PE/ABG references ----------------- The PIOPED Investigators Value of the Ventilation/Perfusion Scan in Acute Pulmonary Embolism. JAMA 1990;263:2753-2759 Overton DT, Bocka JJ The alveolar-arterial oxygen gradient in patients with documented pulmonary embolism. In: Arch Intern Med (1988 Jul) 148(7):1617-9 It has been reported that the finding of a normal PaO2 level on arterial blood gas analysis does not exclude the diagnosis of acute pulmonary embolism. We wished to determine whether a more thorough evaluation of the blood gases would prove more helpful; specifically, whether it is possible for a patient with acute pulmonary embolism to have a normal alveolar-arterial (A-a) oxygen gradient. We studied this question in a patient population in which the diagnosis was definitively made via pulmonary arteriography. Sixty-four patients met all study criteria. In these patients, the A-a gradient ranged from 11.6 to 83.9 mm Hg (mean, 41.8 mm Hg). In three patients, the A- a gradient was normal for age. We conclude that a normal A-a oxygen gradient does not exclude the diagnosis of acute pulmonary embolism, and should not preclude further diagnostic procedures if there is a high index of suspicion. Commentary: This article retrospectively reviews 64 patients who had proven PE on PA-gram, and in whom the A-a gradient on room air was available. They found that only 3/64 patients (4.7%) with documented PE had a normal A-a gradient corrected for age (correction formula: normal A-a gradient = 4 + [age/4]). The article breaks down the history and findings for each patient with a normal ABG. You can thus determine if you would have considered them very low likelihood for PE. --Jon Handler, M.D. Worsley DF, Palevsky HI, Alavi A A detailed evaluation of patients with acute pulmonary embolism and low- or very-low-probability lung scan interpretations. In: Arch Intern Med (1994 Dec 12-26) 154(23):2737-41 BACKGROUND: To determine the clinical characteristics of patients with pulmonary embolism (PE) and low- or very-low-probability ventilation-perfusion lung scan interpretations. METHODS: A retrospective analysis of the data obtained during the Prospective Investigation of Pulmonary Embolism Diagnosis study was performed. The clinical characteristics of patients with acute PE and low- or very-low-probability lung scan interpretation (false-negative interpretations) were compared with patients who had low- or very-low- probability lung scan interpretations and no evidence of acute PE (true-negative interpretations). RESULTS: Of the 1493 patients who gave consent to participate in the Prospective Investigation of Pulmonary Embolism Diagnosis study, 399 patients had angiographic or autopsy evidence of acute PE. Pulmonary embolism was excluded in 960 patients. Patients with false-negative lung scan interpretations more commonly had a history of immobilization (P < .0001), trauma to the lower extremities (P < .003), recent surgery (P < .002), or central venous instrumentation (P < .04) compared with patients with true- negative lung scan interpretations. In patients with low- or very-low- probability lung scan interpretations and none of the above-mentioned risk factors, the prevalence of PE was only 4.5%. In contrast, for patients with low- or very-low-probability lung scan interpretations and two or more of the above-mentioned risk factors, the prevalence of PE was 21%. CONCLUSIONS: Patients with a history of immobilization, trauma to the lower extremities, recent surgery, or central venous instrumentation were more likely to have false- negative lung scan interpretations. Therefore, this population warrants special attention when deciding on the need for peripheral venous studies or angiography in patients with low- or very-low- probability lung scan interpretation. Goodman LR Curtin JJ Mewissen MW Foley WD Lipchik RJ Crain MR Sagar KB Collier BD Detection of pulmonary embolism in patients with unresolved clinical and scintigraphic diagnosis: helical CT versus angiography. In: AJR Am J Roentgenol (1995 Jun) 164(6):1369-74 OBJECTIVE. This study was designed to prospectively compare helical CT with pulmonary angiography in the detection of pulmonary embolism in patients with an unresolved clinical and scintigraphic diagnosis. SUBJECTS AND METHODS. Twenty patients with an unresolved suspicion of pulmonary embolism were evaluated with contrast-enhanced helical CT and with selective pulmonary angiography. An average of 11 hr separated the two studies. The CT scans were obtained during one 24- sec or two 12-sec breath-holds. CT scans were interpreted without knowledge of the results of scintigraphy or angiography. Selective pulmonary angiograms were obtained with knowledge of the findings on the ventilation/perfusion scan only. The sensitivity and specificity of CT were compared with those of angiography for central vessels (segmental and larger) only and for all vessels. RESULTS. Eleven of the 20 patients had proved pulmonary embolism (seven in central vessels and four in subsegmental vessels only). When only central vessels were analyzed, CT sensitivity was 86%, specificity was 92%, and the likelihood ratio was 10.7. However, when subsegmental vessels were included, CT results were 63%, 89%, and 5.7, respectively. CONCLUSION. In our subset of patients, helical CT was only 63% sensitive. Subsegmental emboli are difficult to diagnose. Pulmonary angiography remains the study of choice. CT has a limited role in the evaluation of acute pulmonary embolism. Caracci BF Rumbolo PM Mainini S Walker HS 3d Peterson GJ How accurate are ventilation-perfusion scans for pulmonary embolism? In: Am J Surg (1988 Dec) 156(6):477-80 To evaluate the accuracy of ventilation-perfusion scanning in the diagnosis of pulmonary embolism, the pulmonary arteriograms of 55 patients suspected of pulmonary embolism were compared with their ventilation-perfusion scans. The clinical presentation was consistent for the diagnosis of pulmonary emboli in all 55 patients. The scans were divided into the following four categories according to standard guidelines: normal, low, intermediate, and high probability. Three patients had normal scans confirmed by arteriography; 34 patients had high-probability scans, but only 22 (65 percent) were positive at arteriography. The remaining 18 patients had low or intermediate- probability scans (9 patients each), 5 of whom (28 percent) were positive at angiography (sensitivity 82 percent, specificity 57 percent). This study demonstrates the inaccuracy of ventilation- perfusion scanning for the evaluation of pulmonary embolus and may represent the variability of interpretation at individual institutions. Perhaps each institution may need to compare the results of ventilation-perfusion scanning and angiography to optimally select and treat patients. Oudkerk M van Beek EJ van Putten WL Buller HR Cost-effectiveness analysis of various strategies in the diagnostic management of pulmonary embolism. In: Arch Intern Med (1993 Apr 26) 153(8):947-54 BACKGROUND: Since the clinical diagnosis of pulmonary embolism is unreliable, various objective diagnostic methods (or combinations thereof) are advocated. Pulmonary angiography is the accepted reference method but is considered less suitable for initial screening due to its invasive nature. Therefore, at least nine different diagnostic management strategies employing invasive and noninvasive diagnostic tests are used in clinical practice. We assessed the cost-effectiveness of these strategies to help identify the optimal approach. METHODS: Based on assumptions derived from published data, we calculated mortality, morbidity, adequacy for the indication of anticoagulant therapy, and associated diagnostic and therapeutic costs using a decision analytic model. Additionally, a cost-effectiveness analysis was performed using incremental costs per additional life saved. RESULTS: The analysis identified three separate categories of diagnostic management strategies: (1) Treatment of all patients (or of those with an abnormal perfusion scan) results in the lowest mortality and morbidity rates but the highest costs due to inappropriate treatment of 55% to 70% of patients. (2) Pulmonary angiography strategies with or without prior perfusion-ventilation lung scintigraphy and ultrasonography of the legs have comparable low mortality and morbidity rates, costs savings of approximately 40%, and inappropriate treatment in fewer than 5% of patients. (3) Use of perfusion-ventilation scintigraphy with or without ultrasonography results in unacceptably high mortality rates. CONCLUSIONS: At present, the optimal diagnostic management strategy should include pulmonary angiography. Use of perfusion-ventilation lung scintigraphy and ultrasonography results in a 40% to 50% reduction in the number of patients requiring pulmonary angiography and is cost-effective. Stein PD Hull RD Saltzman HA Pineo G Strategy for diagnosis of patients with suspected acute pulmonary embolism. In: Chest (1993 May) 103(5):1553-9 STUDY PROTOCOL: Two separate groups of clinical investigators have provided new information and divergent approaches to the management of acute pulmonary embolism (PE). In this position paper, investigators from both groups (Prospective Investigation of Pulmonary Embolism Diagnosis [PIOPED] and Canadian study groups) have utilized the combined scientific database in order to rationalize seemingly polarized diagnostic recommendations into a single practical algorithm. METHODS: An in-depth review established the relative risks of deep venous thrombosis (DVT) and the related accuracy of diagnostic tests. In PIOPED, 640 of 887 patients at risk for PE had either an intermediate probability ventilation/perfusion (V/Q) scan or a V/Q scan probability that was discordant with the prior estimate of probability by clinical assessment. The risk of PE in these patients was 16 to 88 percent (average, 34 percent). In this group, we calculated the probability of PE assuming that tests for DVT had been performed and that 50 percent of patients with PE have detectable proximal DVT. By calculation, 108 in 640 patients of whom the diagnosis of PE was uncertain, would have shown proximal DVT. In 239 of these 640 patients, tests for DVT would have been negative and the risks of PE in these patients is calculated to be less than 10 percent. RESULTS: Therefore, we calculate that in 347 of 640 patients, confident recommendations for treatment or no treatment could have been given without pulmonary angiography. Accordingly, in the PIOPED study group of 887 patients, the need for pulmonary angiography would have been reduced from 640 (72 percent) to 293 patients (33 percent). CONCLUSION: In conclusion, a diagnostic strategy that includes the clinical evaluation, V/Q scan, and evaluation for DVT would decrease the number of patients who require pulmonary angiography from 72 to 33 percent. Cooper TJ Hayward MW Hartog M Survey on the use of pulmonary scintigraphy and angiography for suspected pulmonary thromboembolism in the UK [see comments] In: Clin Radiol (1991 Apr) 43(4):243-5 A survey of UK radiologists in 360 acute hospitals was undertaken to assess the current use of pulmonary angiography and radionuclide lung scanning in the investigation of suspected pulmonary thromboembolism (PTE). Replies were received from 340 (94%) hospitals, of which 48 (out of 50) were teaching hospitals or centres which perform cardiothoracic surgery. Lung scintigraphy was provided by 200 (59%) hospitals and angiography by 120 (35%), with 99 (29%) providing both. Twenty-two hospitals which could undertake angiography had not performed more than one angiogram for suspected PTE in the preceding 3 years. Average numbers of lung scans performed in each hospital were 21.8 per month, in contrast with an average of 4.1 angiograms (including digital subtraction angiography) per year performed for suspected PTE over the previous 3 years. The total number of V/Q lung scans performed for the diagnosis of PTE per year was approximately 47,000 compared with 490 pulmonary angiograms. Quinn RJ Butler SP A decision analysis approach to the treatment of patients with suspected pulmonary emboli and an intermediate probability lung scan [see comments] In: J Nucl Med (1991 Nov) 32(11):2050-5; discussion 2056 There remains no clear consensus as to the appropriate further investigation and management of the patient suspected of pulmonary embolism (PE) who has an intermediate lung scan. Clinical assessment is documented as unreliable, yet many of these patients are unlikely to be treated or to have further tests despite a 36% chance of having PE. Using Medical Decision Analysis, four management strategies for such patients have been examined in terms of mortality and morbidity up to 6 mo post-presentation. The strategies were: (1) treat all patients; (2) treat no patients; (3) perform pulmonary angiography; and (4) perform contrast venography. In the last two strategies, the patients with positive examinations are treated; those with negative examinations are not treated. An extensive literature review was performed to provide probability estimates of chance events and outcomes. If all patients are treated, there is 96.8% chance of survival, with an 85.8% chance of survival with no major complications. If no patients are treated, survival is 89.3% and complication-free survival is 89.3%. Angiography and venography results were 96.7%, 93.1% and 94.6% and 89.6%, respectively. We conclude that in patients suspected of PE who have intermediate lung scan results, the optimal strategy is pulmonary angiography since this results in the highest survival with the lowest complications. Catania TA Caride VJ Single perfusion defect and pulmonary embolism: angiographic correlation. In: J Nucl Med (1990 Mar) 31(3):296-301 One hundred and thirty-three ventilation-perfusion scans (V-P) with angiographic correlation were retrospectively reviewed to evaluate the frequency of pulmonary emboli (PE) in single perfusion defects (SPD), regardless of ventilation or radiographic findings. By angiography, 15 of 30 SPD cases had PE. Demographic data and clinical presentation were similar for PE and non-PE patients. However, 9 out of 15 patients with PE had recent surgery compared to none of the non- PE patients. SPD were seen in areas of ventilation and chest x-ray abnormalities in 12 of 15 PE and 11 of 19 non-PE cases. Size of the actual lesion was underestimated by scintigraphy in most cases. In 7 of 15 PE cases, the perfusion defect was larger than the corresponding ventilation abnormality. Most SPD were located at the bases. Twelve of 15 SPD in the PE group were at the posterior basilar segment. In the appropriate clinical setting, SPD carries at least a moderate probability for PE. When the clinical suspicion is high, a pulmonary angiography will be needed to confirm the diagnosis. Bernard EJ Nour R Butler SP Quinn RJ Incidence of pulmonary embolism in single segmental mismatch on lung scanning [see comments] In: J Nucl Med (1994 Dec) 35(12):1928-31 Controversy exists as to whether patients with single segmental mismatch (SSM) on a ventilation/perfusion (VQ) lung scan should be given a low or an intermediate probability of pulmonary embolism (PE). METHODS: Pulmonary angiography was used to evaluate the incidence of PE in SSM at the authors' institution. From January 1991 to January 1993, 1449 VQ scans were performed. RESULTS: With modified Biello criteria, 283 were high probability; 628, low probability; 273, normal; and 273, intermediate probability. Of the intermediate probability scans, 61 had SSM. Forty of these patients underwent pulmonary angiography. Twelve patients had PE in the area of the SSM, giving an incidence of PE of 30%. The risk of PE in SSM in the different lung regions was also analyzed. Twenty-three SSM were in the bases of the lung with a 22% incidence of PE; 17 SSM were either in the midzone or apex with a 41% incidence of PE (p = not significant). CONCLUSION: SSM carries a 30% risk of PE. Accordingly, SSM should be given an intermediate probability of PE and not a low probability of PE. Ritsema GH Polder JJ [Ultrasonography of the leg veins is cost-effective and can be the preferred diagnostic study in suspected pulmonary embolism] Bij vermoeden van longembolie kan echografie van de beenvenen onderzoek van eerste keuze zijn omdat het kosteneffectief is. In: Ned Tijdschr Geneeskd (1995 Jul 8) 139(27):1386-90 (Published in Dutch) OBJECTIVE. To determine whether compression ultrasound examination of both legs, to detect deep venous thrombosis, can be the examination of first choice in patients clinically suspected of pulmonary embolism, assuming that the therapy of thrombosis is the same as the therapy of embolism. DESIGN. Prospective study. SETTING. St. Clara Hospital, Rotterdam, the Netherlands. METHODS. 337 consecutive patients (157 women and 180 men, mean age 65 years, SD: 17.4) suspected of pulmonary embolism underwent compression ultrasound examination of the femoral and popliteal veins to detect deep venous thrombosis, and perfusion lung scintigraphy, with ventilation scintigraphy if indicated. Ultrasound examinations and lung scans were interpreted independently and blinded. A financial analysis of the various strategies was made. RESULTS. Deep venous thrombosis was demonstrated ultrasonographically in four (2%) of 208 patients with a normal lung scan, in four (9%) of 43 patients with a non-diagnostic lung scan, and in 30 (35%) of 86 patients with a high probability lung scan. By starting the diagnostic investigation with ultrasound examination 38 lung scans and chest X-rays would be saved at the cost of 294 extra ultrasound examinations. Because ultrasound examinations are relatively inexpensive a cost reduction of 3.4% would be realised, for the Netherlands approximately 1.4 million guilders. In the United States, however, costs would increase because of the high prices of ultrasound examinations. CONCLUSION. Compression ultrasound examination of both legs can be the examination of first choice in the Netherlands in patients suspected of pulmonary embolism, as it is cost effective. Hull RD Raskob GE Carter CJ Coates G Gill GJ Sackett DL Hirsh J Thompson M Pulmonary embolism in outpatients with pleuritic chest pain. In: Arch Intern Med (1988 Apr) 148(4):838-44 Pleuritic chest pain is a frequent complaint in patients coming to the emergency room, but the proportion of such patients with pulmonary embolism is uncertain. In a prospective study, we evaluated the diagnostic outcomes in 173 consecutive patients who came to the emergency room with pleuritic chest pain. Pulmonary embolism, as demonstrated by angiography or autopsy, was present in 36 (21%). The need for objective testing is clearly indicated by our finding that the sensitivity (85%) and specificity (37%) of predetermined clinical variables for pulmonary embolism were insufficient to allow a definitive treatment decision. Optimal sensitivity and specificity are obtained by using pulmonary angiography in combination with lung scanning. The proportion of patients requiring angiography is substantially reduced, from 43% to 26%, without significant loss of accuracy, if ventilation imaging and impedance plethysmography are used together with perfusion scanning. Cvitanic O Marino PL Improved use of arterial blood gas analysis in suspected pulmonary embolism [see comments] In: Chest (1989 Jan) 95(1):48-51 Since hypoxemia is not known to be a sensitive indicator of acute pulmonary embolism, we performed a retrospective study to determine whether an increased P(A-a)O2 gradient or hypocapnia improved the sensitivity of blood gas analysis in acute embolism. The study group consisted of 78 patients with angiographically documented emboli who had blood gas samples obtained while breathing room air. None had a prior history of cardiopulmonary disease. Hypoxemia was present in 59 patients (76 percent), hypoxemia or hypocapnia in 73 patients (93 percent), an increased P(A-a)O2 gradient in 74 patients (95 percent), and an increased P(A-a)O2 gradient or hypocapnia in 77 patients (98 percent). Only one patient with acute embolism showed a normal P(A- a)O2 gradient and normal PaCO2 breathing room air. These results suggest that a normal P(A-a)O2 gradient and a normal PaCO2 obtained in a patient during room air breathing can be used as evidence against the presence of pulmonary emboli. Henschke CI Whalen JP Evaluation of competing diagnostic tests: sequences for the diagnosis of pulmonary embolism, Part II. In: Clin Imaging (1994 Oct-Dec) 18(4):248-54 Tests for pulmonary embolism (PE) and its most frequent source, deep vein thrombosis, include angiography, ventilation-perfusion nuclear medicine scans, and contrast and radionuclide venography. Although selective angiography is the definitive procedure for diagnosis of PE, its risk of death, although small, as well as the morbidity associated with contrast injections, is high enough that alternative, less accurate but safer, diagnostic procedures are preferred. Part I of this report described the methodology for selection of the single test having the lowest effective cost based on the test cost, its risks, and diagnostic performance. Other than the pulmonary angiogram, however, no single test has a sufficiently high diagnostic performance to be clinically realistic: incorrect diagnoses can result in sudden death. Sequences of tests or algorithms can be used to lower the overall risk of the tests to patients while increasing correct decision making to reach a clinically acceptable level. These points are illustrated by comparing three commonly recommended algorithms for the diagnosis of PE. Additionally, the prevalence of PE affects the choice of the algorithm, and we found that no single algorithm is best for all values of prevalence. In summary, we found that the most cost-effective strategy was to select the particular algorithm having the lowest effective cost for the relevant prevalence value. Use of algorithms also decreased the overall risk of diagnostic test complications and the number of incorrect diagnoses. Stein PD, Terrin ML, Hales CA, et al: Clinical, laboratory, roentgenographic, and electrocardiographic findings in patients with acute pulmonary embolism and no pre-existing cardiac or pulmonary disease [see comments]. Chest 1991;100:598-603. Commentary: This article analyzes a subset of the PIOPED study - closer to our subset of interest. It looked only at those without serious prior cardiopulmonary disease. The number enrolled (a few hundred) is thus much smaller than the total enrolled in PIOPED (originally over 800). Many in PIOPED had significant cardiopulmonary disease and are not representative of the patient population to which we refer in the premise above. The section on A-a gradient in this subset shows that the A-a gradients were similar in both groups (those with and those without PE). However, there is some bias here, as those with an abnormal A-a gradient would tend to be enrolled in the study. Obviously, a lot of people without PE but with some other pulmonary process causing an abnormal A-a gradient were enrolled into the PIOPED study. However, we are not interested in whether there is a difference in A-a gradients between the group with PE and the group without PE. We want to know how many of those with a normal A-a gradient had a PE. Looking at the graph on page 601, only 7-8% with PE had an A-a gradient less than 10. An additional 5% with PE had an A-a gradient of 11-20. Thus, at most, 13% of patients with PE will have an A-a gradient less than 20. Unfortunately, they did not analyze this important group with respect to age. For instance, it is possible that all those with PE and an A-a gradient of 11-20 were very young, and thus this A-a gradient represented an abnormal value for them. However, we just don't know. We can guess that at least some of those with an A-a gradient of 15, 16, 17, 18, 19 and 20 were young enough that this represented an abnormal A-a gradient. In all likelihood less than 10% with PE would have had a normal A-a gradient for their age, but we cannot know that for sure unless sub-analysis is done. --Jon Handler, M.D. Stein PD, Goldhaber SZ, Henry JW; Alveolar-arterial oxygen gradient in the assessment of acute pulmonary embolism. Chest 1995;107:139-143. ABSTRACT: PURPOSE: The purpose of this investigation is to evaluate the utility of the alveolar-arterial (A-a) oxygen gradient in the diagnosis of acute pulmonary embolism (PE) among patients who participated in the Prospective Investigation of Pulmonary Embolism Diagnosis (PIO-PED). METHODS: Pulmonary embolism was diagnosed (n = 280) or excluded (n = 499) by angiography in all patients. Patients were then categorized as (1) the entire cohort, (2) no prior cardiopulmonary disease and no prior PE, and (3) no prior PE or deep venous thrombosis. Normal values of the A-a gradient were defined in three ways: (1) values < or = 20 mm Hg; (2) values < or = age/4 + 4; and (3) values based on age from the literature. RESULTS: When a normal A-a gradient was defined as < or = 20 mmHg, 11 to 14% of patients with PE in the three categories of patients had a normal A-a gradient. When the equation age/4 + 4 was used, 8% to 10% of patients with PE in the three categories of patients had a normal A-a gradient. With age-related values from the literature, 20 to 23% of patients with PE in the three categories of patients had a normal A-a gradient. The A-a gradient was normal in comparable percentages of patients who did not have PE. CONCLUSION: Normal values of the A-a gradient did not exclude the diagnosis of acute PE. Commentary: As predicted, if one corrects the A-a gradient for age using the [age/4 + 4] formula, only 8-10% of patients with a normal A-a gradient will have PE in that patient population. CONCLUSION: Clearly, the literature is consistent in that at most 4-8% of PE's will have an A-a gradient of less than 10. It is almost equally consistent in saying that when adjusted for age, the A-a gradient will be normal in 10% or less of patients with documented PE. The age adjustment formula that appears, in the literature, to be most reflective of risk of PE is the 4+ (age/4) formula. In the patient cited, at age 35 years old, the maximum predicted A-a gradient for age is about 13. Therefore, this patient is right at the highest end of "normal" predicted A-a gradient for age. The risk of PE using that fact alone, with no other information, based on the available literature, suggests a PE risk of about 10%. Jonathan A. Handler, MD jah505@nwu.edu Very nice summary Jon. However the conclusion that the risk of PE in this patient is 10% is incorrect. It's not that 10% of patients with normal A-a gradients have PE rather it's that 10% of patients with PE have normal A-a gradients. His a priori probability of PE 'without any further information' would be very small. H. Louzon MD ??? article Try as I might, I can't find the exact citation in my records. The article exists, however, and can be found by looking in the Mayo Clinic Proceedings article quoted below as "Article 3" - it cites this article. This article retrospectively reviews all patients with documented PE on PA-gram over many years at two institutions - the A-a gradient on room air had to be on the chart. They found that only 4/78 patients (5.1%) had documented PE and a normal A-a gradient. Of interest, if one added the requirement of a PaCO2 being more than 36 on top of a normal A-a gradient, only 1/78 patients (1.3%) had PE. They used an age adjustment for what is considered a normal A-a gradient - a scale which correlated closely with the 4 + [age/4] formula. ARTICLE #3: Quoting article #2 above, Rosenow concluded in a very recent Mayo Clinic Proceedings (I believe Jan. 1995) article that in the presence of both a normal A-a gradient and a PaCO2 > 36 one could essentially exclude the possibility of PE.