POINT-OF-CARE GUIDES

Suspected Pulmonary Embolism: Part I. Evidence-Based Clinical Assessment

MARK H. EBELL, M.D., M.S., Athens, Georgia

This guide is one in a series that offers evidence-based tools to assist family physicians in improving their decision-making at the point of care. The series is published in partnership with Family Practice Management. Part II, "Evidence-Based Diagnostic Testing," offers an encounter form for patients with suspected pulmonary embolism and will appear in the next issue of AFP.

Clinical Question

How can the history and physical examination help physicians diagnose pulmonary embolism?

Evidence Summary

Individual signs and symptoms often are not accurate enough to rule in or rule out a disease. However, combining signs and symptoms into a "clinical decision rule" can be much more helpful in making the diagnosis.

A number of investigators have developed and validated clinical decision rules for the diagnosis of pulmonary embolism.^1-6 One of the most carefully tested decision rules, created by Wells and colleagues,³ consists of seven signs and symptoms (Table 1). The strength of the Wells model is that it does not require a chest radiograph or arterial blood gas measurements; instead, it relies on a careful history and physical examination. The model saves time by focusing the assessment on the most important clinical features. A limitation of the Wells model is that the judgment of whether an alternative diagnosis is less likely than pulmonary embolism is not standardized.

TABLE 1 Wells Model for Clinical Diagnosis of Pulmonary Embolism Signs and symptoms Points Clinical signs and symptoms of DVT (leg swelling and pain with palpation of the deep veins) 3.0 PE as likely or more likely than an alternative diagnosis* 3.0 Heart rate >100 beats per minute 1.5 Immobilization (bed rest, except to access the bathroom, for at least 3 consecutive days) or surgery in the previous 4 weeks 1.5 Previous objectively diagnosed DVT or PE 1.5 Hemoptysis 1.0 Malignancy (treatment that is ongoing, within the past 6 months, or palliative) 1.0 Total points: _____ Total points Risk group Number of patients (%) Probability of PE <2 points Low 527 (56.6) 1.3% (95% CI, 0.5% to 2.7%) 2 to 6 points Moderate 339 (36.4) 16.2% (95% CI, 12.5% to 20.6%) >6 points High 64 (6.8) 37.5% (95% CI, 28.7% to 53.7%) DVT = deep venous thrombosis; PE = pulmonary embolism; CI = confidence interval. * -- Based on the history and physical examination, chest radiography, electrocardiogram, and any blood tests that were considered necessary. Adapted with permission from Wells PS, Anderson DR, Rodger M, Ginsberg JS, Kearon C, Gent M, et al. Derivation of a simple clinical model to categorize patients' probability of pulmonary embolism: increasing the model's utility with the simpliRED d-dimer. Thromb Haemost 2000;83:416-20, with additional information from reference 3.

A second decision rule, developed by Wicki and associates,⁴ has been cross-validated only within the sample on which it was based (Table 2). This model does not rely on a judgment of whether an alternative diagnosis is less likely than pulmonary embolism, and it uses information from a chest radiograph and arterial blood gas measurements.

TABLE 2 Wicki Model for the Clinical Diagnosis of Pulmonary Embolism Variable Points Age (years)     60 to 79 1   >= 80 2 Previous PE or DVT 2 Recent surgery 3 Pulse rate >100 beats per minute 1 Paco2 (mm Hg)     < 36 2   36 to 39 1 Pao2 (mm Hg)     <48.8 4   48.8 to 60.0 3   60.1 to 71.3 2   71.4 to 82.6 1 Chest radiograph     Plate-like atelectasis 1   Elevation of hemidiaphragm 1 Total points: _____ Score range Risk group Number of patients (%) Probability of PE 0 to 4 points Low 486 (49.3%) 10% (95% CI, 8% to 13%) 5 to 8 points Moderate 437 (44.3) 38% (95% CI, 34% to 43%) >= 9 points High 63 (6.4%) 81% (95% CI, 69% to 90%) PE = pulmonary embolism; DVT = deep venous thrombosis; Paco2 = partial pressure of arterial carbon dioxide; Pao2 = partial pressure of arterial oxygen; CI = confidence interval. Adapted with permission from Wicki J, Perneger TV, Junod AF, Bounameaux H, Perrier A. Assessing clinical probability of pulmonary embolism in the emergency ward: a simple score. Arch Intern Med 2001;161:96,97.

What role does a clinician's judgment play, without relying on any clinical decision rule? Wicki and associates⁴ assessed the likelihood of pulmonary embolism in 986 patients suspected of having this condition. Of the 368 patients who were placed in a low probability group, 9 percent eventually were diagnosed with pulmonary embolism. Of the 523 patients who were judged to have a moderate probability of pulmonary embolism, 33 percent eventually were diagnosed with the condition. Of the 94 patients who were judged to be at high risk, 66 percent were eventually diagnosed with pulmonary embolism. These results are quite similar to those obtained with the Wicki clinical decision rule. Another study⁷ also found similar results.

When used alone, none of the clinical decision rules is sufficient to rule in or rule out pulmonary embolism. However, stratification of patients into low-, moderate-, and high-risk groups provides an important basis for further evaluation. If the risk assessment based on a clinical decision rule differs from your instinctual clinical assessment, it seems prudent to rely on the assessment that places the patient in the highest risk group. For example, if the Wells rule places a patient in the low-risk group, but you have a higher index of suspicion based on your global assessment of the patient, the patient should be classified as moderate risk.

In the February 1, 2004, issue of American Family Physician, the second part of this "Point-of-Care Guide" will discuss how a clinical estimate of the probability of pulmonary embolism fits into a noninvasive strategy for further evaluation. It also will provide an encounter form for use in assessing the patient with suspected pulmonary embolism.

REFERENCES

1. Hoellerich VL, Wigton RS. Diagnosing pulmonary embolism using clinical findings. Arch Intern Med 1986;146:1699-704.
2. Wells PS, Anderson DR, Rodger M, Ginsberg JS, Kearon C, Gent M, et al. Derivation of a simple clinical model to categorize patients' probability of pulmonary embolism: increasing the model's utility with the SimpliRED d-dimer. Thromb Haemost 2000;83:416-20.
3. Wells PS, Anderson DR, Rodger M, Stiell I, Dreyer JF, Barnes D, et al. Excluding pulmonary embolism at the bedside without diagnostic imaging: management of patients with suspected pulmonary embolism presenting to the emergency department by using a simple clinical model and d-Dimer. Ann Intern Med 2001;135:98-107.
4. Wicki J, Perneger TV, Junod AF, Bounameaux H, Perrier A. Assessing clinical probability of pulmonary embolism in the emergency ward: a simple score. Arch Intern Med 2001;161:92-7.
5. Musset D, Parent F, Meyer G, Maitre S, Girard P, Leroyer C, et al. Diagnostic strategy for patients with suspected pulmonary embolism: a prospective multicentre outcome study. Lancet 2002;360: 1914-20.
6. Miniati M, Prediletto R, Formichi B, Marini C, Di Ricco G, Tonelli L, et al. Accuracy of clinical assessment in the diagnosis of pulmonary embolism. Am J Respir Crit Care Med 1999;159:864-71.
7. Perrier A, Miron MJ, Desmarais S, de Moerloose P, Slosman D, Didier D, et al. Using clinical evaluation and lung scan to rule out suspected pulmonary embolism: Is it a valid option in patients with normal results of lower-limb venous compression ultrasonography? Arch Intern Med 2000;160:512-6.

Applying the Evidence

Mr. Taylor is a 62-year-old man who complains of increasing shortness of breath over the past 24 hours. He also has swelling of the left leg and pain in the left calf on palpation, but no cough, fever, or other symptoms consistent with an alternative diagnosis such as pneumonia. He denies hemoptysis and has no history of recent immobilization, venous thromboembolism, or malignancy. His heart rate is 92 beats per minute. Chest radiography reveals platelike atelectasis; arterial blood gas measurements yield a partial pressure of arterial oxygen of 68 mm Hg and a partial pressure of arterial carbon dioxide of 37 mm Hg.

First, what is your clinical judgment of this patient's probability of having pulmonary embolism: low, moderate, or high? Second, what is an estimation of this patient's probability of having pulmonary embolism using both the Wells and the Wicki models?

Answer: Wells model: 6 points, moderate probability of pulmonary embolism (16 percent). Wicki model: 5 points, moderate probability of pulmonary embolism (38 percent). Note that the patient is near the bottom of the Wicki range and at the top of the Wells range, so the true probability is somewhere between 16 percent and 38 percent.

Mark H. Ebell, M.D., M.S., is in private practice in Athens, Ga., and is associate professor in the Department of Family Practice at Michigan State University College of Human Medicine, East Lansing. He also is deputy editor for evidence-based medicine of American Family Physician.

Address correspondence to Mark H. Ebell, M.D., M.S., 330 Snapfinger Dr., Athens, GA 30605 (e-mail: ebell@msu.edu). Reprints are not available from the author.