Basic investigations, such as complete blood count, serum electrolytes, renal and liver function tests, coagulation studies, blood sugar levels, and chest radiographs were recorded for all 39 patients. Thrombophilia screen included protein S, protein C, antithrombin, paroxysmal nocturnal hemoglobinuria (flow cytometry with CD55, CD59, and fluorescein-labeled proaerolysin (FLAER) expression), and JAK2 mutation. Antiphospholipid antibody screen included anti-cardiolipin, lupus anticoagulant, and anti-beta2-glycoprotein antibodies. In addition to the confirmatory diagnostic imaging, complication such as PH and variceal bleeding were diagnosed by endoscopy and bowel ischemia by CT. Importantly, we also noted the dates that these investigations were done and compared this to the date of PVT diagnosis to ascertain its relationship to PVT.

Patients were categorized as cirrhotic and non-cirrhotic types, with a further subclassification of the non-cirrhotic type into acute and chronic PVT, depending on the image of cavernoma transformation (formation of collateral blood vessels) or evidence of PH including splenomegaly and esophageal varices, as features of chronic PVT.

Data was analyzed using SPSS Statistics (IBM Corp. Released 2013. IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM Corp.). Continuous data was expressed as mean and range. Qualitative data were expressed as percentages. The chi-square test was used to study the statistical significance of qualitative variables.

Results

A total of 39 PVT patients were enrolled in this study; 15 (38.4%) showed cirrhosis of liver and 24 (61.5%) were classified as non-cirrhotic. Non-cirrhotic PVT patients were further subdivided as acute or chronic; 15 (62.5%) had an acute non-cirrhotic PVT, and nine (37.5%) had chronic PVT. PVT was more common in males than females (25 (64.1%) vs. 14 (35.8%), respectively, p = 0.020).

The presenting symptoms were abdominal pain (64.1%), nausea (30.7%), fever (20.5%), gastrointestinal bleeding (hematemesis or melena) (17.9%), and weight loss (10.2%). Demonstrable ascites and splenomegaly were seen in 28.2% and 38.5% of patients, respectively.

Amongst the 18 tested for an underlying thrombophilia, four (40.0%) had protein C deficiency, five (50.0%) had protein S deficiency, and two (33.3%) had antithrombin deficiency. Antiphospholipid antibodies (anti-cardiolipin, lupus anticoagulant, and anti-beta2-glycoprotein antibodies) were found in four patients (25.0%), and V617F JAK2 mutation was positive in three patients (20.0%). Factor V Leiden mutation, prothrombin G20210A mutation, hyperhomocysteinemia, and paroxysmal nocturnal hemoglobinuria (tested in 7, 4, 2, and 1 patients, respectively) were all negative.

The majority of non-cirrhotic PVT patients were tested for thrombophilia whereas only a few cirrhotic patients were tested. Furthermore, many patients were identified with more than one risk factor that may have contributed to the PVT. Local factors were seen eight patients (20.5%) with PVT; six patients had cholangitis, and one patient each had cholecystitis and liver abscess, respectively.

Cirrhosis of liver alone was the only risk factor in 15 patients (38.5%). However, eight patients (53.0%) with cirrhosis had hepatocellular carcinoma (HCC), which is a known risk factor for PVT. Thrombophilia is not routinely tested in cirrhotic patients; however, four patients with cirrhosis were tested for thrombophilia and were found to show protein C and S deficiency, which was expected due to an underlying acquired metabolic synthetic deficiency.

Thirteen patients (33.3%) in this study had an underlying malignancy. Eight (61.5%) cases with HCC all were found in PVT patients with cirrhosis. Additionally, in five patients (38.4%), other malignancies were observed within the acute non-cirrhotic PVT group. Three patients had cholangiocarcinoma, and one patient each had rectosigmoid carcinoma and carcinoma of the head of the pancreas. Furthermore, in three PVT patients (7.6%), no identifiable risk factors were seen.

Diagnosis was confirmed by CT in 29 patients (74.4%). In 12 patients (31.0%) the diagnosis was first established by an ultrasound and later confirmed by CT, but in six patients (15.3%) only an ultrasound was performed. In 12 PVT patients, MRI was done to look for underlying conditions (namely malignancy or hepatic parenchymal details). Four patients had undergone just an MRI, as the diagnosis was already established in other regional hospitals.

Endoscopy was done routinely in patients with cirrhosis as part of their assessment. However, in non-cirrhotic patients, it was done to rule out gastrointestinal bleeding in cases with a history of melena or hematemesis. The most common complication seen was PH in 10 (66.6%) PVT patients with cirrhosis in contrast to three (33.3%) chronic non-cirrhotic PVT patients.

Other complications in PVT patients with cirrhosis were portal hypertensive gastropathy (n = 7, 46.7%), esophageal varices (n = 7, 46.7%), gastric varices (n = 3, 20.0%), and variceal hemorrhage (n = 5, 33.3%). Chronic non-cirrhotic PVT patients had esophageal varices (n = 4, 44.4%), variceal hemorrhage (n = 2, 22.2%), and gastric varices (n = 1, 11.1%). Two (13.3%) patients with acute non-cirrhotic PVT presented with intestinal ischemia.

Eighteen (46.1%) patients with PVT received immediate anticoagulation with heparin followed by warfarin. Of these, 16 (88.8%) were non-cirrhotic and two (11.1%) had cirrhosis. Thrombectomy was done in one patient with PVT because it was complicated with intestinal ischemia; however, it failed. The patient was started on anticoagulation therapy with heparin followed by warfarin. Management with sclerotherapy was performed in all patients with variceal bleeding (17.9%). Transjugular portosystemic shunting was performed successfully in one patient. Patients who had PH (mainly patients with chronic PVT and cirrhosis) also received additional supporting treatment with beta-blocker prophylaxis and variceal banding.

Seven patients (46.6%) with acute non-cirrhotic PVT had a full recovery. Six (15.3%) patients died. Incidentally, 26 patients (66.6%) had no details on follow-up as they were transferred to other local hospitals or were lost to follow-up.

Discussion

Abdominal pain was the most common manifestation seen in almost two-thirds of the patient cohort and is explained by the fact that PVT is an intra-abdominal event and is thus most likely to present with abdominal symptoms. Kocher and Himmelmann reported a similar observation in their cohort of 20 non-cirrhotic PVT patients.9 We observed that patients with non-cirrhotic PVT had more symptoms than cirrhotic PVT patients. Moreover, few non-cirrhotic PVT cases also presented with vomiting and watery diarrhea. Patients presenting with fever usually had intra-abdominal inflammatory processing like cholangitis. Expectedly, splenomegaly and ascites were seen more in cirrhotic patients with PVT. Detectable ascites by ultrasound was also seen in non-cirrhotic patients, which may be as a result or as a sequel of PH or bowel ischemia. Nevertheless, gastrointestinal bleeding (i.e., hematemesis or melena) was commonly seen with established PH and in chronic non-cirrhotic PVT.

Prothrombotic conditions, malignancy, and cirrhosis were the three main causative factors followed by local factors. Inherited prothrombotic disorders, acquired thrombophilic disorders, malignancy, local factors, and cirrhosis of the liver was 28.2%, 17.9%, 33.3%, 20.5%, and 38.4%, respectively. No identified cause was seen in three (7.6%) patients. This was in agreement with other studies of PVT in adults,10,11 where cirrhosis was generally the most common cause followed by neoplasia. In many cases more than one known causative factors were identified and reported by others; however, the leading risk factor for PVT in our study was cirrhosis of liver (including HCC).11,12

Cirrhosis is now believed to be a hypercoagulable state, and its presentation may vary from being asymptomatic to having life-threatening issues.8 Partial PVT is generally asymptomatic, whereas rapid, complete PVT presents typically with abdominal pain and non-bloody diarrhea.8 Hypercoagulability in cirrhosis is multifactorial with increased factor VIII: C levels along with a reduction in albumin and protein C levels often tilting the balance.13 Furthermore, bleeding risk appears to be higher in PVT patients with cirrhosis compared to patients with cirrhosis alone.14

Thrombophilia testing should not be performed during the period of acute thrombosis when these results can be falsely low. Further, these tests should be avoided if the patient is on active anticoagulation therapy. Thrombophilic conditions are usually identified in approximately 60% of patients and as an additional local predisposing factor in 30–40% of cases, with up to 80% cases demonstrating an underlying cause.3 However, a meta-analysis on thrombophilia in PVT showed a prevalence of antithrombin deficiency (3.9%), protein C deficiency (5.6%), and protein S deficiency (2.6%), with odds ratios of 8.89, 17.63, and 8.00, respectively.15 Myeloproliferative disorders (MPD) is also a common risk factor for PVT.3 A meta-analysis showed that the mean prevalence of MPD and JAK2 mutation was 31.5% (95% CI: 25.1–38.8%) and 27.7% (95% CI: 20.8–35.8), respectively.16 This compares well with the data from our cohort, showing that overall MPD with the JAK2 mutation was seen in 20% of patients (3 out of 15). Furthermore, in this cohort, protein C and protein S were tested in nine (23.0%) patients. It is our understanding that since most of our patients were on anticoagulation therapy, not all could be investigated for an underlying thrombophilia leading to the smaller number of patients being investigated. The role of hyperhomocysteinemia is still controversial. In this study, only two patients were tested, with both being found to be negative.

Local factors also play a causative role in PVT. In our study, the most common inflammatory local risk factors leading to PVT were cholangitis (2.5%), cholecystitis (15.3%), and liver abscess (2.5%), although a significant number of patients (33.3%) had malignancy including cholangiocarcinoma, rectosigmoid carcinoma, carcinoma of the head of pancreas, and HCC. In a Swedish study on 254 autopsies, 10% of patients with PVT had abdominal infections or inflammatory disease.5 Furthermore, patients with cirrhosis and hepatic carcinoma had a high risk of PVT with an odds ratio of 17.1 (95% CI: 11.1–26.4).5

Although ultrasound is usually sufficient to make a diagnosis of PVT, CT is the tool of choice.17,18 Furthermore, the sensitivity and specificity of MRI for detecting main PVT are 100% and 98%, respectively.19 Moreover, both CT and MRI provide additional information such as extension of thrombus, evidence of bowel infarction, and the status of adjacent organs. Thus, most cirrhosis patients are routinely followed-up with CT or ultrasound to look for complications or HCC, which would often be incidentally found in PVT as seen in our patient cohort.

PH was the most common complication in 13 patients (33.3%), followed by portal hypertensive gastropathy, esophageal, and gastric varices in agreement with other studies.6–8

It is therefore pertinent to have gastric endoscopy in patients with PVT as portal hypertensive gastropathy is more often present in the acute PVT, secondary to cancer or cirrhosis, whereas large esophageal varices are often present in patients with chronic PVT.20 In chronic PVT, gastric varices were seen in up to 40% of patients.21 In this cohort, seven patients (five patients with cirrhosis and two with chronic PVT) had evidence of variceal hemorrhage on endoscopy either at presentation or having developed 24 hours after admission. Two acute non-cirrhotic PVT patients presented with intestinal ischemia and were operated. One case was found to have developed cirrhosis related to chronic PVT.

Although complete recanalization is uncommon, complete or extensive recovery can be achieved with anticoagulant therapy.3,22 Complete recanalization rates achieved with anticoagulation vary between 42–75% with the risk of extension of thrombus of only 5–7%.3 Initial anticoagulation with heparin is generally followed-up with vitamin K antagonist like warfarin. However, with the recent availability of non-vitamin K antagonist like rivaroxaban, dabigatran, and apixaban, which have better efficacy and lower bleeding complications, the use of warfarin is expected to reduce.23 Patients who received anticoagulation should be maintained on it for at least three months and continued thereafter depending on the underlying cause of thrombosis.

Treatment strategies most often include the use of anticoagulation, while thrombectomy and TIPS being considered as second-line options.24,25 In this study, 46% of patients with PVT received anticoagulation: initially with low molecular weight heparin or heparin followed by warfarin, with one patient being shifted to rivaroxaban. The majority of these patients had non-cirrhotic PVT (87%) with only two patients (13%) having an underlying cirrhosis. One patient was treated with TIPS with no complication. This was in addition to other supportive standard treatments of PH and its complications which include the use of beta blockers, diuretics, and variceal banding. Overall, as it was shown by Kocher and Himmelmann,9 the prognosis of acute non-cirrhotic PVT is good with complete recovery.

The study has many limitations. In particular, its small sample size, retrospective nature, and risk of biases. Although this study data collection spanned over a period of 10 years, we were able to obtain valid data from only 39 patients with PVT from the two institutions. Further, less than 50% of patients got tested for thrombophilia. This is actually a real-life scenario as thrombophilia testing cannot be performed initially during the phase of acute thrombosis (except for DNA studies), and in a majority of these patients who will get treated with anticoagulants, the window of opportunity to test for thrombophilia is only once the patients are off anticoagulants. However, a significant number of patients are lost to follow-up at this point in time. Lastly, one has to be careful in interpreting the thrombophilia testing results in this setting as well as remain alert to the fact that underlying liver cirrhosis will have some false positive results too.

Conclusions

This is the first study on PVT from Oman. We observed putative causative factors in 92.3% of the patient in the present cohort. Therefore, etiological screening is worthwhile in every case with PVT, but thrombophilia screening may not be cost-effective and has several logistic difficulties.

Disclosure

The authors declared no conflicts of interest. No funding was received for this study.

Acknowledgements

We wish to thank the Royal Hospital and Armed Forces Hospital administrations for the use of hospital material for this study.

references

1. DeLeve LD, Valla DC, Garcia-Tsao G; American Association for the Study Liver Diseases. Vascular disorders of the liver. Hepatology 2009 May;49(5):1729-1764.
2. 2. Manzano-Robleda Mdel C, Barranco-Fragoso B, Uribe M, Méndez-Sánchez N. Portal vein thrombosis: what is new? Ann Hepatol 2015 Jan-Feb;14(1):20-27.
3. 3. Chawla YK, Bodh V. Portal vein thrombosis. J Clin Exp Hepatol 2015 Mar;5(1):22-40.
4. 4. Wani ZA, Bhat RA, Bhadoria AS, Maiwall R. Extrahepatic portal vein obstruction and portal vein thrombosis in special situations: Need for a new classification. Saudi J Gastroenterol 2015 May-Jun;21(3):129-138.
5. 5. Ögren M, Bergqvist D, Björck M, Acosta S, Eriksson H, Sternby NH. Portal vein thrombosis: prevalence, patient characteristics and lifetime risk: a population study based on 23,796 consecutive autopsies. World J Gastroenterol 2006 Apr 7;12(13):2115-2119.
6. 6. Kumar A, Sharma P, Arora A. Review article: portal vein obstruction–epidemiology, pathogenesis, natural history, prognosis and treatment. Aliment Pharmacol Ther 2015 Feb;41(3):276-292.
7. 7. Ponziani FR, Zocco MA, Campanale C, Rinninella E, Tortora A, Di Maurizio L, et al. Portal vein thrombosis: insight into physiopathology, diagnosis, and treatment. World J Gastroenterol 2010 Jan;16(2):143-155.
8. 8. Kinjo N, Kawanaka H, Akahoshi T, Matsumoto Y, Kamori M, Nagao Y, et al. Portal vein thrombosis in liver cirrhosis. World J Hepatol 2014 Feb;6(2):64-71.
9. 9. Kocher G, Himmelmann A. Portal vein thrombosis (PVT): A study of 20 non-cirrhotic cases. Swiss Med Wkly 2005 Jun;135(25-26):372-376.
10. 10. Cohen J, Edelman RR, Chopra S. Portal vein thrombosis: a review. Am J Med 1992 Feb;92(2):173-182.
11. 11. Janssen HL, Wijnhoud A, Haagsma EB, van Uum SH, van Nieuwkerk CM, Adang RP, et al. Extrahepatic portal vein thrombosis: aetiology and determinants of survival. Gut 2001 Nov;49(5):720-724.
12. 12. Rosendaal FR. Venous thrombosis: a multicausal disease. Lancet 1999 Apr;353(9159):1167-1173.
13. 13. Tripodi A, Anstee QM, Sogaard KK, Primignani M, Valla DC. Hypercoagulability in cirrhosis: causes and consequences. J Thromb Haemost 2011 Sep;9(9):1713-1723.
14. 14. North Italian Endoscopic Club for the Study and Treatment of Esophageal Varices. Prediction of the first variceal hemorrhage in patients with cirrhosis of the liver and esophageal varices. A prospective multicenter study. N Engl J Med 1988 Oct;319(15):983-989.
15. 15. Qi X, De Stefano V, Wang J, Bai M, Yang Z, Han G, et al. Prevalence of inherited antithrombin, protein C, and protein S deficiencies in portal vein system thrombosis and Budd-Chiari syndrome: a systematic review and meta-analysis of observational studies. J Gastroenterol Hepatol 2013 Mar;28(3):432-442.
16. 16. Smalberg JH, Arends LR, Valla DC, Kiladjian JJ, Janssen HL, Leebeek FW. Myeloproliferative neoplasms in Budd-Chiari syndrome and portal vein thrombosis: a meta-analysis. Blood 2012 Dec;120(25):4921-4928.
17. 17. Ueno N, Sasaki A, Tomiyama T, Tano S, Kimura K. Color Doppler ultrasonography in the diagnosis of cavernous transformation of the portal vein. J Clin Ultrasound 1997 Jun;25(5):227-233.
18. 18. Subramanyam BR, Balthazar EJ, Lefleur RS, Horii SC, Hulnick DH. Portal venous thrombosis: correlative analysis of sonography, CT and angiography. Am J Gastroenterol 1984 Oct;79(10):773-776.
19. 19. Shah TU, Semelka RC, Voultsinos V, Elias J Jr, Altun E, Pamuklar E, et al. Accuracy of magnetic resonance imaging for preoperative detection of portal vein thrombosis in liver transplant candidates. Liver Transpl 2006 Nov;12(11):1682-1688.
20. 20. Sogaard KK, Astrup LB, Vilstrup H, Gronbaek H. Portal vein thrombosis; risk factors, clinical presentation and treatment. BMC Gastroenterol 2007 Aug;7:34.
21. 21. Sarin SK, Lahoti D, Saxena SP, Murthy NS, Makwana UK. Prevalence, classification and natural history of gastric varices: a long-term follow-up study in 568 portal hypertension patients. Hepatology 1992 Dec;16(6):1343-1349.
22. 22. Delgado MG, Seijo S, Yepes I, Achécar L, Catalina MV, García-Criado A, et al. Efficacy and safety of anticoagulation on patients with cirrhosis and portal vein thrombosis. Clin Gastroenterol Hepatol 2012 Jul;10(7):776-783.
23. 23. Pannach S, Babatz J, Beyer-Westendorf J. Successful treatment of acute portal vein thrombosis with rivaroxaban. Thromb Haemost 2013 Oct;110(4):626-627.
24. 24. Fonseca AL, Cleary MA, Cholewczynski W, Sumpio BE, Atweh NA. Omental vein catheter thrombolysis for acute porto-mesenteric vein thrombosis. Ann Vasc Surg 2013 May;27(4):497.e1-497.e4.
25. 25. Luo J, Yan Z, Wang J, Liu Q, Qu X. Endovascular treatment for nonacute symptomatic portal venous thrombosis through intrahepatic portosystemic shunt approach. J Vasc Interv Radiol 2011 Jan;22(1):61-69.