Incidental Finding of Malignancy by CT Angiogram of Chest in Patients Highly Suspected for Pulmonary Embolism.

Introduction: Venous thromboembolism (VTE), including both pulmonary embolism (PE) and deep venous thrombosis (DVT) is often a silent disease, symptoms are often nonspecific, and the first manifestation may be fatal PE. The link between malignancy itself, and its treatment, type, stage and location, and clinical recurrent VTE has been well recognized since 1865, when Armed Trousseau first reported it. 10–20% of patients with acute VTE have malignancy. VTE may appear before the cancer has become symptomatic and may lead to an earlier diagnosis of cancer; the most common of them include those of the breast, colon, and lung. VTE is the 2nd leading cause of mortality in cancer patients, 2nd only to the cancer itself. Patients with concurrent VTE and malignancy have a more than 3-fold higher risk of recurrent VTE and death, and 4- to 8-fold higher risk of dying after an acute VTE than patients with VTE but no malignancy.

Spiral computed tomography angiogram of chest (CTA) was first introduced in 1990s. It has the ability to study pulmonary vessels down to 2-3 mm in diameter and depict PE directly. Only 30–40% of all patients with suspected PE have a confirmed diagnosis. CTA can also provide significant additional information or an alternate diagnosis.

Objectives: To determine the frequency of an incidental finding of occult malignancy by CTA of chest in patients highly suspected for PE.

Materials and methods: 136 patients, suspected of having PE, were retrospectively studied. We included all the patients, examined by CTA of the chest to diagnose PE from October 2005 through April 2006 at BHMC. We used chart review to evaluate the contribution of CTA to the final clinical diagnosis. The baseline characteristic parameters included age, gender, CTA findings, and pathological examinations.

Results: We studied both positive and negative CTA images for searching of evidence of occult malignancy.134 (97.8%) patients had at least one diagnostic finding. Of the 136 patients, highly suspected for PE, 33 (24.3%) patients were confirmed to have PE. A total of 11 (8.1% of total studied) patients were identified with new diagnosis of malignancies. Of these, cancer was detected in 3 (9.1%) patients with PE, and in 8 (7.8%) patients without PE. Thromboembolic risk factors were identified in 2 (66%) patients with positive PE and cancer (e.g. 14 hour flight, immobilization> 3 days). In 63.6% (#7) of patients cancer was diagnosed by initial CTA, and in 36.4% (#4) of patients with follow up studies and clinical evaluations for unresolved initial symptoms during a range of 1 to 7 months (mean of 3.4 months), 3 of the latter group being negative for PE. 72.8% (# 8) of patients had metastatic disease at the time of initial CTA. The most common cancers were gastric adenocarcinoma (# 3, 27.3%), and breast cancer (# 2, 18.2%) of all the patients with cancers. The most common alternative diagnosis other than malignancies, detected by CTA included pleural effusion in 53 (39%) and atelectasis in 48 (35.3%) patients. Only 3 (2.2%) CT angiograms were completely normal.

Conclusion: CTA was able to identify occult malignancy in 8.1% of total patients, suspecting PE. Since screening for occult malignancies has been generally underutilized, cancer should be included in the differential etiologies of acute PE, and basic comprehensive evaluation to search for underlying cancer in high-risk patients should be considered.