NHANES III equations enhance early detection and mortality prediction of bronchiolitis obliterans syndrome after hematopoietic SCT

Abstract

Bronchiolitis obliterans syndrome (BOS) is a serious complication of chronic GVHD (cGVHD) following HSCT (hematopoietic SCT). The clinical diagnosis of BOS is based on pulmonary function test (PFT) abnormalities including: FEV1<75% predicted and obstructive FEV1/VC ratio, calculated using reference equations. We sought to determine if the frequency of clinical diagnoses and severity of BOS would be altered by using the recommended NHANES III vs older equations (Morris/Goldman/Bates, MGB) in 166 cGVHD patients, median age 48 (range: 12-67). We found that NHANES III equations significantly increased the prevalence of BOS, with an additional 11% (18/166) meeting diagnostic criteria by revealing low FEV1 (<75%) (P<0.0001), and six additional patients by obstructive ratio (vs MBG). Collectively, this led to an increase of BOS incidence from 17 (29/166) to 29% (41/166). For patients with severe BOS, (FEV1<35%), NHANES III equations correctly predicted death 71.4% vs 50% using MGB. In conclusion, the use of NHANES III equations markedly increases the proportion of cases meeting diagnostic criteria for BOS and improves prediction of survival.

Conflict of interest statement

Conflict of Interest: The authors declare no conflict of interest.

Figures

Figure 1

FEV1 and FEV1/FVC ratio by reference equations. The number of subjects discordant for FEV1 and FEV1/FVC between MGB and NHANES III are shown. The FEV1 discordance was significant (with all subjects previously categorized as normal by MGB and abnormal by NHANES III.

Figure 2

Percent of cohort diagnosed with BOS by reference equations and diagnostic criteria. This graph demonstrates that NHANES III captures more patients with BOS regardless of the clinical diagnostic criteria, including (1) initial BOS definitions: FEV1 120%, FEV1/FVC

Figure 3

Severity of patients who ‘newly’ met BOS diagnostic criteria using NHANES III equations. As compared with the FEV1% predicted of the total cohort (by NHANES III), those who ‘newly’ met BOS diagnostic criteria due to FEV1 decline (to

Figure 4

Mortality of BOS patients with respect to severity by FEV1% predicted. On the Y axis is the number of patients who died and on the X axis, the BOS cohort is divided into two main groups: (1)the older definitions including those with FEV1 <75% predicted and a ratio of FEV1/FVC <0.7 threshold for identification of BOS and(2)newer definitions including those with FEV1 <75% predicted and a ratio with FEV1/VC <0.7 or the ratio met criteria for obstruction with less than fifth percentile. For all cause mortality on this study, NHANES III equations better categorized these patients as very severe (FEV1<35%) using either the older or newer diagnostic criteria for BOS (with the largest proportion of severe BOS patients (black) defined by NHANES III).