square test with Yates's correction where appropriate. All data are reported as mean±standard deviation values.

Results

We identified 41 patients; 22 (53.7%) patients were in Group 1. There were no differences in age, gender, or body mass index between the two groups (Table 1).

Results of diagnostic testing and treatment approaches are shown in Table 2. All patients had documented elevation of serum amylase and lipase activities at the time of diagnosis. All patients underwent preoperative ultrasound; 4 had no evidence of stone or sludge but were found to have a dilated common bile duct (CBD) and stones on final pathology (n=1) or choledocholithiasis on computed tomography (n=1) or endoscopic retrograde cholangiopancreatography (ERCP) (n=2). Preoperative computed tomography was used in 10 patients, and magnetic resonance computed pancreatography was used in 4. Hepatobiliary iminodiacetic acid scan was not performed in any patients prior to surgery. Four patients (3 in Group 2) had necrotizing pancreatitis and pancreatic “cyst” formation. One was treated with percutaneous drainage of the cyst followed by interval cholecystectomy. A second underwent cyst gastrostomy at the time of cholecystectomy. The third had documented resolution prior to interval cholecystectomy. The patient in Group 1 underwent a postoperative ultrasound that showed resolution of the “cyst.” The use of ERCP is depicted in Table 2. The 2 patients who did not have stone extraction were found to have cholelithiasis only. A separate patient in Group 2 had successful extraction of a 1-cm-diameter CBD stone on a second preoperative ERCP.

Two of the 7 patients in Group 2 who were readmitted for recurrent pancreatitis prior to their operation were scheduled to have outpatient follow-up in the surgery clinic but canceled their appointments. A third was readmitted the night before planned cholecystectomy, requiring a 3-day hospital stay to “cool down” prior to the operation. Open cholecystectomy was performed in 2 patients in Group 2: one at the surgeon's discretion and the other in combination with an open repeat fundoplication. Single-incision laparoscopic surgery was used in 4 patients in Group 1 and 1 in Group 2; the procedure in the remaining cases was performed using standard four-port laparoscopy. Intraoperative cholangiogram (IOC) was abnormal in 2 patients in Group 1. One patient had findings of choledocholithiasis that was cleared with irrigation of the duct, and another showed no passage of contrast into the duodenum. The latter went onto ERCP and stone extraction.

Data for outcomes and morbidity are shown in Table 3. Of note is that 5 patients in Group 2 undergoing other procedures at the time of cholecystectomy have been excluded from the analysis of operating time and length of stay. Total hospital days attributed to pancreatitis tended to be less in Group 1 compared with Group 2. Postoperative length of stay was not different. Two patients in Group 1 were readmitted postoperatively: one for gastroenteritis and another for recurrent symptoms requiring repeat ERCP from which the child was diagnosed with a choledochal cyst that was later excised. In the latter patient, preoperative imaging consisting of ultrasound and ERCP both demonstrated a dilated CBD with choledocholithiasis, and choledochal cyst was not suspected. Two patients in Group 2 were readmitted. One of these was for abdominal pain where postoperative complication was ruled out with normal labs, computed tomography scan, and ultrasound. The other was for recurrent pancreatitis requiring ERCP and stone extraction. An IOC had not been performed in this patient, but preoperative ultrasound was normal, and pathology showed no stones. There were no intraoperative complications noted in either group.

Three patients from Group 2 deserve further discussion given their unusual presentations or comorbid diseases that likely discouraged surgical referral and cholecystectomy during the first hospital stay. Patient A was an 11-year-old boy with significant reflux and history of aspiration pneumonia who was referred for fundoplication. Given his recent history of pancreatitis and documented cholelithiasis, he underwent laparoscopic cholecystectomy as well as Nissen fundoplication. Patient B was a 2-year-old girl with a history of congenital heart disease, tetrasomy 13, and reflux status post–Nissen fundoplication and gastrostomy who was referred to our clinic after hospitalization for gallstone pancreatitis. History obtained during the visit revealed ongoing symptoms of reflux as well. Therefore, she underwent open repeat fundoplication and cholecystectomy. Patient C was a 14-year-old girl with gallstone pancreatitis who had an ultrasound showing cholelithiasis and a normal CBD diameter. In addition, her total bilirubin was normal. She was discharged home with plans for outpatient ERCP and then cholecystectomy. ERCP done 1 month later showed cholelithiasis only. The following day she returned to the emergency room with symptoms and laboratory findings consistent with recurrent pancreatitis. She was admitted and underwent laparoscopic cholecystectomy without complication.

Discussion

Recurrent pancreatitis in patients with gallstones has been well defined in the adult population but not addressed in the pediatric realm. The findings from this study would suggest the rate of recurrent disease in children treated with interval cholecystectomy is as high as it is in adults.^6–8 Total length of stay was increased by about 5 days in patients who underwent cholecystectomy on second admission, while postoperative length of stay was identical. Rate of readmission was also similar between the two approaches.

Variable practice patterns were observed in our study. Only 53.7% of patients underwent cholecystectomy during index admission. Similar rates have been reported in the adult literature. ¹⁰ Institution of a treatment pathway in this study improved rates of cholecystectomy from 48% to 78%. ¹⁰ The effectiveness of this type of protocol should be further investigated in the pediatric population.

ERCP has been shown to reduce morbidity and mortality in adults with severe biliary pancreatitis. ¹⁴ ERCP in our series was performed by an adult gastroenterologist with extensive experience. Two of 12 (16.7%) patients had complications after undergoing preoperative ERCP, including pancreatitis and abdominal pain requiring overnight observation. Both procedures were performed as an outpatient 24 and 25 days, respectively, after discharge from their index admission. It could be argued that patients that are well enough to be discharged home have likely passed their stone and no longer benefit from ERCP. Complication rates for ERCP in children have been reported to be as low as 4.8%. ¹⁵ No patient who underwent preoperative ERCP had recurrent pancreatitis. IOC is an effective means of examining for choledocholithiasis if preoperative ERCP has not been performed. A review of 48 children suspected of having choledocholithiasis concluded that cholecystectomy with IOC with postoperative ERCP as needed should be performed in this patient population. ¹⁶ It was felt by the authors that this approach reduces the use of potentially unnecessary preoperative ERCP. Choledocholithiasis was found in 10 of 12 patients undergoing preoperative ERCP in our study. Eleven of them had clear indications for ERCP, including jaundice, elevated bilirubin, or radiographic evidence of CBD stones. IOC was positive for CBD stones in only 2 of 11 patients in our study despite suggestion of CBD stones in 9 of these patients. This discrepancy is likely due to either the quick resolution of symptoms in patients in Group 1 or the later time of the IOC relative to ERCP in Group 2. Although no recommendations regarding the use of ERCP or IOC can be made from this study, the authors feel that preoperative ERCP should be reserved for children with objective evidence of choledocholithiasis.

Total hospital length of stay in our study was influenced by the severity of the initial presentation of pancreatitis. One patient in Group 1 and 3 patients in Group 2 presented with complicated pancreatitis and cystic changes of the pancreas. Only 2 patients had cysts large enough to require drainage procedures. Previous studies have shown higher rates of conversion to open surgery, longer postoperative stays, ¹⁷ and higher complication rates ¹⁸ in patients undergoing cholecystectomy for severe pancreatitis during index admission. Patients with peripancreatic fluid collections are probably best observed until the fluid collection either resolves or persists beyond 6 weeks, at which time definitive drainage can be performed. ¹⁸

This is the largest reported series of cholecystectomy for biliary pancreatitis in children. Given high rates of recurrent pancreatitis, cholecystectomy should be offered during the initial hospitalization as is recommended in adults.