Abstract
A 44-year-old man presented with acute onset painless haemorrhage from his ileostomy. He had undergone panproctocolectomy for ulcerative colitis 17 years previously. Immediately following admission, he experienced further massive haemorrhage, developing hypovolaemic shock. Active resuscitation including suturing of a bleeding stomal varix was successful. An ultrasound scan of the liver was reported to be normal and the patient was discharged for further outpatient investigation. However he reattended 8 days later with further significant haemorrhage. Investigation with CT revealed portal hypertension and cirrhosis. The patient is currently awaiting a transjugular intrahepatic portosystemic shunt (TIPS) procedure and transjugular biopsy to alleviate his portal hypertension and yield a tissue sample for histological diagnosis. The case highlights the need for clinicians to have a high index of suspicion when presented with bleeding stomas. Further investigation is warranted and may reveal significant underlying hepatic disease. Additionally, further procedures may be necessitated to prevent recurrence of haemorrhage.
BACKGROUND
In patients presenting with significant stomal haemorrhage, portal hypertension must be suspected and underlying cirrhosis actively excluded. If portal hypertension and chronic liver disease are identified, patients should be referred to a liver failure unit for management so as to prevent recurrence of potentially life-threatening bleeding. Clinicians need to have a high index of suspicion when dealing with this presentation.
CASE PRESENTATION
A 44-year-old man presented on the emergency intake with acute onset painless bleeding from his ileostomy. The stoma had been bleeding for 3–4 h prior to admission during which time he had changed his stoma bag three times, with an estimated blood loss of 1100 ml. He had experienced previous episodes of small volume intermittent bleeding over a 5-month period but had not sought medical attention. There was no associated abdominal pain or discomfort, and his bowel motions were normal with no visible evidence of blood.
The patient had undergone a panproctocolectomy for ulcerative colitis 17 years previously. Past medical history included type 2 diabetes and he was on appropriate oral medication for this condition. He also suffered from hypertension and was regularly taking a calcium channel blocker. There was no history of diabetes or hypercholesterolaemia. Of note, the patient admitted to consuming 35 to 45 units weekly for several years. He was not on any antiplatelet drugs or anticoagulation and there was no personal or family history of bleeding diathesis.
On examination, he was pale, with no icterus. His body mass index (BMI) was not significantly raised. There was borderline tachycardia with a pulse rate of 100 beats/min and the blood pressure was 127/71 mm Hg. Examination of the chest was normal, with no spider naevi. His abdomen had a mid-line laparotomy scar, was soft and no organomegaly was noted. There were no other stigmata of liver failure. His Ileostomy appeared healthy looking with a small amount of blood noted arising from a vein on the inferior aspect of the stoma. This bleeding stopped ceased with local application of pressure.
Initial management consisted of resuscitation with intravenous crystalloid. Laboratory investigations revealed a haemoglobin (Hb) level of 10.3 g/dl. The prothrombin time (PT) and activated partial thromboplastin time (APTT) were within the normal ranges. Liver function testing (LFT) revealed elevated aminotransferases and alkaline phosphatase without a raised bilirubin (table 1). Lipid profile was normal.
Table 1.
Liver function test results
| Liver function test | Value | Units | Reference range |
| Total protein | 65 | g/litre | 60–85 |
| Albumin | 38 | g/litre | 35–50 |
| Alkaline phosphatase | 128 | IU/litre | 5–40 |
| Aspartate aminotransferase | 110 | IU/litre | 10–40 |
| Alanine aminotransferase | 125 | IU/litre | 30–120 |
| γ Glutamyl transferase | 370 | IU/litre | 0–51 |
| Bilirubin | 9 | µmol/dl | 2–14 |
The patient was initially stable following fluid resuscitation. However, the following day he experienced a further large haemorrhage from the same site on his ileostomy. He was given aggressive resuscitation and bleeding was controlled by suturing the prominent bleeding vein. His repeat laboratory indices revealed a Hb of 6.4 g/dl and he was transfused with four units of packed red cells.
As his bleeding had been significant, with the past medical history of ulcerative colitis (and its association with primary sclerosing cholangitis), in addition to the high alcohol intake, an ultrasound scan of his abdomen was arranged. This was to look specifically for signs of cirrhosis and portal hypertension. The ultrasound was reported as normal, with no signs of portal hypertension or portal or splenic vein thrombosis. Following a 24-h period of observation he was discharged with a viewed for further investigations as an outpatient.
While awaiting investigation, the patient reattended 8 days later with further bleeding from his ileostomy. At the time of assessment in the emergency department, the bleeding had stopped. However it was apparent that there had been significant blood loss as his Hb had fallen to 6.3 g/dl. Following initial resuscitation and stabilisation, a further four units of packed red cells were transfused.
Due to the severity of the bleeding, a CT angiogram was performed. This showed portal vein dilatation consistent with portal hypertension, splenomegaly and mild cirrhosis (fig 1). There was no evidence of portal or splenic vein thrombosis. Subsequent outpatient magnetic resonance cholangiopancreatography (MRCP) reiterated the findings of abnormal liver texture, portal hypertension and splenomegaly, but was inconclusive of aetiology.
Figure 1.
Contrast enhanced axial CT scan showing dilated portal vein and splenomegaly.
INVESTIGATIONS
CT angiogram was performed. This showed portal vein dilatation consistent with portal hypertension, splenomegaly and mild liver cirrhosis (fig 1).
MRCP reiterated the findings of abnormal liver texture, portal hypertension and splenomegaly, inconclusive of aetiology.
DIFFERENTIAL DIAGNOSIS
The most likely differential diagnoses for the patient’s portal hypertension were primary sclerosing cholangitis (as an association with ulcerative colitis) and alcoholic liver disease
Other causes of cirrhosis and portal hypertension were also considered and evaluated.
TREATMENT
The patient was treated with resuscitation, local suturing and blood transfusions of four units on the first admission. Following his second admission, he was again resuscitated and transfused a further four units.
OUTCOME AND FOLLOW-UP
The patient has been referred to a specialist liver unit and currently awaiting a transjugular intrahepatic portosystemic shunt (TIPS) procedure to alleviate his portal hypertension and yield a tissue sample for histological diagnosis.
DISCUSSION
The true incidence of bleeding from a stoma is difficult to ascertain. Duchesne et al1 found that 2 out of 164 (1.2%) patients with a stoma experienced bleeding, from cases over a 3-year period. Park et al2 performed a large retrospective analysis of 1616 patients with a stoma, of which 428 had an end or loop ileostomy. They identified stomal haemorrhage as a complication in 0.74% of the overall group. Unfortunately the severity of bleeding or formation of varices was not documented in either study.
Primary sclerosing cholangitis has a strong correlation with ulcerative colitis. Saich and Chapman3 suggest around 5% of patients with ulcerative colitis have associated primary sclerosing cholangitis (derived from three studies from the UK, Norway and Sweden).4–6 A recent study in Argentina7 found an overall prevalence of primary sclerosing cholangitis associated with ulcerative colitis to be 2.9%, reaching 6.2% in extensive colonic disease. The higher rate in patients with extensive colitis is confirmed by a study in Finland,8 which found a prevalence of 11.8% of patients undergoing proctocolectomy for ulcerative colitis.
The performance of proctocolectomy for ulcerative colitis in the presence of documented primary sclerosing cholangitis is associated with a high rate of development of stomal varices. Wiesner et al9 investigated a series of 19 patients who had proctocolectomy for ulcerative colitis in the presence of primary sclerosing cholangitis and noted 10 of 19 patients developed peristomal varices, with 7 patients requiring blood transfusion for bleeding. In the same study, Wiesner et al9 also noted that none of four patients with liver disease that had ileoanal anastomosis suffered from bleeding, suggesting that anastomosis is preferable to ileostomy formation. This was followed by a larger study of 117 patients undergoing colectomy in the presence of liver disease by Fucini et al.10 They found that in the ileoanal anastomosis group of 55 patients, there were no documented cases of anastomotic varices or bleeding. Of 23 patients that bled from stomal varices, the ulcerative colitis/primary sclerosing cholangitis association accounted for 13 (57%) of these patients.
There has been a growing trend to ileal pouch anal anastomosis as the technique has evolved.11 This implies that the overall incidence of bleeding from ileostomy may fall over time. For patients such as ours this may be disadvantageous, as it was the presence of bleeding that alerted us to the presence of associated liver disease and allowed prompt referral to a tertiary liver unit.
During the resuscitation phase, local treatments are often used to decrease bleeding.12 Local measures include pressure, cautery, suture ligation, sclerotherapy and further local surgery (including stomal revision or resiting). Many authors10,12–14 have found that local measures fail to control recurrent bleeding; in these cases portal decompression is required due to underlying liver disease. Surgical shunts and TIPS have both been employed to relieve portal hypertension in bleeding stomal varices. In 1999, a case series reported that TIPS successfully control bleeding from ectopic varices,13 including stomal bleeding. Pharmacotherapy with β blockers has also been employed to control portal hypertension.14,15
In the largest systematic review to date, Speir et al14 examined 74 studies of 234 patients with bleeding stomal varices. They conclude that TIPS is the preferred treatment for portal decompression, with surgical shunts used in TIPS failures. Speir et al also recommend liver transplantation in the treatment of decompensated liver disease.
Three recent case series highlight the need for consideration of underlying liver disease in cases of bleeding from stomal varices. The problem of recurrent bleeding has been described by Farquharson et al15 in their presentation of two cases. Both patients had underlying liver disease causing portal hypertension and in one case this was documented prior to the bleeding. Kabeer et al16 described three cases of bleeding from stomas in the presence of liver disease. Although two patients were terminally ill, they note that recurrent episodes of significant bleeding might have been avoided had the patients had a TIPS procedure earlier. Spier et al14 noted that clinicians may not initially look for varices when faced with bleeding from a stoma, instead attributing bleeding to local irritation. Portal hypertension was not known in three out of eight cases that they reported, being found only on further investigation. Ultrasound is usually the first radiological investigation performed but may be non-contributory and cross-sectional imaging with CT or MRI may be necessary.17 Our case presentation demonstrates this point and emphasises the need for further investigation if there is clinical suspicion.
LEARNING POINTS
Bleeding stomal varices are a hallmark of underlying liver pathology: portal hypertension with or without cirrhosis.
Clinicians need to have a high index of suspicion and a low threshold for investigating and treating these patients.
Without appropriate treatment, there may be recurrence of bleeding which may be extensive and possibly life-threatening.
The indication for the initial stoma may be relevant to future treatment (eg, proctocolectomy for ulcerative colitis, with associated primary sclerosing cholangitis (PSC) causing portal hypertension).
Acknowledgments
The authors acknowledge and thank the patient for his consent and approval of this case report.
Footnotes
Competing interests: None.
Patient consent: Patient/guardian consent was obtained for publication.
REFERENCES
- 1.Duchesne JC, Wang YZ, Weintraub SL, et al. Stoma complications: a multivariate analysis. Am Surg 2002; 68: 961–6 [PubMed] [Google Scholar]
- 2.Park JJ, Del Pino A, Orsay CP, et al. Stoma complications: the Cook County Hospital experience. Dis Colon Rectum 1999; 42: 1575–80 [DOI] [PubMed] [Google Scholar]
- 3.Saich R, Chapman R. Primary sclerosing cholangitis, autoimmune hepatitis and overlap syndromes in inflammatory bowel disease. World J Gastroenterol 2008; 14: 331–7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Perrett AD, Higgins G, Johnston HH, et al. The liver in ulcerative colitis. Q J Med 1971; 40: 211–38 [DOI] [PubMed] [Google Scholar]
- 5.Schrumpf E, Elgjo K, Fausa O, et al. Sclerosing cholangitis in ulcerative colitis. Scand J Gastroenterol 1980; 15: 689–97 [DOI] [PubMed] [Google Scholar]
- 6.Olsson R, Danielsson A, Järnerot G, et al. Prevalence of primary sclerosing cholangitis in patients with ulcerative colitis. Gastroenterology 1991; 100: 1319–23 [PubMed] [Google Scholar]
- 7.Terg R, Sambuelli A, Coronel E, et al. Prevalence of primary sclerosing cholangitis in patients with ulcerative colitis and the risk of developing malignancies. A large prospective study. Acta Gastroenterol Latinoam 2008; 38: 26–33 [PubMed] [Google Scholar]
- 8.Lepistö A, Kärkkäinen P, Järvinen HJ. Prevalence of primary sclerosing cholangitis in ulcerative colitis patients undergoing proctocolectomy and ileal pouch-anal anastomosis. Inflamm Bowel Dis 2008; 14: 775–9 [DOI] [PubMed] [Google Scholar]
- 9.Wiesner RH, LaRusso NF, Dozois RR, et al. Peristomal varices after proctocolectomy in patients with primary sclerosing cholangitis. Gastroenterology 1986; 90: 316–22 [DOI] [PubMed] [Google Scholar]
- 10.Fucini C, Wolff BG, Dozois RR. Bleeding from peristomal varices: perspectives on prevention and treatment. Dis Colon Rectum 1991; 34: 1073–8 [DOI] [PubMed] [Google Scholar]
- 11.Bach SP, Mortensen NJ. Revolution and evolution: 30 years of ileoanal pouch surgery. Inflamm Bowel Dis 2006; 12: 131–45 [DOI] [PubMed] [Google Scholar]
- 12.Roberts PL, Martin FM, Schoetz DJ, Jr, et al. Bleeding stomal varices. The role of local treatment. Dis Colon Rectum 1990; 33: 547–9 [DOI] [PubMed] [Google Scholar]
- 13.Carrafiello G, Laganà D, Giorgianni A, et al. Bleeding from peristomal varices in a cirrhotic patient with ileal conduit: treatment with transjugular intrahepatic portocaval shunt (TIPS). Emerg Radiol 2007; 13: 341–3 [DOI] [PubMed] [Google Scholar]
- 14.Spier BJ, Fayyad AA, Lucey MR, et al. Bleeding stomal varices: case series and systematic review of the literature. Clin Gastroenterol Hepatol 2008; 6: 346–52 [DOI] [PubMed] [Google Scholar]
- 15.Farquharson AL, Bannister JJ, Yates SP. Peristomal varices – life threatening or luminal? Ann R Coll Surg Engl 2006; 88: 6–8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Kabeer MA, Jackson L, Widdison AL, et al. Stomal varices: a rare cause of stomal hemorrhage. A report of three cases. Ostomy Wound Manage 2007; 53: 20–26 [PubMed] [Google Scholar]
- 17.Handschin AE, Weber M, Weishaupt D, et al. Contrast-enhanced three-dimensional magnetic resonance angiography for visualization of ectopic varices. Dis Colon Rectum 2002; 45: 1541–4 [DOI] [PubMed] [Google Scholar]

