Can Splanchnic Venous Thrombosis Predict Subsequent Cancer Diagnosis?

Research has shown that embolic events put patients at an increased risk for developing cancer: Venous thromboembolism (VTE) may represent a marker for occult cancer, and patients with lower-limb deep venous thrombosis or pulmonary embolism have a two- to four-fold increased risk of cancer in the first year following the thromboembolic event. Less is known, however, about the cancer risk associated with splanchnic venous thrombosis (SVT) – clots in the veins that carry blood through the liver and other abdominal organs.

In a recent study published in Blood, Kirstine K. Søgaard, MD, and colleagues sought to determine the link between SVT and the development of cancer – and whether SVT could indicate an undiagnosed cancer – as previous case reports have indicated that SVT is the first sign of liver and pancreatic malignancies.

“As we learn more about the association between many types of thromboses and cancer, we also want to better understand these more rare clots and how they can perhaps signal a hidden cancer,” said Dr. Søgaard. “In this case, we had access to comprehensive data that we believed could provide insights useful to clinicians caring for patients with this condition.”

Dr. Søgaard, of the Department of Clinical Epidemiology at Aarhus University Hospital in Denmark, and investigators analyzed data from the Danish National Patient Registry of 1,191 patients diagnosed with SVT from 1994 to 2011. Information on liver disease, pancreatitis, diabetes, chronic obstructive pulmonary disease, VTE, congestive heart failure, and myocardial infarction diagnosed at any time before SVT were all collected. The SVT patients were followed from the date of first diagnosis of SVT until one of the following occurred: cancer diagnosis, death, emigration, or deadline of December 31, 2011.

To identify subsequent cancer diagnoses, the researchers linked the SVT patients to the Danish Cancer Registry, which records all cancers diagnosed in Denmark since 1943, including the month and year of diagnosis. Risk of cancer was compared to the expected risk in the general Danish population.

Of the SVT patients:

  • 78 percent had portal vein thrombosis (n=924)
  • 12 percent had hepatic vein thrombosis (n=141)
  • 10 percent had mesenteric vein thrombosis (n=126)

During the 1.6-year median follow-up (range, 0-5 years), 183 incident cancers were observed, for an overall standardized incidence ratio (IR) of 4.2 (95% CI 3.6-4.9). In terms of cancer risk at the individual clot sites, the majority of cancers (88%; n=161) were observed in patients with portal vein thrombosis, with an overall standardized IR of 4.7 (95% CI 4.0-5.5].

Notably, more than half of these cancers were diagnosed during the first three months of SVT diagnosis. Three-month and five-year absolute risks of cancer among patients with SVT were 8 percent and 14.8 percent, respectively.

“Our findings support that a diagnostic workup for cancer may be important in patients with portal vein thrombosis and hepatic vein thrombosis,” Dr. Søgaard told ASH Clinical News. “Because recurrence and complication rates after thrombosis are more frequent among cancer patients, detection of underlying cancer will likely influence the treatment of thrombosis.”

The results from this analysis are derived from a population database, meaning that not all details about SVTs, subsequent therapies, nor treatments for those therapies were available. Results could also not be controlled for intensity of follow-up in patients with an SVT diagnosis – and whether patients in the general population or the cancer diagnosis groups received differential follow-up.

The majority of these cancers were liver (absolute risk = 3.5%; standardized IR = 1805; 95% CI 1295-2448), pancreatic (absolute risk = 1.5%; standardized IR = 256; 95% CI 149-409), or blood (absolute risk = 0.7%; standardized IR = 764; 95% CI 329-1505). For blood cancers, the most commonly identified cancers in SVT patients were myeloproliferative neoplasms, and were more frequently diagnosed in patients who had portal and hepatic vein thrombosis.

“We had no information on test results for JAK2V617F mutation,” the authors wrote, “but it is possible that the finding of this mutation was related to the diagnosis of myeloproliferative neoplasms in some patients.”

After determining that SVT confers an additional cancer risk, the researchers also assessed how these abdominal blood clots might affect cancer patients’ survival, comparing their survival outcomes with a matched cohort of cancer patients without SVT. While patients with liver or pancreatic cancers had poor survival regardless of SVT, patients with SVT and these cancers had markedly worse three-month survival outcomes than cancer patients without the clots:

  • Three-month survival after liver cancer diagnosis: 44 percent with SVT versus 55 percent without SVT
  • Three-month survival after pancreatic cancer: 35 percent with SVT versus 53 percent without SVT

While patients with hematologic cancer diagnoses seemed to have better prognoses, the authors explained that an analysis of SVT relative to mortality could not be extrapolated due to the low mortality rate in this patient cohort. Furthermore, the difference between survival rates with and without the clots was not significant, but there was a higher incidence of myeloproliferative neoplasms beyond 12 months following SVT diagnosis – potentially indicating delayed diagnosis.

“We speculated whether abdominal computed tomography (CT) or fluorodeoxyglucose-positron emission tomography/CT scans should be mandatory in the diagnostic workup in patients presenting with SVT,” Dr. Søgaard told ASH Clinical News. “However, before we can propose new screening regimens for occult cancer, we need to know if this will actually improve cancer-associated survival – and whether it is cost-effective.”


Søgaard KK, Farkas DK, Pedersen L, Sørensen HT. Splanchnic venous thrombosis is a marker of cancer and a prognostic factor for cancer survival. Blood. 2015 June 18. [Epub ahead of print]