Screening cancer after venous thromboembolism: How many abnormal tests before diagnosing cancer? An analysis of...

Since Trousseau, we knows that venous thrombemboembolism (VTE) can reveal occult cancer. Different strategies of cancer screening have been evaluated: they are often time-consuming, cause stress and anxiety, and frequently require second-look examinations (due to the risk of false positives), with ultimately a very low yield (about 5%). We evaluated the number of suspect cancer tests before reporting them to the number of cancers finally diagnosed, after a VTE, in the setting of practice's analysis.

We studied retrospectively patients hospitalized for a VTE and with a cancer screening, between 2011 and 2012. Screening cancer was defined by performing at least one of the following tests: PSA, fecal occult blood test, mammography, abdominopelvic iconography (abdominal ultrasound and/or abdominal CT scan). We recorded the suspected cancer tests, the cancers diagnosed, their stage and the survival. These results were expressed as a percentage with a 95% confidence interval.

Out of the 491 patients treated for a VTE, screening cancer was performed on 295 patients (median age 66.2 years). Nineteen PSA (16.7%, 95% CI [10.3-25]) were abnormal, with 2 localized prostate cancers. Nineteen fecal occult blood tests (15.3%, 95% CI [9.5-23]) were positive, with 2 local cancers. Five mammograms suspected cancer (4.7% 95% CI [1.6-10.8]) for one confirmed. Thirty-eight abdomino-pelvic iconographies (14.4% 95% CI [10.4-19.2]) were suspect, with 7 confirmed cancers, 6 being metastatic at times of diagnostic.

Among the 607 tests performed, 81 were suspected of cancer (13.3%) for only 12 cancers confirmed (2.0%). Screening cancer exposes patients to several false positive tests.