Abstract
Introduction
Bleeding stomal varices after ileal conduit urinary diversion are rare, but they can develop in patients with portal venous hypertension caused by cirrhosis.
Case presentation
We report the case of a 68-year-old man who developed stomal haemorrhage two months after radical cystectomy and ileal conduit urinary diversion to treat invasive bladder cancer. Alcoholic cirrhosis and portal venous hypertension were considered to be the causes of varices and bleeding. We chose to control the stomal varices using sclerotherapy. The stomal varices disappeared and no bleeding recurred during one year of follow up.
Conclusion
We believe that sclerotherapy is a suitable treatment for bleeding stomal varices.
Introduction
Ileal conduit urinary diversion is a simple, safe procedure that is commonly performed after radical cystectomy. 1 Postoperative complications of this surgery include parastomal hernia, bowel obstruction, stomal atrophy and necrosis, internal hernia, peristomal dermatitis, 2 and, very rarely, ileal stoma variceal haemorrhage.
Bleeding stomal varices after ileal conduit urinary diversion are predominantly found in patients with cirrhosis or liver metastases, which result in portal hypertension. 3 The pathogenesis of this disease is related to the digestive varices caused by portal hypertension and systemic haemodynamic disorders caused by cirrhosis. In addition, ileal blood circulation near the portal area is more likely to be affected by portal pressure. Thus, ileal stomal varices can occur in patients with mild or moderate portal hypertension but with no obvious symptom or sign of cirrhosis. 4 Poor blood coagulation due to impaired liver function and long-term stimulation by ileal bladder urine also increase the likelihood of variceal bleeding.
Here, and with the patient’s consent, we present a case of rare ileal stoma variceal haemorrhage and explore the clinical symptoms and suitable treatment of this condition.
Case report
A 68-year-old man presented at our hospital in July 2011 with a complaint of bleeding stomal varices after ileal conduit urinary diversion. The patient had undergone this procedure following radical cystectomy because of invasive bladder cancer two months previously. He had a long history of excessive daily alcohol consumption. His liver function and prothrombin time (PT) were normal before cystectomy. Abdominal ultrasound showed slight shrinkage of the liver and a normal portal vein diameter. We thus thought that the liver problem was much less serious than the invasive bladder cancer and decided to perform the surgery. We persuaded the patient to stop drinking in the perioperative period. The patient recovered well, but he resumed excessive drinking after he left the hospital, despite our advice.
Physical examination this time revealed varicose veins and caput Medusae on the ileostomy mucosa. The stomal varices ejected blood, and routine blood examination showed that the patient was anaemic (haemoglobin concentration, 71 g/L). A coagulation test indicated an extended PT. Biochemical examination revealed abnormal liver function. An abdominal ultrasound was performed and showed shrinkage of the liver, larger-than-normal portal vein diameter, splenomegaly, a widened splenic vein and peritoneal effusion. We thus considered the bleeding stomal varices to be caused by alcoholic cirrhosis.
We immediately gave the patient an intravenous transfusion of two units of blood. We dressed the bleeding location with gauze containing 1% epinephrine, but the effect of this treatment was not satisfactory.
Through discussion, we decided to perform sclerotherapy, which had not been used previously for stomal varices caused by ileal conduit urinary diversion. The patient provided informed consent. We checked the deep interior of the stoma and found no obvious varices. Thus, we injected the sclerosant lauromacrogol (polyoxyethylene lauryl ether) into the prominent varicose veins using combined intravenous and submucosal injection. We injected 2 ml sclerosant intravenously and 1 ml each at three submucosal injection sites, resulting in a total injected amount of 5 ml. This procedure stopped the bleeding. We repeated sclerosant injection one week later to consolidate the effect. The stomal varices disappeared with this treatment. The patient ultimately followed our recommendation and stopped drinking excessively, and no stomal bleeding recurred during one year of follow up.
Discussion
Bleeding stomal varices caused by ileal conduit urinary diversion are uncommon. Variceal rupture can cause repeated painless stomal bleeding. Continuous bleeding can lead to further deterioration of liver function, thereby initiating a vicious cycle. The amount of blood loss can be life threatening.
Visual examination is the most direct way to diagnose this disease, based on the presence of caput Medusae (due mainly to distended varicose veins) and ileal stomal bleeding. Diagnosis requires knowledge of whether the patient has a history of liver disease and portal hypertension. Endoscopy can be performed to determine whether varicose veins are present in the deep regions of the stoma. Varicose veins can also be detected in the venous phase of mesenteric angiography. 5
No standard treatment for bleeding stomal varices after ileal conduit urinary diversion has been established. Compression and ligation have been used. The use of percutaneous transhepatic puncture via the portal vein to access the varicose vein and perform embolism has also been reported. 6 In our experience, sclerotherapy is an effective, suitable, noninvasive treatment for this disease.
Our use of sclerotherapy for bleeding stomal varices was novel. The intravenous injection of vascular sclerosants can damage the vascular endothelium and promote the formation of blood clots to achieve vascular occlusion. Submucosal injection near the vein can cause fibrosis. Pressure can then be applied to the veins to reduce the blood flow rate, thereby achieving haemostasis.7,8 Through these mechanisms, sclerotherapy can stop bleeding effectively.
Lauromacrogol is a widely used sclerosant in clinical practice. When used for varicose veins, it can be injected intravenously or intravenously and submucosally. This course of treatment is simple and noninvasive.
Other available methods do not appear to effectively manage acute bleeding. The application of 1% epinephrine usually does not stop bleeding and re-bleeding is common. Varicose vein ligation can be used to control acute bleeding, but varicose veins may recur at other branches. In our experience, suturing stops bleeding only temporarily (5–7 days).
Another advantage of sclerotherapy is its noninvasiveness compared with some other treatments. 9 Invasive intervention for varicose veins has also been reported. Lashley et al. 6 reported the occurrence of bleeding stomal varices after ileal conduit urinary diversion in a patient with recurrent bleeding and slightly elevated portal pressure, which rendered the placement of a portosystemic shunt unsuitable. Percutaneous transhepatic puncture via the portal vein was ultimately used to access the varicose vein and perform intervention therapy. The authors reported satisfactory results of this approach. Thouveny et al. 10 reported on seven cases in which ultrasound-guided percutaneous puncture was used to directly access varicose veins, and embolisation was performed to treat colostomy-associated variceal bleeding after general surgery. This treatment was successful in six of seven cases, with two cases of recurrence. In addition to being invasive, such interventions require a highly skilled operator and advanced technology.
Conclusions
Firstly, local sclerosant injections can stop stomal variceal bleeding effectively and noninvasively, at a much lower cost compared with surgical intervention. Secondly, one must recognise that repeated large injections of sclerosants can cause stomal stenosis, intestinal ulcers and other complications. Thirdly, additional follow-up studies are required to determine the long-term effects of this treatment.
Footnotes
Author’s contributions
Shu-Tao Tan carried out the clinical reception, participated in making the treatment plan, and drafted the manuscript.
Acknowledgments
We would like to thank Professor Yun-Xiang Xiao of Peking University First Hospital for his assistance in treatment.
Declaration of conflicting interests
There were no conflicts of interest involved.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
