Summary

A young female patient taking oral contraceptive pills presented with acute abdominal pain. She was diagnosed to have superior mesenteric vein thrombosis. Thrombophilia screening was performed, and she was managed with anticoagulant therapy. This case illustrates that family physician should have a high index of suspicion towards rare but important conditions. Meta - analyses and cost effectiveness analyses suggest selective thrombophilia screening in patients with venothrombotic diseases. Based on current evidence, oral contraceptive pills are contraindicated in patients with prothrombotic diseases, while evidence-based clinical practice

guidelines provide guidance on chemoprophylaxis for patients with established venothrombotic disease.

Keywords: Superior mesenteric vein thrombosis, venothrombotic, thrombophilia, oral contraceptive pills, chemoprophylaxis

摘要

一名服用避孕藥的年青女子因急性肚痛就醫，被診斷為上腸繫膜靜脈栓塞。經血栓形成傾向篩查後，接受抗凝血治療。 這個例子說明家庭醫生對罕有但是重要的疾病要有高度警惕性。綜合研究分析和成本效益分析都提議，應對有靜脈栓塞的 病人進行有選擇性的血栓形成傾向篩查。現有的驗證證明，不應向有血栓形成傾向的病人處方避孕藥，以驗證為基礎的臨床實踐指引，提供已確診靜脈栓塞的病人藥物預防的原則。

主要詞彙：上腸繫膜靜脈栓塞，靜脈血栓塞，血栓形成傾向，口服避孕藥，藥物預防

Case Report

A 33-year-old lady presented with a 1-day history of insidious nonspecific central abdominal discomfort. Examination did not reveal any abnormalities. She was treated symptomatically and was followed up. She presented again the next day with increased severity of abdominal pain and fever. In view of the severity of her symptoms without an explainable cause, she was admitted for further management. Urgent computed tomography (CT) of the abdomen revealed superior mesenteric vein (SMV) thrombosis. There was no clinical evidence of bowel infarction or complications such as peritonitis. Medical treatment with immediate unfractionated heparin followed by warfarin was commenced.

Preliminary assessments revealed normal vital signs, cardiovascular system, central nervous system and lower limbs, clinic blood pressure, liver and renal functions, prothrombin time, activated partial thromboplastin time, International Normalized Ratio ( INR) and protein concentration in urine . Risk factor screening was unremarkable (first episode of thrombosis, unprovoked, no family history, young age, nonsmoker, single, unremarkable obstetric history, no pregnancy, normal body build, no recent surgery/trauma/ immobilization, no past history or alarming symptom suggestive of malignancy). Investigations, including thrombophilia screening by functional assay such as antithrombin, protein C and protein S, activated protein C resistance ratio for factor V Leiden, Januskinase 2 (JAK2) mutation, Ham’s test, Lupus anticoagulant ratio, anti-cardiolipin immunoglobulin M and G, showed unremarkable results. Hence, the diagnosis of idiopathic superior mesenteric vein (SMV) thrombosis was made. The only possible cause was the patient’s use of an oral contraceptive pill containing drospirenone 3mg and ethinyloestradiol 30μg (Yasmin) over the last 6 months. The patient was recommended by a haematologist to stop taking the contraceptive pill and to continue warfarin prophylaxis for 3 to 6 months for the time being. Thrombophilia screening was performed on her family members, and the results were negative.

The patient was discharged and followed up with INR monitoring; the target INR was 2.0-3.0. Subsequent Oesophagogastroduodenoscopy and colonoscopy showed no oesophageal varices or bowel lesion; there was no evidence of portal hypertension, inflammatory bowel disease or tumour. Magnetic resonance (MR) angiography of abdominal aorta and MR venography of abdomen after 3 months of warfarin therapy revealed occlusion of the main trunk of SMV, and recannulization of some of its branches. Warfarin prophylaxis was continued, and sandostatin was prescribed to control the inflow of blood to the mesentery to prevent venous congestion in the patient’s gut. A follow up CT abdomen one year later revealed a resolution of the previously noted thrombosis in SMV, misty mesentery and mesenteric adenopathy. Warfarin was stopped after discussion regarding the pros and cons of continuing this therapy as prophylaxis. General preventive measures on hypercoagulability and haemostasis were recommended . Followup was suggested with particular emphasis on the possible alarming symptoms of recurrent thrombosis. Consultation with a haematologist was suggested in case of high-risk events such as pregnancy or surgery (heparin prophylaxis might have to be considered if not contraindicated).

Discussion

Missed diagnoses of acute abdominal pain are not uncommon in the frontline primary care settings. As a rare differential diagnosis, SMV thrombosis poses a clinical challenge to physicians. Since CT scan is the only diagnostic modality, a high index of suspicion is needed when patients present with unexplained acute abdominal pain, especially if they are at high risk of thrombophilia. In this particular case, it was fortunate that the diagnosis was established early and anticoagulant therapy was commenced immediately.

There are several questions that the author would like to address through literature review, namely:

1. How common is mesenteric vein thrombosis? What are the characteristics and risk factors?
2. What is the evidence for the cost-effectiveness of thrombophilia screening? When is it indicated?
3. Is the use of oral contraceptive pills associated with mesenteric vein thrombosis?
4. What is the rationale of chemoprophylaxis?

Literature review

Background

Mesenteric vein thrombosis (MVT ) is an uncommon disease which can often be lethal. It was first described by Elliot¹ in 1895 and recognized as adistinct clinical entity in 1935 by Warren  and Eberhard.²

MVT accounts for approximately 10-15% of all cases of mesenteric ischaemia. The true prevalence is unknown, but the incidence of symptomatic MVT was reported to be 2 per 100,000 admissions over two decades.³ While local data is unavailable, the incidence appears to be increasing with improvements in imaging techniques that enable early diagnosis.

Causes

Bradbury et al ⁴ classified MVT as primary (idiopathic) or secondary according to the cause of the disease. Primary MVT accounted for 25-55% of cases in early studies,^5,6 although recent reports showed a decline in primary MVT because of improved diagnosis of hypercoagulable states.⁷

Nowadays, an ætiological factor can be identified in almost three quarters of cases; these patients are said to have secondary MVT. The most common ætiological factors are prothrombotic states due to hereditary or acquired conditions of coagulation or malignancy, intra-abdominal inflammatory disorders, postoperative state, liver cirrhosis and portal hypertension. Oral contraceptive use accounts for 9-18% of cases.^7,8 Other contributing conditions are shown in the Table below.

Clinical Presentation

MVT can be acute, subacute or chronic. Patients usually present with central abdominal colicky pain. Other symptoms include nausea, vomiting, diarrhoea, anorexia or haematemesis. Melaena occurs in about 15% of cases, but occult blood is detectable in nearly 50% of patients.³ Fever, tachycardia, abdominal tenderness, decreased bowel sounds and abdominal distension are common physical signs. Rarely, patients may present with complications such as peritonitis or bowel infarction. Basic laboratory tests are not helpful in diagnosis, although leucocytosis, metabolic acidosis and raised lactate dehydrogenase level can be a sign of established peritonitis or bowel infarction. Hence, diagnosis of MVT is often difficult and delayed.

To the author’s knowledge, the prevalence of MVT in patients with acute abdominal pain is unknown. In clinical practice, it is difficult to differentiate MVT from other causes of acute abdominal pain based on history alone. When a patient presents with abdominal pain, the decision to investigate is usually based on the clinician’s discretion unless the cause or diagnosis is obvious. A high index of suspicion should be maintained when clinicians encounter patients with severe abdominal pain but relatively unremarkable physical findings, particularly in those with risk factors of thrombosis. The diagnosis of abdominal pain is sometimes difficult in a primary care setting where imaging modalities such as CT and magnetic resonance imaging (MRI) are not readily available. Referral to secondary care is recommended when there is an urgent surgical or medical condition, when the diagnosis is uncertain, or when specialist treatment is warranted.

Investigations

Conventional contrast-enhanced CT scan is currently the investigation of choice for suspected MVT. Although CT can establish the diagnosis in 90% of patients^7,9,10, it is less accurate in patients with early thrombosis in the small mesenteric vessels. Plain abdominal X-ray often shows nonspecific ileus with dilated fluid-filled loops of bowel. Thumb-printing or air in the bowel wall or portal system may indicate bowel ischaemia or infarction. Ultrasound can help exclude biliary disease or detect free peritoneal fluid, but its diagnostic value may be hampered by the presence of bowel gas. MRI has excellent sensitivity and specificity,¹¹ but it is expensive and not always readily available. Selective mesenteric angiography may establish a definitive diagnosis and can differentiate venous from arterial ischaemia. However, it is rarely indicated unless non-invasive tests fail to establish the diagnosis, or in selective patients planning for intraarterial infusion of vasodilators or thrombolytic agents.¹² Abdominal paracentesis is sometimes helpful when patients with MVT have serosanguineous ascites.

Screening for thrombophilia

Once the diagnosis of MVT is made, patients should be screened for here ditaryor acquired thrombophilia with functional assays for protein C and S deficiency, factor V Leiden and other mutations, lupus anticoagulant , and paroxysmal nocturnal haemoglobinuria, etc. As the levels of antithrombin and protein C and S may be reduced during acute thrombosis and affected by heparin and warfarin therapy, immediate management is not dependent on these results. Rather, it is recommended to check their levels at least 2 weeks after completion of warfarin therapy.

Venous thrombosis is an extremely common condition, and extensive investigations should only be undertaken in patients with features of prothrombotic tendency. This is important not only because of financial consideration, but also because false positive results are possible. For instance, up to 3% of a normal population may have laboratory evidence of protein C deficiency, but these individuals are unlikely to have significant thrombophilia.¹³

There are currently no randomized controlled trials or controlled clinical trials to assess the benefit of testing for thrombophilia on the risk of recurrent thromboembolism.¹⁴ A meta-analysis and cost-effectiveness analysis¹⁵ suggested that selective thrombophilia screening based on previous personal and/or family history of venous thromboembolism (VTE) was more effective than universal screening in all patient groups evaluated. Another systemic review and cost-effectiveness analysis¹⁶ concluded that in terms of determining the duration of anticoagulation management, scenarios were found in which the cost per quality-adjusted life year of thrombophilia testing was just below £20,000.

However, these results are subject to great uncertainty largely because of the lack of knowledge on the increased risk of recurrence with each type of thrombophilia. Results are influenced by the fact that men have a greater risk of recurrence than women, and by the fact that the frequency of adverse events associated with warfarin treatment increases with age. Further research, for example on the likely sensitivity and specificity of the tests for specific types of thrombophilia, is needed to reduce the uncertainty associated with these results. Studies comparing patients with VTE tested for thrombophilia with those whose risk assessment was based on personal and family history of thrombosis would also be beneficial.

There is no consensus on whether thrombophilia screening should be recommended or not from current literature.^17-20 Based on guidelines of the British Committee for Standards in Haematology,²¹ it is worthwhile to screen patients in this case for thrombophilia because of younger age, first episode of unprovoked disease and unusual site of thrombosis. This is further supported by local studies showing a prevalence rate of hereditary thrombophilia of 58%,^22-23 and an incidence of 53.2%.²⁴ As such, screening is justified to identify at-risk patients for appropriate prophylactic anticoagulation therapy and advice aimed at preventing recurrent thromboembolism, although the test is expensive.²³

Oral contraceptive pills and venothrombotic disease

Another research question arising from this case report is the association between the oral contraceptive pill, containing drospirenone 3mg and ethinyloestradiol 30μg, and venothrombotic disease. Mesenteric venous occlusion in adults associated with oral contraceptive use was first described in 1963 by Reed and Coon.²⁵ Oral contraceptive pill use accounted for 9-18% of episodes of MVT in young women.²⁶ In the literature, clinical presentations of MVT in women who took oral contraceptive pills ranged from acute reversible ischaemic colitis to bowel infarction.²⁷ In 1982, a local prospective study²⁸ on thirty-two Chinese and seven Caucasian patients was performed to examine the effects of oral contraceptive pills on certain coagulation and fibrinolytic parameters. Results of the 12-month study showed that alpha 2PI level was unchanged in the Chinese patients, but a significant increase occurred in the Caucasian patients. Enhanced fibrinolytic response to venous occlusion was demonstrated in the Chinese but not in the Caucasian patients. A paper published in BMJ in 2003²⁹ reported a few cases of thromboembolism as a suspected adverse drug reaction to the new oral contraceptive pill containing drospirenone 3mg and ethinyloestradiol 30μg. The authors concluded that the association of the new oral contraceptive pill with a lower risk of thromboembolism had not been proven by research. There are many case reports on thromboembolism and new-generation oral contraceptive pills, suggesting that further studies on the safety profile of the pill containing drospirenone 3mg and ethinyloestradiol 30μg are warranted with regard to venothrombotic events.^30,31 Hence, history of oral contraceptive use is mandatory when physicians encounter unexplained abdominal pain, especially when MVT is suspected. In patients with prothrombotic risk, use of oral contraceptive pills is contraindicated.

Treatment

Treatment of MVT involves anticoagulation alone or in combination with surgery. Immediate anticoagulation with either unfractionated heparin or low-molecular-weight heparin early in the course of the disease, even intraoperatively, clearly improves survival and significantly reduces the risk of recurrence.⁸ Systemic heparin therapy is initiated with bolus injection of 5000IU, followed by a continuous infusion for at least 5 days or until an INR of 2.0-3.0 is reached. Warfarin can be started within 24 hours of heparin therapy. In the absence of an ongoing thrombotic disorder, the duration of anticoagulation may be limited to 6-12 months, as in this patient. The duration of prophylactic long-term anticoagulation depends on the level of prothrombotic risk of the patient. Evidencebased clinical practice guidelines from the American College of Chest Physicians (CHEST) (8^th Edition)³² provide guidance on the proper prevention of VTE. It also stated that there was strong evidence on the favourable benefit-to-risk ratio and cost-effectiveness of appropriately used thromboprophylaxis. In addition to conventional anticoagulants, new oral anticoagulants such as direct thrombin inhibitors and factor Xa inhibitors are emerging. These agents are subjects of active research as alternative agents for oral anticoagulation, and have been recently approved for prophylaxis in Canada and the European Union.³³ For this patient, her thrombotic event had resolved 1 year after warfarin therapy, as demonstrated by serial CT scans. Follow-up functional assay confirmed that she had no inherited prothrombotic risk factor except her previous use of an oral contraceptive pill containing drospirenone 3mg and ethinyloestradiol 30μg. The only concern is for future pregnancy, puerperium and surgery, when prophylactic anticoagulation will have to be discussed with the patient.

Concerning other management modalities, supportive measures include adequate analgesic, fluid, electrolyte and acid base monitoring and nasogastric tube. Broad – spectrum antibiotics are probably reasonable, especially if there is bowel infarction or peritonitis. For early presentation without peritonitis or infarction, the use of broad-spectrum antibiotics is controversial and is neither supported nor refuted by good evidence.³⁴ Lytic therapy with urokinase, streptokinase or tissue plasminogen activator has been found beneficial in some cases. Sandostatin is a somatostatin analogue that causes vasoconstriction on splanchnic blood flow. As MVT leads to venous congestion in the gut in this patient, sandostatin was prescribed to control the inflow of blood to the mesentery. Had the venous congestion led to bowel infarction, surgical resection would have to be unavoidable. Surgical intervention is usually indicated in patients with bowel infarction or peritonitis. The aim of resection is to conserve as much bowel as possible. Follow-up (second-look) laparotomy 24 hours later is to avoid resecting viable bowel.

Outcome and Prognosis

Acute MVT is associated with a mortality rate of 20% to 50%.^8,26,35 With recent advances in imaging techniques making early diagnosis possible, a trend of falling mortality has been observed in some studies.^36-38 Survival of patients with acute MVT depends on multiple factors including age, coexisting conditions, timing of diagnosis and intervention, and has improved over the last four decades. Morbidity is usually related to postoperative complications such as sepsis, wound infection and short gut syndrome. Long-term survival depends on the underlying reason for the thrombosis. Despite the falling mortality, MVT usually has a high recurrence rate. Anticoagulant therapy combined with surgery is associated with the lowest recurrence rate of about 3-5%.35 Among all aetiologies of mesenteric ischaemia , venous thrombosis carries the best prognosis.³⁹

Conclusion

Family physicians should maintain a high index of suspicion towards MVT when patients present with abdominal pain that is disproportionate to the physical signs together with negative initial workup for the common explainable causes of abdominal pain, especially for patients at high prothrombotic risk. Advances in imaging techniques have enabled early diagnosis of MVT. Immediate use of anticoagulant therapy helps to reduce mortality and morbidity. Surgical management is indicated in the event of complications such as peritonitis and infarction. In patients with possible inherited causes, selective thrombophilia screening and lifelong anticoagulation is warranted. Oral contraceptive pills remain an important cause of venothrombotic disease, in which case at least 6-12 months of anticoagulation is recommended. It is important for family physicians to coordinate comprehensive care when managing MVT by timely and optimal interdisciplinary collaboration with radiologists, haematologists, gastroenterologists and surgical specialists.

---

---

References