Background:
There are no randomized controlled trials demonstrating the efficacy of endoscopic decompression in patients with painful chronic pancreatitis.
Objective: To investigate the pain relief effect of pancreatic catheter decompression in patients with chronic pancreatitis and intraductal stones.
Experimental design:
A 24-week parallel controlled randomized trial (ClinicalTrials.gov number: NCT03966781).
Place:
February 2021 to July 2022, Asian Gastroenterology Institute, India.
Participator:
106 patients with chronic pancreatitis.
Interventions:
A combination of extracorporeal shock wave lithotripsy (ESWL) and endoscopic retrograde pancreatography (ERP) was compared with sham.
Measurement Indicators:
The primary endpoint was pain relief on a visual analogue score (VAS) of 0 to 10 at 12 weeks. Secondary outcomes were assessed after 12 and 24 weeks and included complications of 30% pain relief, opioid use, pain-free days, questionnaires, and interventions.
Outcome:
The ESWL/ERP group included 52 patients and the sham group included 54 patients. At 12 weeks, pain relief was better in the ESWL/ERP group than in the sham group (mean difference in change on VAS was -0.7 [95% CI, -1.3 to 0]; P=0.039）。 There was no persistent difference between groups at 24 weeks of follow-up, nor was there a difference in 30% pain relief at 12 and 24 weeks of follow-up. There was no daily increase in pain at 12-week follow-up (median difference 16.2 days [CI, 3.9 to 28.5 days]) and a decrease in the number of days of opioid use (median difference -5.4 days [CI, -9.9 to -0.9 days]) at 12-week follow-up. Safety outcomes were similar between the two groups.
Locality:
Single-center studies and follow-up were limited.
Conclusion:
For patients with chronic pancreatitis and intraductal stones, the combination of ESWL and ERP can provide short-term pain relief.
Major Funding Organisations:
Asian Gastroenterology Institute and Aalborg University Hospital.
This study was carried out using a DORNIER Delta III lithotripter
Pain is a major challenge in chronic pancreatitis and is often associated with pancreatic ductal obstruction (caused by stones and/or strictures). Endoscopic interventions are often used in conjunction with extracorporeal shock wave lithotripsy (ESWL) to relieve pain through decompression of the pancreatic catheter (1). However, the exact relationship between pain and pancreatic duct morphology remains unclear, and the available evidence relies primarily on observational studies and procedural comparisons to assess the effectiveness of these interventions (2-5). In addition, it is well known that invasive interventions such as surgery and endoscopic therapy may trigger significant sham effects for pain-related conditions (6). Therefore, there is a need for rigorously designed randomized trials, including sham-operated groups, to comprehensively assess the efficacy and safety of these interventions.
To fill this knowledge gap, we initiated the SCHOKE (Extracorporeal Shock Wave Lithotripsy and Endoscopic Treatment for Chronic Pancreatic Inflammation Pain) trial to evaluate the pain relief and safety of pancreatic catheter decompression using ESWL and endoscopic retrograde pancreatography (ERP) compared to sham surgery.
way
Trial design and supervision
This is a single-center, single-blind, parallel-group, randomized, sham-controlled superiority trial. The study protocol, which is available on Annals.org, follows the SPIRIT (Standardized Protocol for Interventional Trials Project Recommendation) guidelines and the principles of the Declaration of Helsinki. The study protocol has been previously published (7). An independent data and safety monitoring committee oversaw the trial and consisted of a neurologist, an anesthesiologist and a statistician.
Trial population and setting
We evaluated consecutive patients (aged ≥ 18 years, both men and women) who were presented to the centre pancreas clinic and were identified with chronic pancreatitis (requiring pancreatic calcification, Cambridge III/IV pancreatic duct abnormality, or histologic confirmation) according to Mayo clinic criteria for inclusion from 1 February 2021 to 31 July 2022 (8). Inclusion criteria included chronic abdominal pain with features of chronic pancreatitis, pain intensity greater than 3 on a visual analogue score (VAS) of 0 to 10, meeting criteria for chronic pain (pain ≥ 3 days per week for ≥ 3 months), and occlusion of the pancreatic duct due to intraductal stones with dilation proximal to the site of obstruction. Magnetic resonance cholangiopancreatography or abdominal computed tomography is used to evaluate pancreatic ductal dilation and the presence of intraductal stones. We excluded patients who had previously undergone pancreatic surgery, ESWL, or endoscopic therapy for pancreatic catheters. In addition, we excluded patients with multiple pancreatic ductal stenosis, pancreatic mass lesions, massive ascites, and large accumulation of pancreatic or peripancreatic fluid. For more details on the inclusion and exclusion criteria, please refer to the Supplementary Materials (available on Annals.org).
The study was conducted at the Asian Gastroenterology Institute of India (Hyderabad, India), a large tertiary care referral centre serving patients in India and neighbouring countries. Two members of the research team initially screened potential patients for eligibility at the pancreas clinic. Once enrolled and randomized, the ESWL procedure is performed by two dedicated physicians and a dedicated anesthesiologist who is responsible for performing spinal anesthesia. The ERP procedure is performed by 4 senior endoscopists with at least 10 years of experience.
Surgical procedure
Patients were randomized in a 1:1 ratio to receive a combination of ESWL and ERP or sham treatment. Concealed dispensing with an automatic dispensing system. We randomized in blocks of six patients to ensure a balanced distribution between sham and ESWL/ERP procedures. The randomization process is performed by a physician who is unable to obtain the patient's clinical data.
In the ESWL/ERP group, patients underwent ESWL under spinal anesthesia to remove pancreatic stones. During the procedure, the patient's eyes are gently covered. After spinal anesthesia is done, the German Dornier Delta 3; Dornier MedTech GmbH) performs ESWL, giving up to 5000 shocks at a rate of 90 per minute. If the stone is not fully rubbed during the initial ESWL, additional sessions are scheduled for subsequent days. Complete lithotripsy is defined as the fragmentation of all stones to 3 mm or less (1). On the second day of lithotripsy, the lithotripsy is removed by an ERP procedure. This involves pancreatic sphincter incision and stone removal, and verifies the patency of the pancreatic ducts. Complete pancreatic ductal patency is defined as a reduction of more than 90% in initial stone volume based on the density of stones observed on fluoroscopy during ESWL and subsequent ERP procedures (9). During ERP, if intubation is not possible after a maximum of 5 attempts, a pre-cut pancreatic sphincterotomy will be performed. If pancreatic ductal stenosis is found during the ERP procedure, dilation will be performed prior to insertion of the pancreatic catheter stent. All patients in the ESWL/ERP group were fitted with a plastic pancreatic stent (5 to 7 Fr in diameter and 7 to 10 cm in length) according to our clinical routine. A follow-up ERP procedure is planned after 6 months to replace or remove the pancreatic stent after completion of all study assessments.
In the sham treatment group, patients were given a superficial stinging pain using a sterile hypodermic needle to mimic the sensation of spinal anesthesia before undergoing the sham ESWL procedure. Then, start the lithotripsy without physical contact with the patient. The patient's eyes are gently covered throughout the procedure. Subsequently, after the sham ESWL, the patient underwent a sham ERP procedure, which consisted of insertion of the duodenum using an upper endoscope, but without touching the nipple area and pancreatic ducts.
After the surgery, the patient is observed in the endoscopic recovery area for 6 to 12 hours. If the patient develops abdominal pain during observation, serum lipase is evaluated to determine the occurrence of postoperative pancreatitis.
Patients are advised to continue their usual medical treatments, including antioxidants and pancreatic enzyme supplements, throughout the trial. Pain management was documented at the screening visit and at 12- and 24-week follow-up. In addition, patients were allowed to use NSAIDs or opioids as emergency pain relievers during the trial, and were required to record the type and use of the medication in a pain diary.
终点
The investigators evaluating the treatment outcome were not aware of the treatment allocation.
Based on a VAS of 0 to 10, the daily average clinical pain intensity score was recorded in the patient's pain diary, where the baseline pain intensity score was recorded one week prior to the intervention and subsequently weekly for 24 weeks. The average of pain scores was calculated weekly to adjust for daytime variation in pain intensity. We compared the mean change in pain scores to baseline and used the difference between groups after 12 weeks as the primary endpoint. Weekly telephone interviews with the study coordinator to facilitate accurate recording and adherence to the registry of pain scores.
Pre-specified secondary endpoints were assessed at 12 and 24 weeks of follow-up and included:
i) partial pain relief (defined as a 30% reduction in pain diary score [VAS] compared to baseline),
ii) number of pain-free days during follow-up,
iii) cumulative use of opioid analgesics (number of days of opioid use),
iv) hospitalizations during the follow-up period, and
v) Group differences between baseline and changes in pain diary scores, quality of life (EORTC Quality of Life Questionnaire) and changes in composite pain severity and interference scores of the Brief Pain Inventory–Short Form questionnaire. In addition, at 12- and 24-week follow-up, differences between groups were compared between groups for depression and anxiety symptoms (Hospital Anxiety and Depression Scale) and the proportion of significant improvement in health status (Patient Global Impression of Change Scale). A detailed description of the questionnaire can be found in the supplementary materials.
All primary and secondary endpoints are pre-specified in the study protocol. During our data analysis, three secondary endpoints (days of opioid use, hospitalization, and presence of anxiety and depression symptoms) were redefined (see Supplementary Table 1 for Annals.org). For information on the process for preprocessing data prior to analysis, see Supplementary Table 2 of the Annals.org.
Safety endpoints include assessment of postoperative acute pancreatitis, perforation, bleeding, and infection according to modified Atlanta criteria (10) (as defined in supplementary material).
Statistical analysis
The sample size of the study was designed to detect a minimum difference of 30% in the change in weekly pain diary scores between groups at 12 weeks. We assume a baseline pain diary score of 5 on a VAS of 0 to 10 and expect a response rate of 20% in the placebo group, resulting in a pain score reduction from 5 to 4 on VAS, while a further 30% reduction in the ESWL/ERP group results in a VAS score of 2.8 (11). This corresponds to a group difference of 1.2 on VAS. Based on the assumed standard deviation of 45%, we determined that a study of 48 patients per group was needed to provide 90% power at a two-sided significance level of 0.05 (12). To account for a 10% dropout rate, the sample size was set at 106 patients.
A prespecified interim analysis was performed after randomization and completion of 24-week follow-up in 45 patients (7). No statistically significant differences were observed between the groups (see Annals.org supplementary graph), and the Data and Safety Monitoring Committee recommended that the trial be continued and completed.
The intent-to-treat population included all randomised patients, and trial data were analysed according to the intent-to-treat principle. Categorical variables are reported as counts and percentages, and continuous variables are reported as mean and standard deviation or median and quartile percentiles. The primary outcomes were analysed using a linear mixed model of repeated measures, and summary statistics of mean pain scores (VAS) at each time point were provided with corresponding 95% confidence intervals. Secondary outcomes were analysed using mixed-effects models (same as the primary endpoint), quantile regression models, and risk differences. After the missing values were processed by multiple imputation techniques, the variables were analyzed. For the primary endpoint, a two-sided P value of less than 0.049 was considered statistically significant (for consideration of interim analysis). Additional information on statistical methods can be found in the supplementary material. All analyses were performed using STATA version 17.0 (StataCorp).
The role of funding sources
The study was supported by our institution. The funding agency had no influence on study design, data collection, statistical analysis, manuscript preparation, or publication decisions.
To be continued......