PURPOSE: To determine if vascular pedicle width (VPW), vascular pedicle ratio (VPR), or cardiothoracic ratio (CTR) on portable chest radiographs (CXR), accurately differentiate congestive heart failure (CHF) from others causes of dyspnea, and if any of these measurements have any additive value to serum B-type natriuretic peptide (BNP) levels.
METHODS: Retrospective chart review of patients presenting to the emergency department with dyspnea. Two radiologists independently reviewed the CXRs and obtained measurements of VPW, VPR and CTR. VPR was calculated by dividing VPW by the diameter of the thorax. Areas under the receiver operating characteristic curve (AUC) were computed from logistic regression models.
RESULTS: Seventy-eight patients were analyzed. The mean age was 68.5 (SD 13.2), 65.5% were women and 84.6% were African-American. 57.7% had the diagnosis of CHF. VPW and VPR could not distinguish patients with CHF from patients with other causes of dyspnea (p=0.37 and 0.128 respectively). CTR alone was a good method of distinguishing patients with CHF from patients with other causes of dyspnea (AUC=78%, 95%CI, 67% to 90%, p<0.001). LogBNP predicted CHF with AUC=90% (p<0.001). When logBNP and CTR were analyzed together, CTR did not add any statistically significant predictive accuracy to a model using BNP (p=0.07).
CONCLUSION: VPW and VPR do not appear to differentiate CHF from other causes of dyspnea in this study group. CTR is a good independent predictor of CHF. CTR does not add any statistically significant predictive value to BNP measurement.
CLINICAL IMPLICATIONS: CXR measurements are readily available, inexpensive, and non-invasive. While VPW and VPR cannot be relied upon to differentiate the causes of dyspnea in patients presenting to the emergency room, CTR may provide useful diagnostic information.
DISCLOSURE: Jair Munoz Mendoza, None.