Michael C. Plewa

Prospective multicenter evaluation of the pulmonary embolism rule-out criteria

Summary.  Backgound: Over‐investigation of low‐risk patients with suspected pulmonary embolism (PE) represents a growing problem. The combination of gestalt estimate of low suspicion for PE, together with the PE rule‐out criteria [PERC(−): age < 50 years, pulse < 100 beats min−1, SaO2 ≥ 95%, no hemoptysis, no estrogen use, no surgery/trauma requiring hospitalization within 4 weeks, no prior venous thromboembolism (VTE), and no unilateral leg swelling], may reduce speculative testing for PE. We hypothesized that low suspicion and PERC(−) would predict a post‐test probability of VTE(+) or death below 2.0%. Methods: We enrolled outpatients with suspected PE in 13 emergency departments. Clinicians completed a 72‐field, web‐based data form at the time of test order. Low suspicion required a gestalt pretest probability estimate of <15%. The main outcome was the composite of image‐proven VTE(+) or death from any cause within 45 days. Results: We enrolled 8138 patients, 85% of whom had a chief complaint of either dyspnea or chest pain. Clinicians reported a low suspicion for PE, together with PERC(−), in 1666 patients (20%). At initial testing and within 45 days, 561 patients (6.9%, 95% confidence interval 6.5–7.6) were VTE(+), and 56 others died. Among the low suspicion and PERC(−) patients, 15 were VTE(+) and one other patient died, yielding a false‐negative rate of 16/1666 (1.0%, 0.6–1.6%). As a diagnostic test, low suspicion and PERC(−) had a sensitivity of 97.4% (95.8–98.5%) and a specificity of 21.9% (21.0–22.9%). Conclusions: The combination of gestalt estimate of low suspicion for PE and PERC(−) reduces the probability of VTE to below 2% in about 20% of outpatients with suspected PE.

Factors Associated With Positive D-dimer Results in Patients Evaluated for Pulmonary Embolism

OBJECTIVES Available D-dimer assays have low specificity and may increase radiographic testing for pulmonary embolism (PE). To help clinicians better target testing, this study sought to quantify the effect of risk factors for a positive quantitative D-dimer in patients evaluated for PE. METHODS This was a prospective, multicenter, observational study. Emergency department (ED) patients evaluated for PE with a quantitative D-dimer were eligible for inclusion. The main outcome of interest was a positive D-dimer. Odds ratio (ORs) and 95% confidence intervals (CIs) were determined by multivariable logistic regression. Adjusted estimates of relative risk were also calculated. RESULTS A total of 4,346 patients had D-dimer testing, of whom 2,930 (67%) were women. A total of 2,500 (57%) were white, 1,474 (34%) were black or African American, 238 (6%) were Hispanic, and 144 (3%) were of other race or ethnicity. The mean (+/-SD) age was 48 (+/-17) years. Overall, 1,903 (44%) D-dimers were positive. Model fit was adequate (c-statistic = 0.739, Hosmer and Lemeshow p-value = 0.13). Significant positive predictors of D-dimer positive included female sex; increasing age; black (vs. white) race; cocaine use; general, limb, or neurologic immobility; hemoptysis; hemodialysis; active malignancy; rheumatoid arthritis; lupus; sickle cell disease; prior venous thromboembolism (VTE; not under treatment); pregnancy and postpartum state; and abdominal, chest, orthopedic, or other surgery. Warfarin use was protective. In contrast, several variables known to be associated with PE were not associated with positive D-dimer results: body mass index (BMI), estrogen use, family history of PE, (inactive) malignancy, thrombophilia, trauma within 4 weeks, travel, and prior VTE (under treatment). CONCLUSIONS Many factors are associated with a positive D-dimer test. The effect of these factors on the usefulness of the test should be considered prior to ordering a D-dimer.

Evaluation of Pulmonary Embolism in the Emergency Department and Consistency With a National Quality Measure: Quantifying the opportunity for improvement

BACKGROUND The National Quality Forum (NQF) has endorsed a performance measure designed to increase imaging efficiency for the evaluation of pulmonary embolism (PE) in the emergency department (ED). To our knowledge, no published data have examined the effect of patient-level predictors on performance. METHODS To quantify the prevalence of avoidable imaging in ED patients with suspected PE, we performed a prospective, multicenter observational study of ED patients evaluated for PE from 2004 through 2007 at 11 US EDs. Adult patients tested for PE were enrolled, with data collected in a standardized instrument. The primary outcome was the proportion of imaging that was potentially avoidable according to the NQF measure. Avoidable imaging was defined as imaging in a patient with low pretest probability for PE, who either did not have a D-dimer test ordered or who had a negative D-dimer test result. We performed subanalyses testing alternative pretest probability cutoffs and imaging definitions on measure performance as well as a secondary analysis to identify factors associated with inappropriate imaging. χ(2) Test was used for bivariate analysis of categorical variables and multivariable logistic regression for the secondary analysis. RESULTS We enrolled 5940 patients, of whom 4113 (69%) had low pretest probability of PE. Imaging was performed in 2238 low-risk patients (38%), of whom 811 had no D-dimer testing, and 394 had negative D-dimer test results. Imaging was avoidable, according to the NQF measure, in 1205 patients (32%; 95% CI, 31%-34%). Avoidable imaging owing to not ordering a D-dimer test was associated with age (odds ratio [OR], 1.15 per decade; 95% CI, 1.10-1.21). Avoidable imaging owing to imaging after a negative D-dimer test result was associated with inactive malignant disease (OR, 1.66; 95% CI, 1.11-2.49). CONCLUSIONS One-third of imaging performed for suspected PE may be categorized as avoidable. Improving adherence to established diagnostic protocols is likely to result in significantly fewer patients receiving unnecessary irradiation and substantial savings.

Clinical features from the history and physical examination that predict the presence or absence of pulmonary embolism in symptomatic emergency department patients: Results of a prospective, multicenter study

STUDY OBJECTIVE Prediction rules for pulmonary embolism use variables explicitly shown to estimate the probability of pulmonary embolism. However, clinicians often use variables that have not been similarly validated, yet are implicitly believed to modify probability of pulmonary embolism. The objective of this study is to measure the predictive value of 13 implicit variables. METHODS Patients were enrolled in a prospective cohort study from 12 centers in the United States; all had an objective test for pulmonary embolism (D-dimer, computed tomographic angiography, or ventilation-perfusion scan). Clinical features including 12 predefined previously validated (explicit) variables and 13 variables not part of existing prediction rules (implicit) were prospectively recorded at presentation. The primary outcome was venous thromboembolism (pulmonary embolism or deep venous thrombosis), diagnosed by imaging up to 45 days after enrollment. Variables with adjusted odds ratios from logistic regression with 95% confidence intervals not crossing unity were considered significant. RESULTS Seven thousand nine hundred forty patients (7.2% venous thromboembolism positive) were enrolled. Mean age was 49 years (standard deviation 17 years) and 67% were female patients. Eight of 13 implicit variables were significantly associated with venous thromboembolism; those with an adjusted odds ratio (OR) greater than 1.5 included non-cancer-related thrombophilia (OR 1.99), pleuritic chest pain (OR 1.53), and family history of venous thromboembolism (OR 1.51). Implicit variables that predicted no venous thromboembolism outcome included substernal chest pain, female sex, and smoking. Nine of 12 explicit variables predicted a positive outcome of venous thromboembolism, including patient history of pulmonary embolism or deep venous thrombosis in the past, unilateral leg swelling, recent surgery, estrogen, hypoxemia, and active malignancy. CONCLUSION In symptomatic outpatients being considered for possible pulmonary embolism, non-cancer-related thrombophilia, pleuritic chest pain, and family history of venous thromboembolism increase probability of pulmonary embolism or deep venous thrombosis. Other variables that are part of existing pretest probability systems were validated as important predictors in this diverse sample of US emergency department patients.

Potential Impact of Adjusting the Threshold of the Quantitative D-dimer Based on Pretest Probability of Acute Pulmonary Embolism

OBJECTIVES The utility of D-dimer testing for suspected pulmonary embolism (PE) can be limited by test specificity. The authors tested if the threshold of the quantitative D-dimer can be varied according to pretest probability (PTP) of PE to increase specificity while maintaining a negative predictive value (NPV) of >99%. METHODS This was a prospective, observational multicenter study of emergency department (ED) patients in the United States. Eligible patients had a diagnostic study ordered to evaluate possible PE. PTP was determined by the clinicians unstructured estimate and the Wells score. Five different D-dimer assays were used. D-dimer test performance was measured using 1) standard thresholds and 2) variable threshold values: twice (for low PTP patients), equal (intermediate PTP patients), or half (high PTP patients) of standard threshold. Venous thromboembolism (VTE) within 45 days required positive imaging plus decision to treat. RESULTS The authors enrolled 7,940 patients tested for PE, and clinicians ordered a quantitative D-dimer for 4,357 (55%) patients who had PTPs distributed as follows: low (74%), moderate (21%), or high (4%). At standard cutoffs, across all PTP strata, quantitative D-dimer testing had a test sensitivity of 94% (95% confidence interval [CI] = 91% to 97%), specificity of 58% (95% CI = 56% to 60%), and NPV of 99.5% (95% CI = 99.1% to 99.7%). If variable cutoffs had been used the overall sensitivity would have been 88% (95% CI = 83% to 92%), specificity 75% (95% CI = 74% to 76%), and NPV 99.1% (95% CI = 98.7% to 99.4%). CONCLUSIONS This large multicenter observational sample demonstrates that emergency medicine clinicians currently order a D-dimer in the majority of patients tested for PE, including a large proportion with intermediate PTP and high PTP. Varying the D-dimers cutoff according to PTP can increase specificity with no measurable decrease in NPV.

12-Lead ECG Findings of Pulmonary Hypertension Occur More Frequently in Emergency Department Patients With Pulmonary Embolism Than in Patients Without Pulmonary Embolism

STUDY OBJECTIVE Acute pulmonary embolism can produce abnormalities on ECG that reflect severity of pulmonary hypertension. Early recognition of these findings may alter the estimated pretest probability of pulmonary embolism and prompt more aggressive treatment before hemodynamic instability ensues, but it is first important to test whether these findings are specific to patients with pulmonary embolism. We hypothesize that ECG findings consistent with pulmonary hypertension would be observed more frequently in patients with pulmonary embolism. METHODS Secondary analysis of a prospective, observational cohort of emergency department patients who were tested for pulmonary embolism. ECGs were ordered at clinicians discretion and interpreted at presentation. RESULTS Six thousand forty-nine patients had an ECG, 354 (5.9%) of whom were diagnosed with pulmonary embolism. The frequency, positive likelihood ratio (LR+) and 95% confidence interval (CI) of each predictor were as follows: S1Q3T3 8.5% with pulmonary embolism versus 3.3% without pulmonary embolism (LR+ 3.7; 95% CI 2.5 to 5.4); nonsinus rhythm, 23.5% versus 16.6% (LR+ 1.4; 95% CI 1.2 to 1.7); inverted T waves in V1 to V2, 14.4% versus 8.1% (LR+ 1.8; 95% CI 1.3 to 2.3); inversion in V1 to V3, 10.5% versus 4.0% (LR+ 2.6; 95% CI 1.9 to 3.6); inversion in V1 to V4, 7.3% versus 2.0% (LR+ 3.7; 95% CI 2.4 to 5.5); incomplete right bundle branch block, 4.8% versus 2.8% (LR+ 1.7; 95% CI 1.0 to 2.7); tachycardia (pulse rate >100 beats/min), 28.8% versus 15.7% (LR+ 1.8; 95% CI 1.5 to 2.2). Likelihood ratios and specificities were similar when patients with previous cardiopulmonary disease were excluded from analysis. CONCLUSION Findings of acute pulmonary hypertension were infrequent overall but were observed more frequently in patients with the final diagnosis of pulmonary embolism compared with patients who do not have pulmonary embolism.

Effects of simulated mouth-to-mouth ventilation during external cardiac compression or active compression-decompression in a swine model of witnessed cardiac arrest.

STUDY OBJECTIVE To assess the effects of simulated mouth-to-mouth (MTM) ventilation on blood gases, gas exchange, and minute ventilation during external cardiac compression (ECC) or active compression-decompression (ACD) in a swine model of witnessed cardiac arrest and bystander CPR. METHODS Twenty swine were anesthetized, intubated, ventilated with room air, and monitored for aortic and right atrial pressure and blood gas sampling. After 1 minute of ventricular fibrillation cardiac arrest, ECC or ACD was manually performed at a rate of 100 per minute for 12 minutes. Animals in the room air group had their endotracheal tubes open to air, whereas those in the MTM group were mechanically ventilated with a gas mixture of 16% oxygen and 4% carbon dioxide. Arterial and venous PO2, PCO2, and pH values; oxygen consumption (VO2); carbon dioxide production (VCO2); and minute ventilation (VE) were measured at baseline and 1, 5, 9, and 13 minutes after induction of cardiac arrest. RESULTS MTM ventilation did not alter arterial or venous PO2 values in comparison with room air but did result in higher arterial PCO2 values at 5 and 9 minutes (although the mean PCO2 was 40 mm Hg or less [5.3 kPa] in all groups) and significant central venous hypercarbic acidosis at 9 and 13 minutes. Arterial PO2 values were greater in the ACD than the ECC groups at 5, 9, and 13 minutes, although all groups maintained acceptable PO2 (mean values > or = 60 mm Hg [8.0 kPa]) through 9 minutes of CPR and through 13 minutes in all but the ECC-room air group. PCO2 values were lower in the ACD groups beyond 1 minute, with the ACD-room air group showing extreme hyperventilation (mean PCO2 < or = 20 mm Hg [2.7 kPa]). MTM ventilation resulted in negative VO2 and VCO2 for the first few minutes, reflecting changes in pulmonary gas stores. As equilibrium was approached, VO2 and VCO2 approached zero in all groups, reflecting low cardiac output. MTM ventilation did not improve VE over room air at any time during ACD. It did improve VE during ECC, but only at the 12th interval. CONCLUSION In this swine model of witnessed CPR, simulated MTM ventilation was not beneficial for blood gases, gas exchange, or ventilation during ECC or ACD CPR.

Independent evaluation of a simple clinical prediction rule to identify right ventricular dysfunction in patients with shortness of breath

BACKGROUND Many patients have unexplained persistent dyspnea after negative computed tomographic pulmonary angiography (CTPA). We hypothesized that many of these patients have isolated right ventricular (RV) dysfunction from treatable causes. We previously derived a clinical decision rule (CDR) for predicting RV dysfunction consisting of persistent dyspnea and normal CTPA, finding that 53% of CDR-positive patients had isolated RV dysfunction. Our goal is to validate this previously derived CDR by measuring the prevalence of RV dysfunction and outcomes in dyspneic emergency department patients. METHODS A secondary analysis of a prospective observational multicenter study that enrolled patients presenting with suspected PE was performed. We included patients with persistent dyspnea, a nonsignificant CTPA, and formal echo performed. Right ventricular dysfunction was defined as RV hypokinesis and/or dilation with or without moderate to severe tricuspid regurgitation. RESULTS A total of 7940 patients were enrolled. Two thousand six hundred sixteen patients were analyzed after excluding patients without persistent dyspnea and those with a significant finding on CTPA. One hundred ninety eight patients had echocardiography performed as standard care. Of those, 19% (95% confidence interval [CI], 14%-25%) and 33% (95% CI, 25%-42%) exhibited RV dysfunction and isolated RV dysfunction, respectively. Patients with isolated RV dysfunction or overload were more likely than those without RV dysfunction to have a return visit to the emergency department within 45 days for the same complaint (39% vs 18%; 95% CI of the difference, 4%-38%). CONCLUSION This simple clinical prediction rule predicted a 33% prevalence of isolated RV dysfunction or overload. Patients with isolated RV dysfunction had higher recidivism rates and a trend toward worse outcomes.

Collaboration among emergency medicine physician researchers and statisticians: resources and attitudes.

We examined the statistical resources within emergency medicine residency programs, and the attitudes of emergency medicine physician researchers toward activities wherein collaboration with a statistician is useful. Anonymous surveys were mailed to 104 emergency medicine physician researchers (1/program). Sixty-four (62%) responses were analyzed. Sixty-seven percent of respondents were their programs research director. Their highest level of statistical training was self-taught/nondegree course work for 88% of respondents. Forty-two percent said they were the person used most often by their program for statistical expertise. One-quarter of programs employed a full-time statistician. Collaboration among researchers and statisticians was considered sometimes or always useful for protocol development (aims 84%, design 99%, outcomes 99%, procedures 73%, sampling 97%, inclusion criteria 93%, number of subjects 100%); data entry 73%; statistical analysis 100%; and manuscript preparation 86%. Although most emergency medicine residencies lacked statistical resources within their program, physician researchers expressed positive attitudes toward collaboration with a statistician for all aspects of research.

A computer model of gas exchange during cardiopulmonary resuscitation

This paper presents a computer model of gas exchange during cardiopulmonary resuscitation (CPR) that permits independent adjustment of inspired air content (16% O2 and 4.5% CO2 present in mouth-to-mouth (MTM) ventilation or ambient air), shunt, deadspace, diffusion impairment, cardiac output, and ventilation. The model contains 15500 acini, each with its own blood supply. Gas exchange occurs at each perfused and ventilated acinus. Arterial P(O2) and P(CO2) are calculated from the summed arterial blood flow using standard formulae. The model and simulations show that MTM ventilation provides inadequate oxygenation when the victim is at high altitude or has diffusion impairment. They also show that analysis of inspired and expired gas concentrations to measure gas exchange primarily measures wash in and wash out of gas when cardiac output is low and that this explains the negative oxygen consumption and carbon dioxide production measured in a previous study.