Prospective Investigation Of Pulmonary Embolism Diagnosis Research Articles

The Low-Probability Lung Scan

PULMONARY embolism cannot be diagnosed on clinical grounds; it can only be suspected. In addition, pulmonary embolism often develops in patients who have other conditions that could explain some or all of the symptoms. For example, viral pneumonia may produce symptoms remarkably similar to those of pulmonary embolism. Additional testing is therefore needed to identify those patients who actually do have embolic disease. Angiography is the gold standard for diagnosis, but this invasive test is associated with risk factors of its own. Thus, the Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED) investigators wisely sought a less invasive method to determine which patients could be spared the risk of angiography. Specifically, they studied the current role of ventilation/perfusion (Ve/Pe) scanning in the diagnosis of pulmonary embolic disease. When these investigators issued their report, they stressed that for the best results, Ve/Pe scan findings should be combined with an assessment of the

Neural Network in the Clinical Diagnosis of Acute Pulmonary Embolism

The purpose of this investigation was to test the hypothesis that computer-based pattern recognition can accurately assess the likelihood of acute pulmonary embolism (PE) based on readily obtainable clinical characteristics. Data were obtained from 1,213 patients who participated in the collaborative study of the Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED). Characteristics of the history, physical examination, electrocardiograph, chest radiograph, and arterial blood gases of patients with suspected acute PE were presented to a back propagation neural network. The 1,213 patients were divided into training set A (n = 606) and test set B (n = 607). These groups were then reversed into training set B (n = 607) and test set A (n = 606). A receiver operating characteristic (ROC) curve was constructed from PIOPED clinical assessment, and from neural network clinical assessment in groups A and B. Areas under the respective ROC curves were 0.7450, 0.7477, and 0.7324. All differences were not significant. Areas under ROC curves for PIOPED clinical assessment combined with ventilation/perfusion (V/Q) scan results were compared with neural network clinical assessment combined with V/Q scan results in groups A and B. The respective ROC areas were 0.8324, 0.8203, 0.8496 (all differences not significant). These data show that neural networks were able to predict the clinical likelihood of PE with an accuracy comparable to experienced clinicians.

Mismatched Vascular Defects: An Easy Alternative to Mismatched Segmental Equivalent Defects for the Interpretation of Ventilation/Perfusion Lung Scans in Pulmonary Embolism

The purpose of this investigation was to test the hypothesis that ventilation/perfusion (V/Q) lung scans in patients with suspected acute pulmonary embolism (PE) can be evaluated on the basis of the total number of mismatched vascular defects, irrespective of whether such defects are moderate or large size segmental defects. Lung scan data from the national collaborative study of the Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED) were assessed in 383 patients with acute PE and 681 patients in whom suspected PE was excluded. The predictive value of the cumulative number of mismatched moderate size segmental defects (irrespective of the number of mismatched large segmental defects) was nearly the same as that of mismatched large segmental defects (irrespective of the number of mismatched moderate size segmental defects). This suggests that the diagnostic value of mismatched moderate size segmental defects is the same as mismatched large segmental defects. Lung scans evaluated on the basis of the number of mismatched vascular defects (moderate and/or large segmental defects) were compared with V/Q scans evaluated on the basis of the number of mismatched segmental equivalents. The maximum likelihood estimates of the areas under the receiver operating characteristic (ROC) curves for the number of mismatched vascular defects and for mismatched segmental equivalents were similar (0.8512 vs 0.8530) (NS). Stratification according to the presence or absence of prior cardiopulmonary disease permitted a more accurate assessment of both clinical groups. Evaluation of V/Q scans by vascular defects and by segmental equivalents showed similar areas under the ROC curves. In conclusion, the number of mismatched vascular defects is as powerful for the assessment of V/Q scans as the number of mismatched segmental equivalents. The number of mismatched vascular defects, however, is easier to interpret, and permits a more objective evaluation.

Acute pulmonary embolism: artificial neural network approach for diagnosis.

To investigate use of an artificial neural network (ANN) as a computer-aided diagnostic (CAD) tool for predicting pulmonary embolism (PE) from ventilation-perfusion lung scans and chest radiographs. The data base consisted of cases extracted from the collaborative study of the Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED). Initially, scan findings from 1,064 patients (383 with PE, 681 without PE) were used to train and test the network by using the "jackknife" method. Then, a receiver-operating-characteristic analysis was applied to compare the performance of the network with that of the physicians involved in the PIOPED study. The ANN significantly outperformed the physicians involved in the PIOPED study (two-tailed P value = .01). The findings suggest that an ANN can form the basis of a CAD system to assist physicians with the diagnosis of PE.

Stratification of Patients According to Prior Cardiopulmonary Disease and Probability Assessment Based on the Number of Mismatched Segmental Equivalent Perfusion Defects: Approaches to Strengthen the Diagnostic Value off Ventilation/Perfusion Lung Scans in Acute Pulmonary Embolism

A categorical diagnosis of "high probability" or "intermediate probability" encompasses a spectrum of diagnostic probabilities of pulmonary embolism (PE) that is not communicated to the referring physician. The diagnostic value of ventilation/perfusion lung scans, in the present investigation, was strengthened by use of a table to determine the likelihood of PE in individual patients on the basis of the observed number of mismatched segmental equivalent perfusion defects. In addition, we tested the hypothesis that stratification of patients according to the presence or absence of prior cardiopulmonary disease may enhance the ventilation/perfusion scan assessment of the probability of PE among both of these clinical categories of patients. Data were derived from the collaborative study of the Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED). Ventilation/perfusion lung scans were evaluated in 378 patients with acute PE and 672 patients in whom suspected PE was excluded. Among patients with no prior cardiopulmonary disease, > or = 1.0 mismatched segmental equivalents was indicative of PE in 102 of 118 (86 percent) vs 113 of 155 (73 percent) among patients with prior cardiopulmonary disease (p < 0.02). Among patients with prior cardiopulmonary disease, > or = 2 mismatched segmental equivalents were required to indicate > or = 80 percent probability of PE. Stratification on the basis of the presence or absence of prior cardiopulmonary disease, therefore, enhanced the ability of ventilation/perfusion scan readers to assign an accurate positive predictive value and specificity to individual patients based on the observed number of mismatched segmental equivalent defects. Among patients with no prior cardiopulmonary disease, fewer mismatched segmental equivalent defects were required to indicate a high probability of PE than were required by PIOPED criteria. The findings from some of these patients, by PIOPED criteria, would have indicated intermediate probability. Some indeterminate probability readings, therefore, will be eliminated among patients stratified with no prior cardiopulmonary disease.

Prospective validation of the stripe sign in ventilation-perfusion scintigraphy.

The data base of the Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED) was used to test the accuracy of the stripe sign for the exclusion of embolic perfusion defects on lung scintigrams. Lung scan readings showed the presence and location of this sign in 50 (4.7%) of 1,064 patients. Perfusion defects showing the stripe sign were not associated with pulmonary embolism in the same lung zone (upper, middle, or lower third of each lung) in 93% (79 of 85) of instances. Thirty-eight percent (n = 32) of lung zones with the stripe sign had associated chest radiographic abnormalities, and 69% (n = 59) had ventilation scan abnormalities. Formulation of the scan diagnosis according to PIOPED criteria showed fewer indeterminate readings in patients with the stripe sign and without pulmonary embolism when the stripe sign was used. Use of the sign changed diagnosis in less than 1% of the total population, however, because of its low overall prevalence. The stripe sign is a useful adjunct to standard criteria in the interpretation of pulmonary scintigrams for evaluation of suspected acute pulmonary embolism.

Value of ventilation/perfusion scans versus perfusion scans alone in acute pulmonary embolism

The value of ventilation/perfusion (V/Q) lung scans in the diagnosis of acute pulmonary embolism (PE) was described in the results of the collaborative study of the Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED). 1 The effects of preexisting cardiac or pulmonary disease and of a normal chest radiograph on the diagnostic value of V/Q scans 2,3 were also described among patients in PIOPED. The present report assesses the value of V/Q scans compared with that of perfusion scans alone in the diagnosis of acute PE.

Complications and validity of pulmonary angiography in acute pulmonary embolism.

The Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED) addressed the value of ventilation/perfusion scans in acute pulmonary embolism (PE). The present study evaluates the risks and diagnostic validity of pulmonary angiography in 1,111 patients who underwent angiography in PIOPED: Complications were death in five (0.5%), major nonfatal complications in nine (1%), and less significant or minor in 60 (5%). More fatal or major nonfatal complications occurred in patients from the medical intensive care unit than elsewhere: five of 122 (4%) versus nine of 989 (1%) (p less than 0.02). Pulmonary artery pressure, volume of contrast material, and presence of PE did not significantly affect the frequency of complications. Renal dysfunction, either major (requiring dialysis) or less severe, occurred in 13 of 1,111 (1%). Patients who developed renal dysfunction after angiography were older than those who did not have renal dysfunction: 74 +/- 13 years versus 57 +/- 17 years (p less than 0.001). Angiograms were nondiagnostic in 35 of 1,111 (3%), and studies were incomplete in 12 of 1,111 (1%), usually because of a complication. Surveillance after negative angiograms showed PE in four of 675 (0.6%). Angiograms, interpreted on the basis of consensus readings, resulted in an unchallenged diagnosis in 96%. The risks of pulmonary angiography were sufficiently low to justify it as a diagnostic tool in the appropriate clinical setting. Clinical judgment is probably the most important consideration in the assessment of risk.

Clinical characteristics of patients with acute pulmonary embolism

Among 117 patients with pulmonary embolism (PE) and no prior cardiac or pulmonary disease who participated in the National Heart, Lung, and Blood Institute Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED) study, combinations of clinical characteristics were identified that were present in nearly all in whom the diagnosis of PE was made. 1 We now assess these as well as additional combinations of characteristics in the entire population of patients with acute PE who participated in PIOPED, irrespective of the presence of prior cardiopulmonary disease.

Recent advances in the diagnosis and lytic therapy of pulmonary embolism.

During the past year, there have been 2 major advances in the management of pulmonary embolism (PE). First, the Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED) investigators published the results of their comparison of ventilation-perfusion lung scanning and pulmonary angiography. This multimillion-dollar trial sponsored by the National Heart, Lung, and Blood Institute indicated that lung scanning is surprisingly insensitive to the diagnosis of PE. High-probability lung scanning detects fewer than half of the cases of PE that are found at pulmonary angiography. The PIOPED results force us to conclude that increased utilization of both ultrasonography of the deep leg veins and pulmonary angiography is warranted in order to detect the majority of cases of venous thromboembolism. Second, in June 1990, the Food and Drug Administration approved recombinant human tissue-type plasminogen activator (rt-PA) for use in the treatment of acute PE. The dosing regimen is 100 mg of rt-PA as a continuous peripheral intravenous infusion administered over 2 h. The convenience, rapid effect, and relative safety of this therapeutic approach may result in increased use of thrombolysis for management of PE.

An Evaluation of Preoperative and Postoperative Ventilation and Perfusion Lung Scintigraphy in the Screening for Pulmonary Embolism After Elective Orthopedic Surgery

One hundred two patients undergoing elective knee or hip arthroplasty were studied with radionuclide ventilation scans (V) and perfusion scans (Q) preoperatively (preop) and postoperatively (postop) to assess their relative value in the diagnosis of asymptomatic pulmonary embolism (PE) after orthopedic surgery. Postop Q were read in combination with preop V and Q and postop V using prospective investigation of pulmonary embolism diagnosis (PIOPED) criteria. Of 25 postop Q interpreted as either high or intermediate probability for PE, preop Q were judged useful in 96%; the postop V were useful in 78%; and the preop V were not helpful in any of the cases. Of 63 postop Q interpreted as low probability, preop Q were useful in 74%; the postop V were useful in only 33%; and the preop V were useful in only one case. When postop Q were read as normal (14 cases), none of the three auxiliary studies were found to be useful. Overall, postop V were more helpful than preop Q in only 2%, and preop V contributed significantly in only 1%. This experience suggests that preop Q alone is the most useful adjunct to the postop Q in the postoperative evaluation for PE. The authors conclude that to screen for asymptomatic PE after elective orthopedic surgery, preop Q should be performed in all cases, preop V are not necessary, and postop V need be performed only if a baseline preop Q is not available.

The Diagnosis of Pulmonary Embolism

To the Editor.— The PIOPED (Prospective Investigation of Pulmonary Embolism Diagnosis) study, 1 being a large, multicenter study, will undoubtedly have an effect on the use of lung ventilation/perfusion scans by many other physicians. However, the conclusions of the study are seriously flawed. All ventilation imaging was done only in posterior view and with xenon 133. Ventilation imaging is best done in the views that show perfusion defects. If aerosol is used, then the same six views as on perfusion imaging can be obtained. The drawback to this is the limitation of aerosol imaging to assessing only early or wash-in ventilation. Xenon 127 gives images with much better resolution than does xenon 133 and, unlike xenon 133, can be used after perfusion imaging without interference from technetium Tc 99m photons. Computer subtraction of the perfusion images is therefore not necessary, as it is if xenon 133 is used after technetium

Value of the ventilation/perfusion scan in acute pulmonary embolism. Results of the prospective investigation of pulmonary embolism diagnosis (PIOPED). The PIOPED Investigators

Value of the ventilation/perfusion scan in acute pulmonary embolism. Results of the prospective investigation of pulmonary embolism diagnosis (PIOPED). The PIOPED Investigators