There is paucity of literature concerning the work up of patients with isolated low diffusion capacity (DLCO) on pulmonary function testing (PFT). Our hypothesis was that considerable testing in these patients results in only limited diagnostic yield. This descriptive study investigates the characteristics of these patients and the outcomes of diagnostic testing.
Of 2666 PFTs performed over a 6-year period, 232 (8.7%) had an isolated low DLCO (defined as either <80% predicted with VI/SVC >90% or <70% predicted with VI/SVC >85%). Fifty-three of these patients were evaluated by a staff pulmonologist. Sixteen were excluded as a prior diagnosis explained the low DLCO. Charts of the remaining 37 patients were reviewed and the following data abstracted: age, gender, BMI, history of hypertension, diabetes, obstructive sleep apnea (OSA) and cigarette smoking, and results of chest radiograph (CXR), chest CT scan, echocardiography, ventilation-perfusion (V/Q) scan and serologies for connective tissue disorders.
The mean age was 58 years (range 33-90), and 29 patients (78%) were women. Sixteen patients (43%) had hypertension, 6 (16%) diabetes, 11 (30%) OSA, and 17 (46%) reported a history of smoking. The 10 test abnormalities listed below occurred in 7 patients resulting in the following diagnoses: 1 sarcoidosis, 1 heart failure, 1 bullous disease, 2 interstitial lung disease, and 2 pulmonary artery hypertension. Of interest, 75% of the echocardiograms (12 patients; 32% of patients total) showed left ventricular hypertrophy (LVH). On statistical analysis, neither clinical characteristics nor the number of tests performed increased the likelihood of obtaining a diagnosis (all p values >0.18).
A precise diagnosis explaining an isolated low DLCO was elusive in 81% of our patients, and performing more tests did not increase yield. The finding of OSA, hypertension, and LVH in a significant proportion of our cohort raises questions regarding their relationship, if any, to gas transfer abnormality.
A diagnostic algorithm, evaluated by a prospective trial, should be created for patients with isolated low DLCO to facilitate an efficient and cost-effective work-up. CXRCTEchoV/QSerologyPatients tested35171685Test abnormal16111410Test contributing to an explanation for low DLCO5 (17%)2 (12%)3 (19%)00
J.P. Moralejo, None.