Ten basic principles for optimizing the diagnostic yield of noninvasive diagnostic tests for patients with hematological malignancies and ARF
| Do not perform all tests in every patient. Instead, rely on the pretest probability (DIRECT approach) to select tests with either high sensitivity for confirming a strongly suspected diagnosis or high negative predictive value for ruling out a diagnosis that has a low pretest probability. |
| Identify the diagnosis. Every patient must have a diagnostic workup, which may involve a single diagnostic test when the diagnosis is obvious (echocardiography, pleural drainage, and sputum test). |
| Avoid cognitive bias. Do not be satisfied with a high pretest probability. Document the diagnosis using rapid diagnostic tests or invasive tests. Consider every possible differential diagnosis. |
| In patients with hypoxemic ARF, give priority to noninvasive diagnostic tests. When a diagnosis can be made noninvasively, do not perform bronchoscopic-bronchoalveolar lavage |
| Do not confuse a positive test with a diagnostic test. PCR can detect tiny amounts of viral DNA, whose presence does not confirm pneumonia due to the corresponding virus. Identified pathogens may be commensals or colonizing pathogens. |
| A negative test can contribute substantially to the diagnosis. For a diagnosis with a low pretest probability, a test with a high negative predictive value can allow withdrawal of a treatment, thereby decreasing toxicity and cost. |
| The best diagnostic test assesses the lung compartment predominantly affected by the disease. The diagnostic yield of bronchoscopic-bronchoalveolar lavage is high in patients with ground-glass opacities. These opacities indicate bronchoscopic-bronchoalveolar lavage, unless the patient is considered too unstable or too hypoxemic or a noninvasive diagnostic test can achieve the same diagnostic yield. |
| When bronchoscopic-bronchoalveolar lavage is deemed mandatory for a patient with severe hypoxemia, early intubation that allows for the investigation is acceptable. Early intubation is an option when the patient is responding poorly, to avoid delaying necessary bronchoscopic-bronchoalveolar lavage. In these patients who are severely ill, the adverse effect on outcomes of failure to identify the cause of ARF deserves close consideration. |
| The risk-to-benefit ratio of each procedure must be established by consensus. All alternatives and innovative options should be considered. |
| Diagnostic strategies can be more effective than diagnostic tests. Integrating clinical data with a series of efficient tests can substantially increase diagnostic yields. |
| Do not perform all tests in every patient. Instead, rely on the pretest probability (DIRECT approach) to select tests with either high sensitivity for confirming a strongly suspected diagnosis or high negative predictive value for ruling out a diagnosis that has a low pretest probability. |
| Identify the diagnosis. Every patient must have a diagnostic workup, which may involve a single diagnostic test when the diagnosis is obvious (echocardiography, pleural drainage, and sputum test). |
| Avoid cognitive bias. Do not be satisfied with a high pretest probability. Document the diagnosis using rapid diagnostic tests or invasive tests. Consider every possible differential diagnosis. |
| In patients with hypoxemic ARF, give priority to noninvasive diagnostic tests. When a diagnosis can be made noninvasively, do not perform bronchoscopic-bronchoalveolar lavage |
| Do not confuse a positive test with a diagnostic test. PCR can detect tiny amounts of viral DNA, whose presence does not confirm pneumonia due to the corresponding virus. Identified pathogens may be commensals or colonizing pathogens. |
| A negative test can contribute substantially to the diagnosis. For a diagnosis with a low pretest probability, a test with a high negative predictive value can allow withdrawal of a treatment, thereby decreasing toxicity and cost. |
| The best diagnostic test assesses the lung compartment predominantly affected by the disease. The diagnostic yield of bronchoscopic-bronchoalveolar lavage is high in patients with ground-glass opacities. These opacities indicate bronchoscopic-bronchoalveolar lavage, unless the patient is considered too unstable or too hypoxemic or a noninvasive diagnostic test can achieve the same diagnostic yield. |
| When bronchoscopic-bronchoalveolar lavage is deemed mandatory for a patient with severe hypoxemia, early intubation that allows for the investigation is acceptable. Early intubation is an option when the patient is responding poorly, to avoid delaying necessary bronchoscopic-bronchoalveolar lavage. In these patients who are severely ill, the adverse effect on outcomes of failure to identify the cause of ARF deserves close consideration. |
| The risk-to-benefit ratio of each procedure must be established by consensus. All alternatives and innovative options should be considered. |
| Diagnostic strategies can be more effective than diagnostic tests. Integrating clinical data with a series of efficient tests can substantially increase diagnostic yields. |