Continuous volume computed tomography in pulmonary embolism: the answer, or just another test?

Thorax Editorials Continuous volume computed tomography in pulmonary embolism: the answer, or just another test? Few exercises in clinical medicine are more troublesome than securing a diagnosis of pulmonary embolism. Because of the protean clinical features of pulmonary embolism, much reliance has been placed on the direct and indirect imaging of clot within the pulmonary arteries and/or peripheral venous system. This has prompted the cynical observation that radiologists and other imagers devise tests of dubious value faster than clinicians can discard them. Increasingly elaborate diagnostic algorithms have been recommended, which in reality are seldom followed. The ideal diagnostic test would reliably and safely show clot within the pulmonary arteries. Pulmonary arteriography is regarded as the final arbiter, simply because it directly images embolus within the pulmonary arteries. The assumption that life threatening emboli are not missed by pulmonary arteriography seems to hold true.'2 However, it is worth assaying this gold standard from time to time; although false positives are difficult to prove and probably scarce, false negatives can occur despite extensive intra-vascular thrombus.3 In the UK ventilation-perfusion (V/Q) scanning is more widely available than pulmonary arteriography. The PIOPED study2 has shown that, when the suspicion of pulmonary embolism is high, based on clinical and laboratory findings, and the V/Q scan is in the high probability category, the chance of a positive pulmonary arteriogram exceeds 95%. At the opposite extreme, a low suspicion of pulmonary embolism and a normal V/Q scan is associated with a normal arteriogram in 95% of cases. Unfortunately, most patients do not fall into these convenient categories and the problem of establishing which patients require anticoagulation without resort to pulmonary arteriography remains unresolved. It has recently been argued that patients with indeterminate V/Q scan or high V/Q probability with discordant clinical probability should undergo compression ultrasonography of the lower limbs4 with a view to empirical treatment if deep venous thrombosis is shown. However, a negative ultrasonographic test results in the common dilemma of whether to proceed to pulmonary arteriography or to treat with anticoagulation without confirmation of the diagnosis. The recent recognition that modern computed tomo-graphy (CT) scanners are able to detect pulmonary emboli complicates the diagnostic algorithm. Advances in CT technology have allowed a gradual increase in the speed of data acquisition and image reconstruction. In the early 1970s each slice required up to five minutes to acquire, but by the end of the 1980s this …