Article Text

Download PDFPDF

Aetiology of extrahepatic portal vein thrombosis
  1. B W M Spanier1,
  2. J Frederiks1
  1. 1Department of Internal Medicine, Spaarne Ziekenhuis Heemstede, Händellaan 2, 2102 CW Heemstede, the Netherlands
  1. Correspondence to:
    B W M Spanier;
  1. H L A Janssen2
  1. 2Department of Gastroenterology and Hepatology, University Hospital Rotterdam, Dr Molewaterplein 40, 3015 GA Rotterdam, the Netherlands; devlaming{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

I read with interest the article by Janssen et al (Gut 2001;49:720–4) regarding the aetiology and determinants of survival of extrahepatic portal vein thrombosis (EPVT). Among others, this retrospective study investigated the systemic risk factors for EPVT. We would like to add hyperhomocysteinaemia as a relatively new, not yet mentioned, risk factor.1–3 This is illustrated by a recent case in our clinic.

A 54 year old woman was hospitalised because of worsening of upper abdominal pain which started two weeks before admission and was continuously present. Her personal and family history for thromboembolic processes was uneventful and she did not smoke. She did not use vitamin supplements. One day prior to admission she was using the progesterone norethisterone (Primolut N) because of vaginal blood loss.

Combined portal-splenic vein thrombosis was diagnosed using colour Doppler ultrasonography and computed tomography. There were already some venous collaterals in the hilar area of the liver; hence the thrombosis would have been present for at least several weeks.

After investigations for thrombophilias were carried out, intravenous heparin and oral anticoagulant therapy were started and the patient improved.

The patient was found to be heterozygous for the prothrombin gene G20210A mutation and for the methyltetrahydrofolate reductase mutation. Mild hyperhomocysteinaemia (fasting/six hour post methionine load values 18/91 μmol/l) was also detected. Plasma vitamin B6, B12, and folate levels were normal. The patient is currently on lifelong oral anticoagulation therapy and has not yet started to use vitamin supplements.

Mild hyperhomocysteinaemia is a hypercoagulability risk factor for the development of EPVT which, at the time of the study of Janssen et al, had not been recognised as a prothrombotic factor. The association of hyperhomocysteinaemia and prothrombin gene mutation in EPVT has been documented only once in the literature. It is still unclear if the association is additive or synergistic for the development of thrombosis.1 We would like to recommend determination of homocysteine levels in patients with idiopathic EPVT as vitamin B6 and folate supplementation is a cheap and safe therapy in preventing deleterious vascular complications.


Author’s reply

Spanier and Frederiks describe the role of diagnosing hyperhomocysteinaemia in a patient with portal vein thrombosis. Their case not only illustrates the potential importance of hyperhomocysteinaemia but also the concurrence of multiple risk factors in portal vein thrombosis.1,2 Two inherited thrombotic risk factors (methylenetetrahydrofolate and prothrombin gene G20210A mutation) predisposed the patient to a thrombotic event which became clinically manifest after the recent start of progesterone.

There is clear evidence that hyperhomocysteinaemia is an independent risk factor for cardiovascular diseases. Hyperhomocysteinaemia can be diagnosed by genetic testing for the methylenetetrahydrofolate mutation and by measurement of increased plasma homocysteine levels, both fasting and after loading with methionine. The pathophysiological mechanism of homocysteine induced vascular disease is not well understood. It is even unclear whether homocysteine itself or a related metabolite or cofactor is primarily responsible for the thrombogenic effects of hyperhomocysteinaemia in vivo.

In our study on risk factors and determinants of survival for portal vein thrombosis, we did not investigate hyperhomocysteinaemia. As mentioned by Spanier and Frederiks, this relates primarily to the fact that in most of our patients hyperhomocysteinaemia was not recognised as an established thrombogenic risk factor at the time of diagnosis. Furthermore, many patients exhibited comorbidity or concurrent risk factors for portal vein thrombosis, which may lead to poor nutritional status and therefore interfere with plasma homocysteine measurement. These factors include the presence of liver cirrhosis, malignancy, and infections.3 Although testing for the point mutation (C677T) in the methylenetetrahydrofolate reductase gene avoids the problem of acquired hyperhomocysteinaemia, this genetic defect by itself does not appear to be a significant independent risk factor for atherothrombotic disease.4 From a therapeutic viewpoint, testing for hyperhomocysteinaemia is interesting, also for portal vein thrombosis, because vitamin supplementation (with folic acid, pyridoxine, and vitamin B12) is generally effective in reducing homocysteine concentrations. However, it is not known if this therapy confers a risk for either extensive splanchnic thrombosis or other thrombotic manifestations which can develop in patients with portal vein thrombosis.5 Ongoing prospective controlled trials are investigating the potential beneficial effect of homocysteine lowering treatment on cardiovascular morbidity and mortality in subjects with hyperhomocysteinaemia. Before advocating widespread screening for hyperhomocysteinaemia in patients with portal vein thrombosis, it would be preferable to have a better understanding of the clinical efficacy of these therapeutic interventions.