Lung scintigraphy is used widely for diagnosis of pulmonary embolus (PE). Technegas ventilation imaging has many advantages over other methods, but little outcome data exists on this technique. The aims of this study were to better define the role of lung scintigraphy in the management of patients with suspected PE and to evaluate technegas ventilation imaging by following patient outcomes. METHODS A group of 717 out of 834 consecutive patients, referred to a university teaching hospital for lung scintigraphy to confirm or refute the diagnosis of PE, was followed for 18-30 mo to determine clinical outcome. The follow-up endpoints were death as a result of PE, death as a result of hemorrhage after treatment for PE, uncomplicated survival, survival with subsequent PE, nonfatal hemorrhage after treatment for PE and recurrence of PE in treated patients. Ventilation imaging was performed using technegas, and perfusion imaging was performed using intravenous 99mTc macroaggregated albumin. The modified PIOPED (Prospective Investigation of Pulmonary Embolism Diagnosis) diagnostic criterion was used for interpretation of lung scintigraphy. RESULTS Diagnostic results included 3.5% normal studies, 67.4% assessed as low probability for PE, 10% as moderate probability for PE and 19.1% as high probability for PE. A total of 231 patents received therapy with heparin, followed by warfarin, including those receiving anticoagulation therapy for other conditions. Ninety-six percent of patients with normal and low probability studies (n = 508) had good outcomes, 6 patients died as a result of PE and 12 subsequently developed PE. The odds ratio for death by PE in this group was 0.2. Of the 72 moderate probability studies, 39 patients were untreated. In this group there was 1 death due to PE, and PE subsequently developed in 2 patients. None of the remaining 33 treated patients died, but 4 patients experienced bleeding complications. The odds ratio for death by PE in the moderate probability group was 0.7. In those patients with high-probability studies, there were 8 deaths by PE, 6 deaths by hemorrhage, 11 nonfatal hemorrhages and 7 patients who experienced recurrences of PE. The odds ratios in this group were 6 and 10 for death by PE, or death by PE and the treatment of PE, respectively. CONCLUSIONS The use of the modified PIOPED diagnostic classification is valid for technegas lung scintigraphy. Using technegas, normal/low-probability and high-probability results are highly predictive of respective outcomes. Technegas lung scintigraphy reduces the number of indeterminate studies.