Work

NHANES III Equations for PFT Interpretation Significantly Alters BOS Diagnoses, Severity, and Prognosis After Hematopoietic Stem Cell Transplantation (HSCT)

Year: 2011

Type: article

Abstract: Bronchiolitis obliterans syndrome (BOS) is a deadly manifestation of chronic graft-versus-host disease (cGVHD) following HSCT. The proposed BOS definition includes pulmonary function test (PFT) parameters of: FEV1 < 75% predicted, FEV1/VC ratio < 0.7, (JAMA 2009) that are determined by reference equations. NHANES III is the recommended standard, however, older equations are still used. We sought to determine if the clinical diagnosis and prognosis of BOS could be predicted by FEV1 and whether the reference equations influenced these results in a cohort of cGVHD patients on a NIH protocol. FEV1% predicted and FEV1/FVC for 166 consecutive patients were compared using both NHANES III and our prior reference equations (Morris/Goldman/Bates-MGB). The resulting 4 groups were assessed for degree of agreement using McNemar's exact test for paired categorical data. Mortality was determined using clinical records and SSDI. Assessing only FEV1% predicted, 11% (18/166) of patients were classified as normal by MGB(> 75%), yet abnormal by NHANES III(< 75%) while 0/166 were classified as abnormal by MGB(> 75%), yet normal by NHANES III(< 75%) (p < 0.0001). Evaluating only FEV1/FVC ratios, 95% (157/166) were concordant by confidence interval. The discordant ratios were divided evenly between > 0.7 (n = 4) and < 0.7 (n = 5) (p = 1.00). Using FEV1/VC ratios outside the CI identified 6 more patients who then met criteria for BOS, while 1 was removed because MGB ratio of < 0.7 was normal by NHANES III. Using the amended BOS definition with NHANES III classification, 12 additional patients (29%) would have been diagnosed with BOS, and 1 would have been removed from the BOS cohort. NHANES III better predicted oxygen requirement: 7/8 with FEV1 < 40% by NHANES III and 5/8 FEV1 < 40% by MGB. FEV1 and 2-minute walk velocity association calculated by either method were weak and comparable (MGB (r = 0.25) or NHANES III (r = 0.24)). For patients with > 1 year of follow-up, 35% of them died (14/40). FEV1 < 35% was a significant predictor of mortality with either equation (OR = 20.7, p = 0.001 for NHANES III vs. 18.0, p = 0.01 for MGB equation). However, the NHANES III equation correctly predicted death in 9 patients (64%) compared with 6 (42.8%) patients when MGB equation was used. In conclusion, using NHANES III reference equations markedly increased the number of cases meeting diagnostic criteria for BOS and BOS cases categorized using NHANES III improved prediction of oxygen requirement and survival.

Source: Transplantation and Cellular Therapy

DOI: 10.1016/j.bbmt.2010.12.531