Background

Diagnosing pulmonary embolism (PE) represents a challenge in emergency care because it can present with nonspecific symptoms and can be potentially, but not commonly, lethal. However, detecting PE should also be balanced with the potential for over‐testing, over‐diagnosis, and over‐treatment of PE, particularly when a diagnosis of PE is sought in very low‐risk patients.¹ PEs are commonly encountered in the emergency department (ED), with nearly 200,000 ED patients receiving a diagnosis of PE in 2016.² Accurate and timely identification of patients with PE in the ED can minimize complications and morbidity. The challenge in diagnosing PE in the ED is the appropriate selection of patients for diagnostic testing and risk stratification based on clinical findings. The following is not meant to be a comprehensive review of all aspects of the diagnosis of PE in the ED; rather, it is a compilation of clinically relevant questions, and studies providing objective data for specific clinical questions.

There are multiple tests for PE, including D‐dimer (enzyme‐linked immunosorbent assay [ELISA] or whole‐blood assay), computed tomographic pulmonary angiogram (CTPA), ventilation–perfusion (V/Q) scan, and pulmonary angiogram. It is important in the case of suspected PE to choose testing strategies that employ blood tests first such as D‐dimer in lower‐risk patients rather than starting with high‐radiation costly tests such as CTPA or V/Q, when D‐dimer is appropriate. Choice of a test traditionally focuses on assessment of pretest probability of the disease based on objective clinical criteria where lower‐risk patients can be risk‐stratified using D‐dimer, or even no testing, to exclude PE and higher risk patients traditionally receive a chest CTPA or V/Q scan. Because pulmonary angiogram, long considered the criterion standard test for PE, has a 1.5% incidence of serious complication just from receiving the test, it is rarely used to make a diagnosis and/or guide patient management.

Clinical question

In patients with suspected acute PE, can a risk stratification tool be used to identify patients who need no further testing for PE?

While D‐dimer can be helpful in ruling out PE without imaging if applied to low or intermediate‐risk patients, overuse of the D‐dimer to screen for possible PE can have negative consequences because of its low specificity. Using D‐dimer in high‐risk patients should be avoided because it is not sensitive enough to completely rule out PE. As patients with elevated D‐dimers commonly undergo CTPA to rule out PE,³ and the number of negative CTPAs for PE is high (90–95% in the United States), selecting which patients need any testing to rule out PE is clinically helpful.^4,5 Kline et al.⁶ aimed to derive and test clinical criteria for patients that did not require D‐dimer testing, creating a prediction rule, called the pulmonary embolism rule out criteria (PERC). The authors sought to find variables that were associated with the absence of PE. In the derivation study, the prevalence of PE was 11%. The final model that was derived consisted of eight criteria (Table 50.1). If all are negative, the patient does not require any testing for PE and PE can be ruled out on clinical grounds. In other words, a patient lacking any of the PERC criteria is so low‐risk for PE that they do not require a D‐dimer or further imaging to reduce the risk further and to continue to evaluate these very low‐risk patients further will ultimately harm more patients than will be helped by over‐testing as explained in Chapter 1 with test‐treatment thresholds.

Clinical question

What is the Wells criteria for pretest probability of PE?

The initial derivation of the Wells criteria for PE involved the development of a scoring system to calculate pretest probability in patients with suspected PE.¹⁰ The authors used a randomly selected sample of 80% of the patients who had participated in a prospective cohort study of patients with suspected PE who also received D‐dimer testing (SimpliRED), and performed logistic regression analysis using 40 clinical variables to create a clinical prediction rule. Seven variables ended up being predictive of PE and were termed the Wells criteria for PE (Table 50.2). They created cut points for the new rule and two separate scoring systems; one that classifies patients as having low, intermediate, or high probability of PE, and a second that dichotomizes the Wells criteria into PE likely and PE unlikely. The goal of the dichotomous scoring was to simplify the risk stratification process and include a subgroup of patients who were PE unlikely and had a negative D‐dimer, and therefore would have a PE prevalence of less than 2%. Notably, studies have determined that the testing threshold for PE is in the approximately 1–2% prevalence range.^11,12 For more information on test‐treatment thresholds, see Chapter 1. The authors then applied these probabilities to the remaining 20% of the sample (to validate the scoring system).

A validation study of the dichotomized Wells score in 3306 patients found that when a normal quantitative D‐dimer (VIDAS or Tinaquant) was combined with a PE unlikely probability, the 3‐month incidence of VTE was 0.5% (95% CI 0.2–1.1). The authors concluded that a combination of a score ≤4 and a negative D‐dimer may have a negative predictive value where safe discharge is possible in patients with suspected PE.¹³ While a negative D‐dimer combined with low pretest probability or PE Unlikely Wells score has been demonstrated to be safe, Warren et al

Related posts:

Adult Septic Arthritis Testicular Torsion Palpitations Penetrating Trauma to the Extremities and Vascular Injuries First‐Episode Seizure Small Bowel Obstruction