For some clinicians, CTPA has an illusion of certainty in the diagnosis of pulmonary embolism. We analyzed the frequency and causes of false positive CTPA.
We reviewed the reports of 6303 consecutive CTPA performed between 11/7/06 and 12/31/08 at three Northeastern US community hospitals. Official reports were reviewed for the presence or absence of pulmonary embolism (PE). A chest radiologist with 18 years experience reviewed positive CTPA for the following features: 1) presence of PE, 2) reasons for false positive exam, 3) exam quality, 4) location of the largest PE.
608 (10%) CTPA were officially reported positive for PE. On review, 68/608 (11%) were either false positive (FP) or probable false positive (PFP) exams. Of FP or PFP exams, 46/68 (68%) were at the subsegmental level. Of reported subsegmental PE (SSPE), 46/168 (27%) were FP or PFP exams. The majority (56/68, 82%) of FP or PFP exams were caused by technical factors including motion artifact, pulsation artifact, streak artifact, volume averaging, suboptimal bolus and quantum mottle. FP or PFP exams ( 31/56, 55%) were more likely than TP or PTP exams (49/510, 10%) to be of diminished quality that could impact the detection of SSPE. In many of the FP or PFP exams, the official report indicated uncertainty for PE.
A significant percentage of SSPE reported on CTPA were FP or PFP studies. Most FP or PFP exams were caused by artifacts in diminished quality CTPA. The official report often indicated the uncertainty for PE in these studies.
Clinicians need to be aware of the high rate of FP or PFP exams when only SSPE are reported in the context of diminished image quality CTPA .
Wallace Miller Jr, Other Supported by the Sharpe-Strumia Research Foundation of the Bryn Mawr Hospital; No Product/Research Disclosure Information