Endoscopic retrograde cholangiopancreatography in the treatment of pancreaticopleural fistula in children

Jing Zhang, Liu-Cun Gao, Shu Guo, Tian-Lu Mei, Jin Zhou, Guo-Li Wang, Fei-Hong Yu, Yong-Li Fang, Bao-Ping Xu

Abstract

Key Words: Pancreaticopleural fistula; Childhood; Endoscopic retrograde cholangiopancreatography; Magnetic resonance cholangiopancreatography; Diagnostic; Treatment

INTRODUCTION

MATERIALS AND METHODS

Objective

To explore the treatment response to ERCP for PPF in children.

Setting, design, and sample size

From December 2007 to May 2019, the clinical data of seven children with PFF in our department were retrospectively analyzed. The patients comprised three boys and four girls ranging in age from 2 to 10 years (mean age, 6.57 ± 3.26 years). Their main symptoms were chest distress and pain (n= 3), intermittent fever (n= 3), dyspnea (n= 3), and abdominal pain and distention (n= 4). Five patients had massive pleural effusion, and two had moderate pleural effusion. Three patients had pleural effusion on the right side, one had effusion on the left, and three had effusion on both sides. One patient had a history of abdominal trauma, but no patients had a history of abdominal surgery.

Diagnostic criteria

All seven patients were confirmed to have PPF by laboratory and imaging examinations. The laboratory examinations mainly included pancreatic and pleural effusion biochemical examinations. The imaging examinations mainly included Bultrasound, enhanced computed tomography, magnetic resonance cholangiopancreatography, and ERCP.

2.1 妊娠晚期羊水過少單因素分析 兩組文化程度、經產情況、胎次、胎數及胎兒畸形率比較，差異均無統計學意義(P>0.05)；年齡≥35歲、孕周>41周、妊高癥、過期妊娠、羊水污染、胎兒宮內生長受限、胎膜早破發生率為妊娠晚期羊水過少的影響因素(P<0.05)。見表1。

Treatments

All seven children initially received conservative treatment, including fasting, a somatostatin prescription to inhibit pancreatic secretion, anti-infection medication, and nutritional support. After conservative treatment, the body temperature normalized and pleural effusion disappeared in one patient, while a poor response was seen in six patients. Therefore, two patients were treated by surgery and five underwent ERCP, however, one of the five patients who underwent ERCP required surgery because of ERCP intubation failure.

Literature review

Peer-reviewed English-language publications were retrieved from the PubMed database using the search term “[Pancreaticopleural Fistula] OR [PPF],” and Chinese publications were retrieved from the Wanfang and China National Knowledge Infrastructure databases using the search term “Pancreaticopleural Fistula.” The time limit for the literature search was January 2009 to December 2019.

Statistical analysis

SPSS 22.0 software (IBM Corp, Armonk, NY, United States) was used to analyze the correlation between the length of hospital stay and conservative treatment. Descriptive data are expressed as mean ± standard deviation. The effects of surgical treatment and ERCP were compared by at-test, andP< 0.05 indicated a statistically significant difference.

RESULTS

Diagnostic results

All seven patients with pleural effusion had hemothorax. Four had a leukocyte count of > 500 × 106/L, and five had a pleural effusion protein concentration of > 30 g/L. The concentration of amylase in the pleural fluid was substantially increased in all patients (> 1000 U/L; reference, < 150 U/L); five patients had a pleural fluid amylase concentration of 1000 to 50000 U/L, and two had a pleural fluid amylase concentration of > 50000 U/L. Table 1 shows that five of the seven patients had a high serum amylase concentration (mean, 792.8 ± 409.97 U/L). The serum lipase concentration was increased in all seven children (mean, 1826.1 ± 1650.21 U/L), and one patient had alarge amount of ascites with an amylase concentration of 13053 U/L. Table 2 shows that all seven patients had negative results of acid-fast staining and bacterial culture of the pleural effusion, and no tumor cells were found in the pathological examination. All seven patients were diagnosed with PPF by magnetic resonance cholangiopancreatography. Pulmonary imaging showed a large amount of pleural effusion in all children; the effusion was present on the right side in three children, on the left side in one, and on both sides in three.

第一類：“脫離家庭環境的兒童”。指暫時或永久性脫離家庭環境的兒童，或為其最大利益不能在家庭環境下繼續生活的兒童。這包括被遺棄兒童，被拐賣兒童，孤兒，父母被剝奪監護權的兒童和流浪兒童。

Table 1 Laboratory findings in patients with pancreaticopleural fistula

Treatments and outcomes

All seven patients with PPF were initially treated with conservative therapy for 10 to 60 d (mean, 34.67 ± 22.03 d). Six of them had a recurrent fever and continuous pleural effusion following the conservative treatment. Therefore, five patients underwent ERCP, and one of these patients was transferred to surgery after ERCP intubation failed. The remaining four children who underwent ERCP recovered well without a recurrent fever after the procedure. Their body temperature normalized within 2 to 4 d, and they began to eat within 4 to 6 d. Pump infusion of a somatostatin was continued for 4 to 20 d, and the amylase concentration recovered to normal in 4 to 23 d. The patients underwent 2 to 22 d of closed thoracic drainage; the one child who underwent drainage for 22 d required prolonged drainage because of obstruction of the ERCP tent by small stones. The hospitalization stay after ERCP ranged from 12 to 30 d among these four patients (mean, 18 ± 10.39 d) (Table 3).

Association between overall hospital stay and duration of conservative treatment

SPSS software was used to fit the overall hospital stay and duration of conservative treatment, and a positive linear correlation was obtained (R2= 0.9992) (Figure 1).

Literature review

Articles describing clinical operations for PPF published worldwide during the past decade were reviewed and summarized (Table 4)[6-40]. In total, 37 case reports were found among 35 non-duplicated publications. The 37 patients comprised 25 adults and 12 children. Among seven patients who received conservative treatment, one died of a poor response. Thirteen patients received surgical treatment, and among the 17 patients who received ERCP treatment, three were converted to surgical treatment because of a poor response to ERCP. The duration of conservative treatment ranged from 7 to 60 d (mean, 30.76 ± 17.4 d). The postoperative hospital stay of patients who underwent surgical treatment ranged from 5 to 30 d (mean, 16 ± 10.95 d), and the postoperative hospital stay of patients who underwent ERCP ranged from 12 to 30 d (mean, 18.7 ± 6.88 d). There was no significant difference in the postoperative hospital stay between the two groups (P> 0.05).

DISCUSSION

Treatment status of PPF worldwide

PPF is a rare complication of chronic pancreatitis. The main symptoms of PPF are chest pain, tachypnea, and dyspnea, and the condition is difficult to diagnose. In 1976, Cameronet al[41]considered PPF to be caused by entry of pancreatic secretions into thebody cavity rather than the duodenum. In the present study, PPF originated from a ruptured main pancreatic duct or leaking pseudocyst. If the front of the pancreatic duct is damaged, extrapancreatic secretions will leak into the abdominal cavity, resulting in pancreatic ascites; if the duct is damaged at the rear, extrapancreatic secretions will leak into the mediastinum through the posterior peritoneumviathe aorta or esophageal hiatus; and if the secretion penetrates the pleura, it will cause fluid accumulation (with or without bleeding) in one or both thoracic cavities[41-44]. In adults, PPF is usually secondary to chronic alcoholic pancreatitis. However, the cause is unclear in children.

Table 2 Clinical symptoms, treatments, and outcomes of seven children with pancreaticopleural fistula

CT: Computed tomography; ERCP: Endoscopic retrograde cholangiopancreatography; MRCP: Magnetic resonance cholangiopancreatography.

PPF can be treated by conservative therapy with medication, surgery, or endoscopic technology[9,44]. In previous research, 31% to 65% of adult patients with PPF fully responded to octreotide combined with total parenteral nutrition treatment and usually took 2 to 3 wk to recover[6,45]. However, because of the repeated occurrence of pleural effusion in children, a closed thoracic drainage tube should be placed. During conservative treatment, children may develop malnutrition, catheter infection, septicemia, and other complications that are difficult to treat[46]. Children who undergo failed conservative treatment need further surgical and endoscopic treatment. Surgery is one of the main treatment methods for PPF. The purpose of surgical treatment is to connect the pancreaticojejunal channel to drain fully the pancreatic juice. The most common surgical treatment is pancreatojejunostomy. Frey’s operation can be performed when a pancreatic head mass compresses the pancreatic duct and biliary tract; this procedure involves pancreatectomy and longitudinal pancreatojejunostomy[47]. Placement of an ERCP stent is a new nonsurgical treatment for PPF. An ERCP stent can open the proximal end of the pancreatic duct, smoothly drain the pancreatic juice, allow the pancreatic juice to flow to the duodenum with low resistance, and close the fistula that is abnormally connecting the pancreatic duct and pleura[29].

In the present study, we summarized 37 cases of PPF treatment published in the past decade (25 adults and 12 children). The proportions of adults and children who received conservative treatment, surgical treatment, and ERCP treatment were 16.67% and 25%, 50% and 41.7%, and 42.7% and 33.3%, respectively. However, conservative treatment produced a limited response. One of seven patients who received conservative treatment died, and the success rate was only 16.67%. Surgery washistorically the most frequently used treatment but was invasive. With the development of minimally invasive ERCP in recent years, ERCP is now being increasingly used in the treatment of patients with PPF, especially children.

Table 3 Comparison of therapeutic effect of endoscopic retrograde cholangiopancreatography vs conservative treatment

Optimal PPF treatment method

PPF is a rare disease, and no systematic study has been performed to determine the best treatment; therefore, no consensus has been reached regarding the optimal therapy. Conservative treatment has a low success rate and is associated with many complications, and patients often need secondary surgery or endoscopic treatment. Both surgery and endoscopic treatment can effectively treat PPF. However, no systematic study has been performed to compare the efficacy of the two treatments. We herein performed a preliminary comparison of surgery and endoscopic treatment of PPF by summarizing the treatment results and prognosis of seven children treated in our hospital and both adults and children described in previous publications worldwide. The first case of PPF cured by surgery was reported in 1960[48]. The first adult with PPF cured by ERCP was reported by Saeedet al[5]in 1993. Current research data show that more adults and children with PPF choose ERCP treatment.

All seven patients with PPF in this study initially received conservative treatment, but the responses were poor. Among the five patients who received ERCP treatment, one was converted to surgery because of incubation failure; the treatment success rate was thus 80%. Two patients underwent surgery (one was lost to follow-up after transfer to another hospital), and both recovered. The mean postoperative hospital stay for the two patients who underwent surgery and the four patients who underwent ERCP was 25 d and 19.25 d, respectively. The preliminary conclusion was that the recovery time was shorter after ERCP than after surgical treatment. However, because of the small number of cases, the hospital stay of the two treatment methods could not be statistically analyzed. The present study also showed that patients with PPF who undergo ERCP require a very short time until they start to eat, discontinue somatostatin pump maintenance, return to a normal amylase concentration, and discontinue closed thoracic drainage.

Because children very rarely develop PPF, the present study summarized the clinical outcome data for both adults and children with PPF worldwide during the past decade for a comprehensive analysis. The mean postoperative hospital stay of patients treated with surgery and ERCP was 16 ± 10.95 d and 18.7 ± 6.88 d, respectively (P> 0.05). There was no significant difference in the postoperative hospital stay between the two treatment methods, and the curative effect of the twomethods was equivalent. The success rate of ERCP treatment (80%) was slightly lower than that of surgical treatment (100%), which may have been due to the small number of patients. In some studies, the duration of using ERCP to cure PPF was 4 to 12 wk with different success rates. The success rates reported by Khanet al[49], Paiet al[50], and Varadarajuluet al[51]were 100%, 96.4%, and 50.0% (the low success rate was due to stent placement failure or failure to pass through the pancreatic duct rupture site), respectively, similar to the surgery success rate (94%) reported by Kinget al[7]. These findings indicate that the success rate of ERCP treatment is not lower than that of surgical treatment. Our data are consistent with the findings of most previous studies; statistical analysis was impossible because of the limited sample size. The results of the literature review of studies published in the past decade indicated that the average recovery time following ERCP was slightly longer than that following surgery. This result might have been related to either variations in techniques between surgery and ERCP or limited information from the publications reviewed. The literature describes multiple surgical procedures (including distal pancreatectomy with splenectomy, pancreatic duct anastomosis with an intestinal loop, pancreaticoduodenectomy, cystogastrostomy, and cystojejunostomy)[14], which are traumatizing and associated with many complications such as leakage, intra-abdominal infections, and fistula recurrence[7]. No further analysis was performed because of the limited number of cases reported. In addition, the results showed that the standard deviation of the ERCP group was smaller, suggesting that the ERCP group had less invasive treatment, a shorter postoperative recovery time, and a lower incidence of complications (infection, bleeding, destruction of pancreatic duct anatomy, repeated fluid accumulation, and pancreatitis). All four patients treated with ERCP reportedly had a good prognosis with no complications. The standard deviation of the postoperative recovery time in the surgery group was larger, indicating that the postoperative recovery time in the surgery group had greater variation and higher uncertainty. In summary, we believe that ERCP can reduce the hospitalization time and should be the preferred treatment for PPF in children.

Table 4 Worldwide cases of pancreaticopleural fistula published in the most recent 10 years

2019[40]M/14 CT Conservative/operation 30 AVG 30.76 ± 17.4 16 ± 10.95 18.7 ± 6.88 P value P > 0.05 CT: Computed tomography; ERCP: Endoscopic retrograde cholangiopancreatography; MRCP: Magnetic resonance cholangiopancreatography; MRI: Magnetic resonance imaging.

Figure 1 The correlation between endoscopic retrograde cholangiopancreatography intervention time and total hospital stay. The linear equation is not a model prediction but only a correlation analysis.

Aswaniet al[3]also reported that after ERCP, patients can quickly transition to the oral feeding stage and have a short recovery time, which reduces the hospital stay and mortality rate compared with a traditional operation. Therefore, existing research suggests that ERCP should be the first choice for patients with PPF who have a poor response to conservative treatment, and only after failure of conservative treatment and ERCP treatment should surgical treatment be considered. Because of the limited number of patients in the present study, further prospective studies are needed to compare the cost-effectiveness and long-term results of ERCP and surgery.

Best operation time for ERCP

Patients with PPF initially receive conservative treatment and will choose surgery or ERCP treatment if their condition does not fully respond. We recommend ERCP as the first-choice treatment. Pleural effusion readily recurs after conservative treatment, potentially resulting in malnutrition, catheter infection, septicemia, and other complications. A longer duration of conservative treatment is associated with a greater risk for the patient. The present study investigated the relationship between the duration of conservative treatment and the overall hospital stay. The fitting analysis of the conservative treatment time and the total length of stay of three patients who received ERCP showed a positive linear correlation and suggested that a shorter conservative treatment time is associated with earlier performance of ERCP and a shorter overall hospital stay. Although conservative treatment has a certain response rate for PPF, the rate is very low, and the treatment cycle is long. Some researchers have proposed that conservative treatment should only be used as the initial stage of PPF treatment to stabilize the condition and should not be used as the treatment plan for PPF[7,52]. For patients with PPF, the duration of conservative treatment should be reduced, and ERCP treatment should be carried out as early as possible.

CONCLUSION

In conclusion, the success rate of ERCP for patients with PPF was similar to that of surgical treatment, and the prognosis was not worse than that of surgical treatment. Compared with traditional surgery, ERCP does not require laparotomy, is a simple operation, induces less trauma and fewer complications, and promotes rapid fast recovery. Thus, it is very suitable for children and advanced-age patients who cannot tolerate surgery or have poor health conditions. Earlier performance of ERCP promotes faster recovery and a shorter total length of stay. Therefore, ERCP is recommended as the first-choice treatment for PPF in children. ERCP should be performed as early as possible if conditions permit during conservative treatment. Because PPF is a rare disease and it is difficult to obtain data on clinical cases, the present study included only seven patients, one of whom was lost to follow-up after discharge. Thus, we were unable to perform a scientific and systematic comparative analysis on the curative effect of surgery and ERCP. The conclusions of this study still need to be validated.

ARTICLE HIGHLIGHTS

Research results

There was no significant difference in the postoperative hospital stays between surgical treatment and ERCP. However, there was a positive linear correlation between the overall hospital stay and ERCP intervention time.

Research conclusions

ERCP is recommended as the first-choice treatment of PPF in children. ERCP should be performed as early as possible if conditions permit.

Research perspectives

Because PPF is a rare disease and it is difficult to obtain data on clinical cases, the present study included only seven patients, one of whom was lost to follow-up after discharge. Thus, we were unable to perform a scientific and systematic comparative analysis on the curative effect of surgery and ERCP. The conclusions of this study still need to be validated.

ACKNOWLEDGEMENTS

We thank the families of these patients for their support in this study.