The role of salvage ERCP for the treatment of post-ERCP pancreatitis

We read with great interest the article by Kerdsirichairat et al. titled “Urgent ERCP with pancreatic stent placement or replacement for salvage of post-ERCP pancreatitis” [1]. The authors reported their experience of performing an urgent endoscopic retrograde cholangiopancreatography (ERCP) to place a pancreatic stent in patients who developed post-ERCP pancreatitis (PEP). PEP was defined by abdominal pain (using the revised Atlanta criteria), serum amylase level at least three times greater than the upper limit of normal, and more than one night of hospitalization (the so-called “Cotton criteria” [2]). Salvage ERCP was attempted 2 hours or more after clinical onset of pain. Compared with before salvage ERCP, the median pain score (P < 0.001), and median amylase (P = 0.003) and lipase (P = 0.001) levels significantly improved 24 hours after the procedure. In addition, systemic inflammatory response syndrome (SIRS) resolved in 24 hours (P < 0.001).

However, possible bias and methodological shortcomings limit the validity of the study and the conclusion. Although not clearly stated in the text, we believe that this study was a retrospective analysis of a prospectively collected database. In addition, out of 3216 patients, 64 (2 %) developed PEP, but 7 of them were excluded because of incomplete information. Only 14 (25 %) of the remaining 57 patients were treated with salvage ERCP, and no information regarding the other 43 patients (75 %) included in the study were given, representing a potential bias. Unexpectedly, the authors did not introduce any comparison group, focusing the analysis instead on internal data (i. e. improvement of PEP after salvage ERCP). We would be interested to know the following: What were the main characteristics of the 14 included patients compared with the remaining 43 patients? Why were these 43 patients not considered for salvage ERCP? What were the selection criteria? Would it be possible for the authors to state whether the outcome of the 14 patients was significantly different from that of the remaining 43 patients who did not undergo salvage ERCP? This information is essential in order to understand whether, and to what extent, the information acquired on the 14 cases can be generalized to the remaining patients.

Most interestingly, the diagnosis of PEP was established according to Cotton consensus criteria (i. e. amylase at least three times the normal level at more than 24 hours after the procedure and requiring admission or prolongation of planned admission to 2 – 3 days) [2]. However, the median onset of PEP was 5 hours after ERCP in patients with a prophylactic pancreatic stent and 2 hours after ERCP in patients without a prophylactic pancreatic stent. According to these data, the onset of PEP did not meet the Cotton criteria for the diagnosis of PEP in almost all of the included cases. Furthermore, the salvage ERCP was performed at a median of 10 hours after the clinical onset of PEP. In other words, most of the patients underwent a so-called “salvage” ERCP before a diagnosis of PEP could be correctly established.

The final results of the study showed that PEP resolved 24 hours after salvage ERCP in all cases, indicating that this strategy was associated with rapid resolution of clinical pancreatitis, SIRS, and elevated lipase and amylase levels after ERCP.

In conclusion, because of the abovementioned limitations (i. e. the presumed retrospective nature of the study, the lack of inclusion/exclusion criteria for salvage ERCP among those with a presumed diagnosis of PEP, the lack of a control group) we agree with the authors that the current data should be considered extremely carefully, until well-performed, prospective, randomized studies can clarify the real effectiveness of urgent salvage ERCP for the treatment of post-procedure acute pancreatitis.

• References

• 1 Kerdsirichairat T, Attam R, Arain M et al. Urgent ERCP with pancreatic stent placement or replacement for salvage of post-ERCP pancreatitis. Endoscopy 2014; 46: 1085-1091
• 2 Cotton PB, Garrow DA, Gallagher J et al. Risk factors for complications after ERCP: a multivariate analysis of 11,497 procedures over 12 years. Gastrointest Endosc 2009; 70: 80-88