2008 Volume 12 Issue 2 Pages 109-117
Although there are numerous screening tests for dysphagia, they mostly focus on aspiration but silent aspiration (SA). We studied the usefulness of a cough test done by inhaling citric acid to screen for SA. Participants in this study consisted of 204 patients (131 men and 73 women) whose average age was 69.90 ± 11.70. They were suspected of having dysphagia and were administered a cough test and VF or VE. They inhaled a mist of 1.0％ citric acid-physiologic saline orally for 1 minute using an ultrasonic nebulizer. A medical assistant observed the number of times each patient coughed：more than five coughs was considered as negative (normal), while less than four coughs was regarded as positive. Using diagnosis of the VF or VE as standards, the sensitivity of the cough test to detect SA was 0.87, the specificity was 0.89, efficiency was 0.89, the positive predictive value (PPV) was 0.74, and the negative predictive value (NPV) was 0.95. We then investigated the usefulness of the cough test by major primary disease. In cerebrovascular disease patients, sensitivity was found to be 0.76, specificity was 0.82, efficiency was 0.79, PPV was 0.73, and NPV was 0.84. In head or neck cancer patients, sensitivity was found to be 1.00, specificity was 0.97, efficiency was 0.98, PPV was 0.93, and NPV was 1.00. In neuromuscular disease patients, sensitivity was found to be 0.83, specificity was 0.84, efficiency was 0.84, PPV was 0.56, and NPV was 0.95. In respiratory disease patients, sensitivity was found to be 0.67, specificity was 0.81, efficiency was 0.76, PPV was 0.67, and NPV was 0.81. In tracheostomy patients, sensitivity was found to be 0.71, specificity was 1.00, efficiency was 0.78, PPV was 1.00, and NPV was 0.50. In dementia patients, sensitivity was found to be 1.00, specificity was 1.00, efficiency was 1.00, PPV was 1.00, and NPV was 1.00. Based on these results, the cough test performed by inhaling citric acid is considered to be a useful tool to screen for SA regardless of disease.