From the Guidelines
The diagnostic yield of a non-induced sputum sample is generally low, ranging from 20-60% for bacterial infections, and is highly dependent on the quality of the specimen and the suspected pathogen. The yield of non-induced sputum samples can vary significantly depending on the patient factors and the type of infection being diagnosed. For example, the sensitivity of non-induced sputum for diagnosing tuberculosis is approximately 20-30% for a single sample, while the sensitivity for Pneumocystis pneumonia is around 10-30% 1.
Factors Affecting Diagnostic Yield
Several factors can affect the diagnostic yield of non-induced sputum samples, including:
- The quality of the specimen, with optimal samples containing <10 epithelial cells and >25 polymorphonuclear cells per low-power field
- The number of samples collected, with multiple samples improving yield
- The time of collection, with early morning specimens typically providing better results
- The patient's ability to cough forcefully from the lungs rather than clearing the throat
Comparison with Induced Sputum
Induced sputum samples have been shown to have a higher diagnostic yield than non-induced sputum samples, particularly for tuberculosis 1. The diagnostic yield of induced sputum increases with multiple specimens, with detection rates by AFB smear microscopy of 91%–98% and mycobacterial culture of 99%–100% reported when 3 or more specimens are obtained 1.
Clinical Implications
Given the low diagnostic yield of non-induced sputum samples, induced sputum or more invasive procedures like bronchoscopy may be necessary when non-induced samples are inadequate or negative despite high clinical suspicion. The choice of diagnostic test should be guided by the suspected pathogen and the patient's clinical presentation, as well as the availability of resources and the potential risks and benefits of each test 1.
From the Research
Diagnostic Yield of Non-Induced Sputum Sample
- The diagnostic yield of a non-induced sputum sample is not directly addressed in the provided studies, as they primarily focus on induced sputum or compare induced sputum to bronchoalveolar lavage (BAL) or spontaneous sputum 2, 3, 4.
- However, a study comparing diagnostic yield of AFB with sputum induction to spontaneous sputum examination in suspected pulmonary tuberculosis found that among patients who were already expectorating, AFB smear and culture positivity results were comparable between spontaneous and induced sputum sampling 3.
- Another study assessed the value of examining multiple sputum specimens in the diagnosis of pulmonary tuberculosis and found that the first sputum smear examination was positive in 82.3% of smear-positive cases, and either the first or the second sputum was diagnostic in 94.9% of these cases 5.
- The diagnostic yield of non-induced sputum samples can be inferred to be lower than that of induced sputum, as induced sputum has been shown to improve diagnostic yield in patients unable to expectorate adequate sputum samples 3, 4.
- Sputum quality has been found to impact the diagnostic performance of GeneXpert MTB/RIF, with salivary sputum showing a higher sensitivity and potentially higher yield than other sample types 6.
Comparison of Diagnostic Yields
- Induced sputum has been shown to have a higher diagnostic yield than spontaneous sputum in some cases, particularly in patients who are unable to expectorate adequate sputum samples 3, 4.
- BAL has been found to have a higher diagnostic yield than induced sputum in patients with suspected pulmonary tuberculosis and negative sputum smears 2.
- The choice of diagnostic method may depend on the clinical context and availability of resources 2.