The use of low-dose computed tomography (LDCT) may modify the clinical diagnosis of elderly patients admitted to the hospital for pneumonia and reduce unnecessary antibiotic therapy, according to a study published by the European Respiratory Society.
Researchers conducted a prospective, interventional cohort study of 200 patients older than 65 years (median age, 84) admitted to the hospital with a suspicion of pneumonia and treated with antimicrobial therapy. The patients had both a chest radiograph and a LDCT within 72 hours of study inclusion. The purpose of the study was to determine the clinician’s pneumonia probability estimates both before and after LDCT.
Clinicians were asked to assess the probability of the patient having pneumonia by using the clinical examination, biological data, and chest radiography results, and then grading the probability of pneumonia on a 5-level Likert scale. A new evaluation of the probability of pneumonia was performed by incorporating the results of the LDCT and the radiologist interpretation, and a decision was made by the clinician whether to continue or discontinue antibiotic therapy at that time.
The high, intermediate, and low probability of pneumonia in this cohort of patients was 56.5%, 35%, and 8.5%, respectively, on the initial assessment of data prior to LDCT. Of note, clinical information obtained through LDCT changed the probability of pneumonia in 45% of patients, with probability results upgraded in 15% and downgraded in 30%. Patients in the intermediate probability group had the largest probability change, with 80% of individuals having changed post LDCT. The overall number of patients correctly reclassified was +16, which corresponded to 8% of the patients included in this study.
There were 113 true-positives, 44 true-negative, 13 false-positive, and 30 false-negative cases, leading to accuracy measures of 0.79 for sensitivity, 0.77 for specificity, 0.90 for positive predictive value, 0.59 for negative predictive value, 3.46 for positive likelihood ratio, 0.27 for negative likelihood ratio, and area under the curve (AUC) of 0.79 (95% CI, 0.73-0.86). The clinician’s diagnostic performance with LDCT access had an AUC of a receiver operating characteristic curve of 0.847 (95% CI, 0.7907-0.9032). Overall, clinicians withdrew antibiotic therapy in 8.5% of patients, including 30% of patients with low post-LDCT probability.
Researchers concluded that the use of LDCT did indeed change the probability of a diagnosis of pneumonia in 45% of patients, with an absolute net reclassification index of 8%. LDCT primarily helped clinicians exclude the diagnosis of pneumonia, discontinue the use of unnecessary antibiotics, and could potentially assist clinicians in reducing the overdiagnosis of pneumonia in the future. Therefore, clinicians should consider the use of LDCT in patients older than 65, especially patients classified as having an intermediate probability of pneumonia.
Prendki V, Scheffler M, Huttner B, et al. Low-dose CT for the diagnosis of pneumonia in elderly patients: a prospective interventional cohort study [published online April 12, 2018]. Eur Respir J. doi:10.1183/13993003.02375-2017