Screening for Occult Malignancy Following First Unprovoked VTE: A Single Centre Experience

There are clearly demonstrated links between unprovoked venous thromboembolism (VTE) and underlying malignancy. Previous studies have shown an incidence between 3 and 13% of subsequent cancer diagnoses in patients with unprovoked VTE. National guidance issued in the United Kingdom, 2012 recommend that all patients with a first unprovoked VTE are investigated for occult malignancy with a thorough history and examination, full blood count, liver function tests, calcium, chest X-ray and urinalysis with directed investigation of any positive findings. Additionally, abdomino-pelvic CT scans (and mammography in women) should be considered for all patients over 40 years with first unprovoked VTE without positive findings on basic investigations.

We retrospectively reviewed all patients diagnosed with unprovoked VTE at King's College Hospital between April 2014 and March 2015, and results were followed up to July 2015. We excluded as provoked VTEs all cases associated with trauma, known malignancy, recent surgery or hospitalisation, prolonged immobilisation, long-haul travel, hormonal therapy, intravenous drug use, pregnancy or the puerperium. We examined extent of investigations performed and reviewed the incidence of occult malignancy in those with a first unprovoked VTE.

Over the period of study, 544 patients were objectively diagnosed with pulmonary embolism (PE) or deep vein thrombosis (DVT). Of these, 140 cases were unprovoked in nature. 75/140 (53.6%) were male, with a median age of 56 years (range 22-97). All 140 patients had initial clinical assessment and bloods tests. 113 (80.7%) patients also had chest X-ray screening performed. Of the remaining 27 patients, 4 were not followed up in our centre. 75 (53.6%) patients had tumour markers taken, 74 (52.9%) patients had abdominal imaging (of which 61, 82.4% had CT abdomen and pelvis, remainder ultrasound) and 3 women had mammography. Tumour markers were abnormal in 26/75 (34.7%). Abdominal imaging was abnormal in 33/66 (50.0%) patients without a subsequent diagnosis of malignancy, with 18/66 (27.3%) requiring additional investigation to definitively exclude malignancy. 8/136 (5.9%) cases of occult malignancy were identified (see Table for characteristics). The majority of patients found to have occult malignancy were diagnosed at an advanced stage, with high subsequent mortality rates and minimal opportunity for intervention.

Our findings compare favourably with the findings of the SOME trial with a low incidence of occult malignancy and questionable value of routine abdomino-pelvic imaging for otherwise asymptomatic patients with first unprovoked VTE. Such screening is likely to incur anxiety for patients, incidental findings and higher costs without demonstrable patient benefit. Abnormal tumour markers were common and non-specific and should not be performed routinely following unprovoked VTE. Targeted investigation for individuals with suggestive clinical features or abnormalities on baseline bloods, chest X-ray or urinalysis should be considered.