Patients with sinistral portal hypertension frequently form varices most commonly in the fundus of the stomach. This location is explained by the venous drainage of the spleen via the short gastric veins. In contrast, gastric and oesophageal varices are more commonly precipitated by hepatic sinusoidal hypertension, but these patients usually exhibit other stigma of chronic liver disease [2
] which are absent in presinusoidal portal hypertension. We herein describe a case of sinistral hypertension with recurrent upper gastrointestinal bleeding resulting from splenic venous thrombosis, as a late complication of splenic flexure colonic perforation. This complication has not been described previously to our knowledge.
Colonoscopy is an invaluable tool in assessing diseases of the rectum and colon; however, it is in an invasive procedure and not without complications. Perforation and bleeding occur in 0.12% of patients undergoing colonoscopy, and peritoneal abscesses have been reported [3
]. Such patients may benefit from prompt laparotomy with primary repair or limited resection with anastomosis to minimise morbidity and mortality [4
]. Conservative management may on occasion be entirely appropriate, but on rare occasion this course may lead to added risk of further complications, as illustrated by the case described.
Colonic perforation with development of a peritoneal abscess leading to splenic vein thrombosis is clearly a very rare iatrogenic cause of sinistral portal hypertension and variceal bleeding. More common causes of splenic vein thrombosis include pancreatic tumours, pancreatitis, pancreatic pseudocysts, splenic vein stenosis, and polycystic disease of the liver and pancreas [5
]. If the common risk factors for upper GI haemorrhage are absent, including NSAID use and cirrhosis, then OGD and imaging of the abdomen are required to rule out splenic vein thrombosis.
In conclusion, conservative management of an iatrogenic colonoscopic perforation at the splenic flexure led to late presentation of life-threatening sinistral portal hypertension with gastric fundal varices secondary to splenic vein thrombosis and sinistral portal hypertension. In cases of active bleeding, splenectomy is indicated and can be curative [1
]. We advocate more prompt operative intervention in cases of splenic flexure colonic perforation, where the perforation is localized to the lesser sac. Primary repair following iatrogenic gastrointestinal perforation may have prevented a late complication that was more challenging to manage from a surgical perspective.