Protocol-based Medical Management of Post-ERCP Pancreatitis
Protocol-based Medical Management of Post-ERCP Pancreatitis
Acute pancreatitis is a common and dreaded complication of endoscopic retrograde cholangiopancreatography (ERCP) that continues to disturb endoscopists since the inception of this technique. Post-ERCP pancreatitis has been reported in up to 40% of patients, depending on the criteria used for diagnosis, differences in patient populations, endoscopic techniques used, and accuracy of the follow up. A recent survey of prospective studies from 1977 to 2006 involving 16855 patients, reported that post-ERCP pancreatitis occurred in 3.47% of patients. However, in high-risk conditions, such as sphincter of Oddi dysfunction, the incidence of acute pancreatitis can be as high as 31%. Even though post-ERCP pancreatitis is usually a mild disease with a favorable outcome in the majority of cases, it can cause significant discomfort to the patient, increase treatment costs, and be the focus the attention (and tension) of the treating endoscopist until the patient recovers completely.
The pathogenesis of post-ERCP pancreatitis is not completely understood. It seems to be an inflammatory response to mechanical, chemical, thermal, hydrostatic, enzymatic, allergic, and microbiological insults that result from papillary instrumentation and/or injection of contrast media for imaging of the pancreatic duct. These inciting factors lead to premature intracellular activation of pancreatic proteolytic enzymes, autodigestion, and release of inflammatory cytokines that produce both local and systemic effects.
There have been several efforts to develop an effective preventive strategy that will eliminate or reduce the risk or severity of post-ERCP pancreatitis. Most of the pharmacological agents tested are not universally effective, hence a combined strategy is being advocated to minimize the occurrence of post-ERCP pancreatitis. This includes the identification of patient- and procedure-related factors associated with high risk, refinement of endoscopic methods, search for an effective pharmacological agent for prophylaxis, and prophylactic placement of a pancreatic duct stent in high-risk patients.
A number of prospective studies have evaluated patient-related risk factors for post-ERCP pancreatitis. These have found that patients under 50 years of age, female sex, normal bilirubin, suspected sphincter of Oddi dysfunction, small bile duct diameter, past history of post-ERCP pancreatitis, and absence of chronic pancreatitis are associated with high risk of post-ERCP pancreatitis. Extra caution and appropriate prophylactic measures should be adopted in these patients. These factors can act synergistically, thereby putting patients with multiple risk factors at high risk for post-ERCP pancreatitis.
Just as there are patient-related risk factors, various procedures have also been associated with higher risk of post-ERCP pancreatitis, as documented by meta-analysis and multivariate analyses. These include ≥1 injection of contrast into the pancreatic duct, pancreatic sphincterotomy, balloon dilatation of an intact biliary sphincter, difficult or failed cannulation, failed clearance of bile duct stones, and precut sphincterotomy. Other factors, such as pancreatic acinarization, pancreatic brush cytology, pain during ERCP, low ERCP volume, and involvement of trainees during ERCP can also be associated with an increased risk of post-ERCP pancreatitis.
A proper understanding of the patient- and procedure-related risk factors has led to a strategy of carefully selecting patients for ERCP, for example, avoiding diagnostic ERCP, using magnetic resonance cholangiopancreatography and endoscopic ultrasound (EUS) for diagnostic purposes, and the use of meticulous endoscopic techniques. The latter include atraumatic manipulation of the papilla, avoidance of pancreatic duct injection for biliary cases, avoidance of acinarization, using an aspirating catheter technique or solid-state manometry system for suspected sphincter of Oddi dysfunction, limited use of precut sphincterotomy, biliary sphincterotomy from 11.00 to 1.00 o'clock position avoiding the pancreatic sphincter, using pure cut electrosurgical current or adjustment of automated generator to lower tissue effect during pancreatic sphincterotomy, avoidance of balloon dilatation of intact biliary sphincter, avoiding placement of biliary stent through intact sphincter in patients with sphincter of Oddi dysfunction, and using soft-tipped guide wires to access bile ducts.
Despite selecting proper patients and using meticulous techniques, post-ERCP pancreatitis occurs when it is least anticipated. Therefore, various pharmacological interventions and short-term placement of a pancreatic duct stent have also been evaluated for its prevention. Drugs that either affect the sphincter pressure or contractility and pancreatic secretion or the inflammatory cascade have been tested in various studies, but the results have been conflicting. For the prevention of post-ERCP pancreatitis: somatostatin, octreotide, gabexate mesylate, ulinastatin, allopurinol, corticosteroids, pentoxifylline, non-steroidal inflammatory drugs (NSAIDs), such as diclofenac and indomethacin, interleukin-10, glyceryl trinitrate, nifedipine, antibiotics, botulinum toxin, locally sprayed lidocaine, N-acetylcysteine, heparin and its derivatives, and β-carotene have been tried in clinical studies, while in animal models, thalidomide, rosiglitazone (PPARγ agonist), and the intraductal administration of a natural killer-1 receptor antagonist have been tested. Recent studies on rectal NSAIDs, including one in the Journal of Gastroenterology and Hepatology, prolonged infusion of a high dose of octreotide and a bolus injection of somatostatin, as shown in a meta-analysis with encouraging results. Further studies and more data are needed before putting these agents to use in routine clinical practice.
Studies have shown that maintaining the flow of the pancreatic juice through the papilla by the placement of a short-term transpapillary narrow caliber pancreatic duct stent is effective in reducing the risk of post-ERCP pancreatitis. While evaluating the cost effectiveness of three strategies (no prophylactic stents,prophylactic stents in all patients, and prophylactic stents in high-risk patients), the selective placement of prophylactic pancreatic stents in high-risk patients was the most cost-effective plan. Placing a pancreatic stent requires considerable expertise, especially after a biliary procedure; a failed pancreatic duct stent insertion may be more harmful than not attempting to place it. Another fear of pancreatic stent placement is the development of stent-induced ductal and parenchymal changes mimicking chronic pancreatitis. This can be disastrous as the majority of these patients have normal pancreatic ducts before stenting. Furthermore, another endoscopy is required to remove the pancreatic stent. These potential problems of stent placement can be avoided by using softer and smaller diameter 3F or 4F stents with an unflanged inner end. The advantages of these stents are lower rates of post-ERCP pancreatitis, decreased frequency of stent-induced changes in the pancreatic duct, and a higher rate of spontaneous dislodgement of the stent obviating the need for repeat endoscopy. However, it should be kept in mind that deploying 3F or 4F stents requires thin (0.018 or 0.025 inches) guide wires and considerable expertise in pancreatic endotherapy. It would be interesting to follow the results of prophylactic stenting from centers with low ERCP load and with relatively less-experienced endoscopists.
For an early diagnosis of post-ERCP pancreatitis and early discharge after an outpatient ERCP, various strategies have been suggested. One recent study proposed that values of serum amylase more than five times the upper limit of normal and serum elastase more than the upper limit of normal at 6h after the procedure are accurate for the prediction of post-ERCP pancreatitis. Serum amylase levels 4h after the ERCP is also considered a rapid and useful predictor of pancreatitis. Based on the values of serum amylase 4h after the ERCP, the following management protocol has been suggested: (i) when the serum amylase level is <1.5 times the upper limit of normal (negative predictive value 100%), the patient can be discharged; (ii) when the serum amylase level is >3.0 times the upper limit of normal (positive predictive value 37%), the patient should be admitted to the hospital; and (iii) when the serum amylase level is between 1.5 and 3.0 times the upper limit of normal, clinical assessment and risk factors should govern decisions on further management.
Although there is extensive literature on the prevention of post-ERCP pancreatitis, there are scant data on its management. Most cases are managed as for any other cause of acute pancreatitis. However, the study by Reddy et al. in this issue of Journal of Gastroenterology and Hepatology has demonstrated the need for protocol-based management of post-ERCP pancreatitis; such an approach is associated with less severe pancreatitis, shorter hospital stay, fewer imaging investigations, and less use of antibiotics. The authors retrospectively analyzed the efficacy of a standardized management protocol in 32 patients with post-ERCP pancreatitis. They compared the outcomes with the results of a non-protocol based management in 13 patients who underwent ERCP at the authors' center, but were managed by referring gastroenterologists or regional health-care facilities. The protocol included an observation period following ERCP, criteria-based admission and discharge policies, adequate hydration and analgesia, proper nutrition either by the enteral or parenteral route depending on bowel motility and the patient's clinical condition, and computerized tomography (CT) of the abdomen in patients with moderate or severe pancreatitis.
The baseline clinical presentation and severity of pancreatitis were similar in both groups. The median duration of hospitalization was significantly less in patients managed as per protocol in comparison to those in the non-protocol group (3 days vs 7 days, respectively; P=0.01). When compared to the patients in the non-protocol group, the patients treated by protocol were significantly less likely to develop moderate or severe pancreatitis (62% vs 12.5%, crude odds ratio [OR]=11.2; 95% confidence interval [CI]=1.9-69; P=0.002) and required a significantly lower number of CT examinations (100% vs 16%; P<0.001) and antibiotics (50% vs 3.1%; P=0.01). Also, patients who underwent pancreatic duct stenting appeared more likely to develop mild pancreatitis if treated per protocol (OR=2.6; 95% CI=0.2-147; P=0.61), and those managed outside protocol did not experience any benefit from pancreatic duct stenting (OR=0.2; 95% CI=0.01-4.0; P=0.29). After adjusting for age, comorbidity, endotherapy, and pancreatic duct stenting, a multivariate logistic regression analysis revealed that the protocol-based management strategy was significantly more often associated with less severe disease (adjusted OR=19; 95% CI=2.6-132; P=0.003).
Aggressive hydration, maintenance of electrolyte balance, nutritional support, and prevention of local and systemic complications are important for the management of acute pancreatitis irrespective of the etiology. The present study has demonstrated that this is also true for post-ERCP pancreatitis with management as per a standard protocol; there is less likelihood of severe pancreatitis developing, thereby reducing the hospital stay. However, it would be interesting to know how patients in the non-protocol group were managed on a day-to-day basis with regards to rates of hydration, nutritional support, and involvement of gastroenterologists in their treatment; this would allow us to identify factors in the management of pancreatitis that led to differences in outcome between the two groups. The answer to this question may come from prospective, randomized studies between the two treatment modalities.
In conclusion, acute pancreatitis is a dreaded complication of ERCP, and while the hunt for the ideal drug for the prophylaxis of post-ERCP pancreatitis is on, a combined strategy of identifying high-risk patients, adopting meticulous endoscopic techniques, and inserting a prophylactic pancreatic duct stent in high-risk patients can reduce this. Once post-ERCP pancreatitis develops, management as per a standard protocol, such as that prescribed by Reddy and colleagues, may be helpful. However, more data on this protocol is needed for the management of post-ERCP pancreatitis.
Acute pancreatitis is a common and dreaded complication of endoscopic retrograde cholangiopancreatography (ERCP) that continues to disturb endoscopists since the inception of this technique. Post-ERCP pancreatitis has been reported in up to 40% of patients, depending on the criteria used for diagnosis, differences in patient populations, endoscopic techniques used, and accuracy of the follow up. A recent survey of prospective studies from 1977 to 2006 involving 16855 patients, reported that post-ERCP pancreatitis occurred in 3.47% of patients. However, in high-risk conditions, such as sphincter of Oddi dysfunction, the incidence of acute pancreatitis can be as high as 31%. Even though post-ERCP pancreatitis is usually a mild disease with a favorable outcome in the majority of cases, it can cause significant discomfort to the patient, increase treatment costs, and be the focus the attention (and tension) of the treating endoscopist until the patient recovers completely.
The pathogenesis of post-ERCP pancreatitis is not completely understood. It seems to be an inflammatory response to mechanical, chemical, thermal, hydrostatic, enzymatic, allergic, and microbiological insults that result from papillary instrumentation and/or injection of contrast media for imaging of the pancreatic duct. These inciting factors lead to premature intracellular activation of pancreatic proteolytic enzymes, autodigestion, and release of inflammatory cytokines that produce both local and systemic effects.
There have been several efforts to develop an effective preventive strategy that will eliminate or reduce the risk or severity of post-ERCP pancreatitis. Most of the pharmacological agents tested are not universally effective, hence a combined strategy is being advocated to minimize the occurrence of post-ERCP pancreatitis. This includes the identification of patient- and procedure-related factors associated with high risk, refinement of endoscopic methods, search for an effective pharmacological agent for prophylaxis, and prophylactic placement of a pancreatic duct stent in high-risk patients.
A number of prospective studies have evaluated patient-related risk factors for post-ERCP pancreatitis. These have found that patients under 50 years of age, female sex, normal bilirubin, suspected sphincter of Oddi dysfunction, small bile duct diameter, past history of post-ERCP pancreatitis, and absence of chronic pancreatitis are associated with high risk of post-ERCP pancreatitis. Extra caution and appropriate prophylactic measures should be adopted in these patients. These factors can act synergistically, thereby putting patients with multiple risk factors at high risk for post-ERCP pancreatitis.
Just as there are patient-related risk factors, various procedures have also been associated with higher risk of post-ERCP pancreatitis, as documented by meta-analysis and multivariate analyses. These include ≥1 injection of contrast into the pancreatic duct, pancreatic sphincterotomy, balloon dilatation of an intact biliary sphincter, difficult or failed cannulation, failed clearance of bile duct stones, and precut sphincterotomy. Other factors, such as pancreatic acinarization, pancreatic brush cytology, pain during ERCP, low ERCP volume, and involvement of trainees during ERCP can also be associated with an increased risk of post-ERCP pancreatitis.
A proper understanding of the patient- and procedure-related risk factors has led to a strategy of carefully selecting patients for ERCP, for example, avoiding diagnostic ERCP, using magnetic resonance cholangiopancreatography and endoscopic ultrasound (EUS) for diagnostic purposes, and the use of meticulous endoscopic techniques. The latter include atraumatic manipulation of the papilla, avoidance of pancreatic duct injection for biliary cases, avoidance of acinarization, using an aspirating catheter technique or solid-state manometry system for suspected sphincter of Oddi dysfunction, limited use of precut sphincterotomy, biliary sphincterotomy from 11.00 to 1.00 o'clock position avoiding the pancreatic sphincter, using pure cut electrosurgical current or adjustment of automated generator to lower tissue effect during pancreatic sphincterotomy, avoidance of balloon dilatation of intact biliary sphincter, avoiding placement of biliary stent through intact sphincter in patients with sphincter of Oddi dysfunction, and using soft-tipped guide wires to access bile ducts.
Despite selecting proper patients and using meticulous techniques, post-ERCP pancreatitis occurs when it is least anticipated. Therefore, various pharmacological interventions and short-term placement of a pancreatic duct stent have also been evaluated for its prevention. Drugs that either affect the sphincter pressure or contractility and pancreatic secretion or the inflammatory cascade have been tested in various studies, but the results have been conflicting. For the prevention of post-ERCP pancreatitis: somatostatin, octreotide, gabexate mesylate, ulinastatin, allopurinol, corticosteroids, pentoxifylline, non-steroidal inflammatory drugs (NSAIDs), such as diclofenac and indomethacin, interleukin-10, glyceryl trinitrate, nifedipine, antibiotics, botulinum toxin, locally sprayed lidocaine, N-acetylcysteine, heparin and its derivatives, and β-carotene have been tried in clinical studies, while in animal models, thalidomide, rosiglitazone (PPARγ agonist), and the intraductal administration of a natural killer-1 receptor antagonist have been tested. Recent studies on rectal NSAIDs, including one in the Journal of Gastroenterology and Hepatology, prolonged infusion of a high dose of octreotide and a bolus injection of somatostatin, as shown in a meta-analysis with encouraging results. Further studies and more data are needed before putting these agents to use in routine clinical practice.
Studies have shown that maintaining the flow of the pancreatic juice through the papilla by the placement of a short-term transpapillary narrow caliber pancreatic duct stent is effective in reducing the risk of post-ERCP pancreatitis. While evaluating the cost effectiveness of three strategies (no prophylactic stents,prophylactic stents in all patients, and prophylactic stents in high-risk patients), the selective placement of prophylactic pancreatic stents in high-risk patients was the most cost-effective plan. Placing a pancreatic stent requires considerable expertise, especially after a biliary procedure; a failed pancreatic duct stent insertion may be more harmful than not attempting to place it. Another fear of pancreatic stent placement is the development of stent-induced ductal and parenchymal changes mimicking chronic pancreatitis. This can be disastrous as the majority of these patients have normal pancreatic ducts before stenting. Furthermore, another endoscopy is required to remove the pancreatic stent. These potential problems of stent placement can be avoided by using softer and smaller diameter 3F or 4F stents with an unflanged inner end. The advantages of these stents are lower rates of post-ERCP pancreatitis, decreased frequency of stent-induced changes in the pancreatic duct, and a higher rate of spontaneous dislodgement of the stent obviating the need for repeat endoscopy. However, it should be kept in mind that deploying 3F or 4F stents requires thin (0.018 or 0.025 inches) guide wires and considerable expertise in pancreatic endotherapy. It would be interesting to follow the results of prophylactic stenting from centers with low ERCP load and with relatively less-experienced endoscopists.
For an early diagnosis of post-ERCP pancreatitis and early discharge after an outpatient ERCP, various strategies have been suggested. One recent study proposed that values of serum amylase more than five times the upper limit of normal and serum elastase more than the upper limit of normal at 6h after the procedure are accurate for the prediction of post-ERCP pancreatitis. Serum amylase levels 4h after the ERCP is also considered a rapid and useful predictor of pancreatitis. Based on the values of serum amylase 4h after the ERCP, the following management protocol has been suggested: (i) when the serum amylase level is <1.5 times the upper limit of normal (negative predictive value 100%), the patient can be discharged; (ii) when the serum amylase level is >3.0 times the upper limit of normal (positive predictive value 37%), the patient should be admitted to the hospital; and (iii) when the serum amylase level is between 1.5 and 3.0 times the upper limit of normal, clinical assessment and risk factors should govern decisions on further management.
Although there is extensive literature on the prevention of post-ERCP pancreatitis, there are scant data on its management. Most cases are managed as for any other cause of acute pancreatitis. However, the study by Reddy et al. in this issue of Journal of Gastroenterology and Hepatology has demonstrated the need for protocol-based management of post-ERCP pancreatitis; such an approach is associated with less severe pancreatitis, shorter hospital stay, fewer imaging investigations, and less use of antibiotics. The authors retrospectively analyzed the efficacy of a standardized management protocol in 32 patients with post-ERCP pancreatitis. They compared the outcomes with the results of a non-protocol based management in 13 patients who underwent ERCP at the authors' center, but were managed by referring gastroenterologists or regional health-care facilities. The protocol included an observation period following ERCP, criteria-based admission and discharge policies, adequate hydration and analgesia, proper nutrition either by the enteral or parenteral route depending on bowel motility and the patient's clinical condition, and computerized tomography (CT) of the abdomen in patients with moderate or severe pancreatitis.
The baseline clinical presentation and severity of pancreatitis were similar in both groups. The median duration of hospitalization was significantly less in patients managed as per protocol in comparison to those in the non-protocol group (3 days vs 7 days, respectively; P=0.01). When compared to the patients in the non-protocol group, the patients treated by protocol were significantly less likely to develop moderate or severe pancreatitis (62% vs 12.5%, crude odds ratio [OR]=11.2; 95% confidence interval [CI]=1.9-69; P=0.002) and required a significantly lower number of CT examinations (100% vs 16%; P<0.001) and antibiotics (50% vs 3.1%; P=0.01). Also, patients who underwent pancreatic duct stenting appeared more likely to develop mild pancreatitis if treated per protocol (OR=2.6; 95% CI=0.2-147; P=0.61), and those managed outside protocol did not experience any benefit from pancreatic duct stenting (OR=0.2; 95% CI=0.01-4.0; P=0.29). After adjusting for age, comorbidity, endotherapy, and pancreatic duct stenting, a multivariate logistic regression analysis revealed that the protocol-based management strategy was significantly more often associated with less severe disease (adjusted OR=19; 95% CI=2.6-132; P=0.003).
Aggressive hydration, maintenance of electrolyte balance, nutritional support, and prevention of local and systemic complications are important for the management of acute pancreatitis irrespective of the etiology. The present study has demonstrated that this is also true for post-ERCP pancreatitis with management as per a standard protocol; there is less likelihood of severe pancreatitis developing, thereby reducing the hospital stay. However, it would be interesting to know how patients in the non-protocol group were managed on a day-to-day basis with regards to rates of hydration, nutritional support, and involvement of gastroenterologists in their treatment; this would allow us to identify factors in the management of pancreatitis that led to differences in outcome between the two groups. The answer to this question may come from prospective, randomized studies between the two treatment modalities.
In conclusion, acute pancreatitis is a dreaded complication of ERCP, and while the hunt for the ideal drug for the prophylaxis of post-ERCP pancreatitis is on, a combined strategy of identifying high-risk patients, adopting meticulous endoscopic techniques, and inserting a prophylactic pancreatic duct stent in high-risk patients can reduce this. Once post-ERCP pancreatitis develops, management as per a standard protocol, such as that prescribed by Reddy and colleagues, may be helpful. However, more data on this protocol is needed for the management of post-ERCP pancreatitis.
Source...