Diagnosis of laryngotracheal stenosis from routine pulmonary physiology using the expiratory disproportion index

The study's objective was to determine the utility of expiratory disproportion index (EDI), the ratio of forced expiratory volume in 1 second (FEV1) to peak expiratory flow rate (PEFR) (EDI = FEV1[L] /PEFR[L/s] × 100), in differentiating between laryngotracheal stenosis (LTS) and other respiratory diagnoses. LTS is an uncommon complication of mechanical ventilation or vasculitis or a manifestation of airway compression or malignancy. It frequently masquerades as asthma and evades timely diagnosis, causing prolonged morbidity and airway-related mortality.

Observational study.

We compared spirometry results of 9,357 healthy subjects and nonstenosis pulmonary patients with 217 cases of LTS. Bootstrap analysis, receiver-operating characteristic (ROC) statistics, and Pearson correlation were used to assess the diagnostic utility of the EDI and its correlation with stenosis severity.

Mean EDI values were 36 ± 7 in nonstenosis cases, 76 ± 17 in benign stenoses, and 69 ± 23 in tracheal cancer (P < .0001). A significant correlation existed between anatomic stenosis severity and EDI (P < .0001; R = 0.61). Area under the ROC curve was 0.98, and at a threshold of >50, EDI had a sensitivity of 95.9% and a specificity of 94.2% in differentiating between stenosis and nonstenosis cases.

EDI can reliably diagnose LTS using routine lung function data. Its simplicity and clinical utility, first recognized by Duncan Empey, are underpinned by a unique physiology whereby PEFR, being determined by total tracheobronchial tree resistance, falls disproportionately compared with FEV1 , which is determined within small intrathoracic airways. EDI provides valuable information about the presence and extent of LTS particularly in nonspecialist clinical settings and its routine inclusion within standard lung function reports could prevent the prolonged morbidity and mortality that currently result from missed and delayed diagnoses.