How and When Should NSAIDs Be Used for Preventing Post-ERCP Pancreatitis? A Systematic Review and Meta-Analysis 

Endoscopic Retrograde Cholangiopancreatography (ERCP) is a vital diagnostic and therapeutic procedure in the medical field that was developed in 1968. It is primarily used for evaluating and treating conditions affecting the biliary and pancreatic ducts, by insertion of an endoscope through the mouth and into the duodenum to access the bile ducts and pancreas. ERCP is valuable for diagnosing diseases such as gallstones, pancreatitis, and tumors, allowing for precise localization and tissue sampling. It also enables the removal of gallstones or the placement of stents to relieve obstructions. The most common complication post-ERCP is pancreatitis, followed by infection, bleeding, or perforation of the bile or pancreatic duct. A variety of medical and surgical strategies have been studied for reducing the risk of post-ERCP pancreatitis (PEP), with the most effective methods being temporary pancreatic duct stenting and rectal NSAID.

In a 2014 systematic review, researchers aimed to determine the most effective NSAID for preventing post-ERCP pancreatitis and identify specific patient populations and administration routes where NSAID usage could be superior. They conducted an extensive search across Scopus, Medline, Cochrane Library, and ISI Web of Knowledge to locate randomized control studies with primary outcomes focused on NSAID efficacy in preventing PEP. They also examined secondary outcomes, including the effectiveness of NSAIDs in reducing the severity of moderate-severe pancreatitis, adverse effects associated with NSAID use, the duration of hospitalization, and mortality rates. Among the 9 studies identified, the results revealed that a single rectal dose of either indomethacin or diclofenac, administered immediately before or after ERCP, significantly reduced the risk of pancreatitis in both high-risk and average-risk patients, regardless of whether a stent was placed. Specifically, the incidence of pancreatitis was 7.4% (80 out of 1077 individuals) among those who received NSAIDs, compared to 14.6% (154 out of 1056 individuals) among those who did not, resulting in a reported risk ratio of 0.51 and a number needed to treat of 14. Analyzing the secondary outcomes, it was found that both indomethacin and diclofenac statistically lowered the risk and severity of pancreatitis, with the rectal route proving to be the only effective route compared to other administration routes like oral and intramuscular. While a decrease in hospital stay was observed in some studies, further investigations are required to confirm this association. Notably, the researchers also observed that these results were not statistically significant for males and the elderly; however, this discrepancy may be attributed to the limited sample sizes in some of the studies.  

The findings of this study provide evidence that the safe administration of rectal NSAIDs before or immediately after ERCP effectively reduces the risk of post-operative pancreatitis. Remarkably, this benefit extends to individuals of all risk profiles, regardless of whether a stent was inserted during the procedure. The comparable efficacy observed in cases with and without stent placement suggests that this approach could offer a cost-effective strategy for preventing PEP.  

It is important to acknowledge certain limitations in the study. The researchers reported one limitation to be the possibility of an erroneous high estimation of risk reduction due to the high prevalence of PEP in the control group which could be attributed to many studies focusing on high-risk individuals, a second limitation to be the inclusion of three low-quality studies though their removal yielded similar results in the main analysis, a third limitation to be different methods in assessing the severity of the condition (e.g Ranson’s criteria or CT findings), and the final limitation being the study’s reliance on relatively small sample sizes.   

Source:

Puig, Ignasi et al. “How and when should NSAIDs be used for preventing post-ERCP pancreatitis? A systematic review and meta-analysis.” PloS one vol. 9,3 e92922. 27 Mar. 2014, doi:10.1371/journal.pone.0092922

Site Evaluator: Andrea Pizarro

During my initial encounter, I presented a detailed history and physical examination (H&P) of a 63-year-old male patient who visited the Emergency Department at the recommendation of his podiatrist. The reason for his visit was to investigate a chronic right hallux ulcer that was healing poorly and to rule out osteomyelitis. The patient described experiencing mild swelling in the affected area and reported difficulty ambulating due to pain, which worsened when he walked. On the physical examination, I observed a 1.5 cm poorly healed ulcer located at the medial aspect of the proximal interphalangeal joint of the right hallux. Additionally, there was mild swelling and erythema extending from the forefoot to the midfoot. Considering the patient’s medical history, which included poorly controlled diabetes type 1, hypertension, peripheral arterial disease, coronary artery disease, and poor perfusion, along with his presenting symptoms and a positive bone-to-probe test, acute osteomyelitis emerged as a highly likely diagnosis. I selected this case to deepen my understanding of osteomyelitis and its diagnostic workup. When presenting the case to Ms. Pizzaro, we delved into the SAFARI procedure that the patient had undergone in 2021 and reviewed my differential diagnoses. Ms. Pizzaro concurred with the proposed management plan.

For my final site visit, I presented the case of a 64-year-old female who arrived at the Emergency Department with epigastric pain that extended to her back. She also experienced nausea and vomiting two days after undergoing an ERCP procedure. The patient’s clinical symptoms, along with an elevated lipase level and findings from the CT scan, strongly indicated the development of acute pancreatitis as a post-surgical complication. In my differential diagnosis, I considered the possibility of duodenal perforation post-ERCP, peptic ulcer, and acute gastritis. Subsequently, she was admitted to the medical ward for further care. I selected this case for presentation not only because I had the opportunity to conduct initial patient rounds before the Physician Assistant, but also because it allowed me to gain additional insights into the common occurrence of acute pancreatitis as a post-ERCP complication. During the discussion with Ms. Pizzaro, we found alignment in our assessment of the differential diagnoses and the proposed management plan.

The third case I submitted but didn’t present was a 42-year-old male who presented in the ED complaining of chest pain. The patient had negative cardiomarkers and a normal sinus rhythm on the EKG. ACS work up was started in the ED and continued on the medical floor. 

One aspect of the H&Ps that I did a bit differently for this rotation was that I included patient education at the end. To write out what I would say to the patient helped me better understand the case in a way that I can explain it clearly to the patient in non-medical terms.  

For my initial visit, I presented an H&P on a 28-year-old Guyanese female who presented at 36-week gestation for her initial OB visit. Her history included a stillbirth at 37 weeks for her first pregnancy due to an undetected clot in the umbilical cord that was discovered upon autopsy, prompting her to seek care in the United States. I chose this case because I thought her history was interesting and she was one of the first patient’s that I performed the fetal Doppler and measured fundal height on by myself. As this was her first initial prenatal visit at Queens hospital, we had ordered all the initial lab work as well as delivering third trimester precautions. Due to her history of an umbilical cord clot, she was referred to follow up in high-risk clinic. 

For my final visit, I submitted two more H&P. The first one was an ED consult of a 27-year-old female complaining of gradual worsening right pelvic pain that started a week prior. The only contributing factor that she could think of was she had protected sexual intercourse 2 days prior to pain onset. She denied vaginal symptoms, nausea, vomiting, fever, chills, chest pain, shortness of breath, appetitive change, or abnormal bowel movements. Her labs were unremarkable and her pregnancy test was negative. A ultrasound was performed which confirmed a right sided hemorrhagic cyst measuring 5x4cm. The plan was symptom control and discharge. I chose to include this case because of the importance of needing to rule out ovarian torsion and ectopic pregnancy in a reproductive aged woman presenting with pelvic pain. The risk of these emergent conditions was of low likelihood given the patient’s symptoms were gradual in onset, she is not in acute distress, and the doppler ultrasound displayed blood flow to both ovaries. Professor Melendez agreed with the workup process and the diagnoses. My second H&P was another ED consult of a 31-year-old Hispanic female 20 weeks gestation who was complaining of decreased fetal movement for more than 12 hours and vaginal bleeding that had started 4 hours prior. She was also complaining of intermittent abdominal pain. Her OB history was remarkable for a c-section due to potential infection and a termination of pregnancy via D&C. Besides for that, her medical history was unremarkable and she had been consistent with her parental visits without any complications. Her lab work was unremarkable. Her transabdominal ultrasound showed that the fetus was in vertex position with the posterior placenta edge noted to be low lying measuring 1.5 cm (15 mm) away from the internal cervical os. A fetal heart rate was heard and fetal movement was confirmed. Given the patient’s symptoms, history of past c-section and D&C, and transabdominal exam findings, the patient’s symptoms were most consistent with posterior marginal, placenta previa. Patient is in no acute distress and is vitally stable. Non-contracting, closed cervix, and fetal well-being reassured. The patient was scheduled for a transvaginal ultrasound to confirm the diagnoses and was referred to the high-risk clinic for close monitoring of the placental position. The patient was also given preterm labor precautions. Professor Melendez overall agreed with the management and care but it prompted a discussion on when or if a speculum exam/digital exam should be performed with only the results of a transabdominal ultrasound and not a transvaginal ultrasound in a suspected case of placental previa.