Splenic Injury Following Endoscopic Retrograde Cholangiopancreatography: A Case Report and Review of the Literature

Abstract

Endoscopy of both the upper and lower gastrointestinal tract remains one of the commonest procedures performed on patients. The common risk factors associated with this procedure are well known. However, some of the rarer complications are not known or are not considered in the immediate postprocedure time period when a patient experiences acute abdominal pain. Herein, we report a case of solid organ injury (splenic capsular tear) resulting from endoscopic retrograde cholangiopancreatography. The risk factors and presentation of this type of injury are reviewed, including its association with chronic pancreatitis, which we describe for the first time.

CASE REPORT

A 38-year-old male was admitted to the hospital under the gastroenterology service with an acute exacerbation of alcohol-induced chronic pancreatitis. A symptomatic pseudocyst had been percutaneously drained 4 months prior to admission. Despite this, the patient complained of ongoing left upper quadrant abdominal discomfort with associated nausea. During the investigation of this abdominal pain an endoscopic retrograde cholangiogram was performed. This study was technically difficult; a pancreatogram could not be obtained; imaging of the bile duct revealed a slight narrowing of the distal common bile duct.

In the time immediately following the endoscopic retrograde cholangiogram, the patient complained of worsening left upper quadrant abdominal discomfort with associated left shoulder tip pain. At this time, the surgery department was consulted. Examination revealed a patient in moderate distress, with a tachycardia of 120 beats per minute, blood pressure of 120/80 mm Hg with no postural changes, and afebrile. The patient was alert and oriented. The physical examination revealed vague tenderness in the left upper quadrant of the abdomen. Laboratory investigations at this time were noncontributory. Multiple views of the abdomen revealed residual contrast in the biliary tree, but no free air. Serial physical examinations were performed, and a computed tomography (CT) examination of the abdomen and pelvis was performed with intravenous and oral contrast. No free air or paraduodenal retroperitoneal air was seen. A small pseudocyst was evident in the tail of the pancreas. A large amount of free fluid in the peritoneal cavity was observed surrounding both the liver and spleen. No evidence of solid organ splenic or liver injury was identified on the CT scan.

Examination after the CT scan was performed revealed increasing abdominal pain, with signs of peritonitis. The patient was resuscitated and taken for emergent laparotomy. At celiotomy the patient was found to have 2000 cc of fresh and old (clotted) blood throughout the peritoneal cavity. Upper abdominal packing was performed as per trauma laparotomy techniques to obtain hemostasis. Complete laparotomy revealed that no hollow viscus injury was present. Kocherization and division of the gastrocolic ligament allowed complete visualization of the pancreas. The gland was hard in keeping with chronic pancreatitic changes, with a small pseudocyst located in the tail of the gland. Some dense adhesions were present between the tail of the pancreas, the stomach, and the spleen as well as the transverse colon and splenic flexure. It appeared that these adhesions were the result of the chronic pancreatitis. No injuries were identified.

A major splenic capsular tear/disruption was identified as the cause of the hemoperitoneum, and in fact was still bleeding vigorously. The spleen was of normal size and consistency, with no evidence of splenomegaly or splenic vein thrombosis. Splenorrhaphy was felt to be inappropriate given the amount of ongoing bleeding, as well as some transient hemodynamic instability the patient had experienced. A splenectomy was performed. Postoperatively, the patient received appropriate vaccinations and recovered uneventfully.

DISCUSSION

Injuries to the spleen following colonoscopy and endoscopy are rare; 2 large series prior to 1975 involving over 12 000 patients failed to report a single case of splenic rupture.^{1, 2} A recent review of splenic rupture following colonoscopy in 1997 revealed 18 cases to date,³ with 7 more cases being reported in the literature since then, bringing the total to 25 cases.^4–9 Two prior reports of upper gastrointestinal endoscopy causing splenic rupture have been reported, and one case of liver hemorrhage.^10–12 Both reports of splenic injury followed endoscopic retrograde cholangiopancreatography.

Patients presenting with this rare injury usually do so within 24 hours; however, it may be delayed up to 3 days.^4–9 Symptoms often include localized peritoneal irritation in the left upper quadrant progressing with time to generalized peritonitis, along with referred pain to the left shoulder. Changes in vital signs often accompany the peritonitis, including tachycardia and orthostatic changes in blood pressure.^4–9

Prior reports emphasize 3 important predisposing factors to the development of splenic injury during endoscopy: excessive traction on the splenocolic or gastrosplenic ligaments, a decrease in the mobility between the spleen and the surrounding organs (most often resulting from adhesions), and finally direct trauma to the spleen.³ In this case, the presence of chronic pancreatitis along with an intervention for a prior pseudocyst caused both adhesions and a decrease in mobility between the stomach, pancreas, and the spleen as a result of fibrosis secondary to prior inflammation. This increases the risk of a traction injury to the spleen as a result of manipulating the endoscope in the stomach, causing the transmission of mechanical forces through the ligamentous attachments of the spleen.

Although uncommon, this type of injury is important for endoscopists to be aware of. Unfortunately, the injury is not often included in the differential diagnosis of postendoscopy pain. As in this case, the CT scan will often not show any direct evidence of splenic injury. The authors suggest that this is because the injury is usually a capsular avulsion, and therefore no parenchymal changes will be seen on CT. The CT will demonstrate free fluid, which will be interpreted as representing a perforated hollow viscus rather than a splenic injury. In this scenario, a diagnostic laparoscopy is a reasonable alternative to determine the nature of the injury. Management should be tailored to the patient's condition as well as the nature of the splenic injury. In patients exhibiting hemodynamic compromise, splenectomy is the safest course of action. However, if the patient is stable, and an isolated capsular/parenchymal tear involving the spleen is identified, nonoperative management as used in traumatic splenic injuries may be undertaken by an experienced clinician. In addition, splenorrhaphy may be considered at laparotomy in a stable patient with an isolated injury amenable to repair.

CONCLUSION

This case illustrates 2 important points. The first being that solid organ injury, especially involving the spleen, should be considered in the postendoscopy patient with increasing abdominal pain. And the second being that the presence of chronic pancreatitis is a risk factor for injury to the spleen during the course of an endoscopic examination.