Ultrasonography is helpful in evaluating the biliary tree, looking for ureteral obstruction, and evaluating vascular structures, but has limited utility in obese patients.
c. Computed tomography (CT) is helpful in possible intracranial bleeding, abdominal and/or pelvic masses, and pulmonary processes, and may help to delineate the lymph nodes and retroperitoneal disorders. CT exposes the patient to radiation and requires the patient to be immobilized during the procedure. Generally, CT requires administration of a radiocontrast dye, which can be nephrotoxic.
d. Magnetic resonance imaging (MRI) identifies soft-tissue planes very well and provides the best imaging of the brain parenchyma. When used with gadolinium contrast (which is not nephrotoxic), MR angiography (MRA) is useful for delineating vascular structures. MRI does not use radiation, but the powerful magnetic field prohibits its use in patients with ferromagnetic metal in their bodies, for example, many prosthetic devices.
e. Cardiac procedures:
i. Echocardiography: Uses ultrasonography to delineate the cardiac size, function, ejection fraction, and presence of valvular dysfunction.
ii. Angiography: Radiopaque dye is injected into various vessels and radiographs or fluoroscopic images are used to determine the vascular occlusion, cardiac function, or valvular integrity.
iii. Stress treadmill tests: Individuals at risk for coronary heart disease are monitored for blood pressure, heart rate, chest pain, and electrocardiogram (EKG) while increasing oxygen demands on the heart, such as running on a treadmill. Nuclear medicine imaging of the heart can be added to increase the sensitivity and specificity of the lest. Individuals who cannot run on the treadmill (such as those with severe arthritis), may be given medications such as Persantine to "stress" the heart.
Interpretation of Test Results: Using Pretest Probability and Likelihood Ratio
Because no test is 100% accurate, it is essential when ordering them to have some knowledge of the test's characteristics, as well as how to apply the test results to an individual patient's clinical situation. Let us use the example of a patient with chest pain. The first diagnostic concern of most patients and physicians regarding chest pain is angina pectoris, that is, the pain of myocardial ischemia caused by coronary insufficiency. Distinguishing angina pectoris from other causes of chest pain relies upon two important factors: the clinical history, and an understanding of how to use objective testing. In making the diagnosis of angina pectoris, the clinician must establish whether the pain satisfies the three criteria for typical anginal pain: (a) retrosternal in location, (b) precipitated by exertion, and (c) relieved within minutes by rest or nitroglycerin. Then, the clinician considers other factors, such as patient age and other risk factors, to determine a pretest probability for angina pectoris.
After a pretest probability is estimated by applying some combination of statistical data, epidemiology of the disease, and clinical experience, the next decision is whether and how to use an objective test. A test should only be ordered if the results would change the posttest probability high enough or low enough in either direction that it will affect the decision-making process. For example, a 21-year-old woman with chest pain that is not exertional and not relieved by rest or nitroglycerin has a very low pretest probability of coronary artery disease, and any positive results on a cardiac stress test arc very likely to be false positive. Any test result is unlikely to change her management: thus, the test should not be obtained. Similarly, a 69-year-old diabetic smoker with a recent coronary angioplasty who now has recurrent episodes of typical angina has a very high pretest probability that the pain is a result of myocardial ischemia. One could argue that a negative cardiac stress test is likely to be a falsely negative, and that the clinician should proceed directly to a coronary angiography to assess for a repeat angioplasty. Diagnostic tests, therefore, are usually most useful for those patients in the midranges of pretest probabilities, in whom a positive or negative test will move the clinician past some decision threshold.
In the case of diagnosing a patient with atherosclerotic coronary artery disease (CAD), one test that is frequently used is the exercise treadmill test. Patients are monitored on an electrocardiogram, while they perform graded exercise on a treadmill. A positive test is the development of ST-segment depression during the test; the greater the degree of ST depression, the more useful the test becomes in raising the posttest probability of CAD. In the example below, of a patient with pretest probability of 50%, the development of 2 mm of ST segment depression raised the posttest probability to 90%.
If one knows the sensitivity and specificity of the test used, one can calculate the likelihood ratio of the positive test as sensitivity/1 - specificity. Posttest probability is calculated by multiplying the positive likelihood ratio by the pretest probability, or plot the probabilities using a nomogram (see Figure 1-1).
Was this article helpful?