Open AccessImproving the detection of pulmonary hypertension in systemic sclerosis using pulmonary function tests
Author(s) -
Schreiber Benjamin E.,
Valerio Christopher J.,
Keir Gregory J.,
Handler Clive,
Wells Athol U.,
Denton Christopher P.,
Coghlan John G.
Publication year - 2011
Publication title -
arthritis & rheumatism
Language(s) - English
Resource type - Journals
eISSN - 1529-0131
pISSN - 0004-3591
DOI - 10.1002/art.30535
Subject(s) - medicine , dlco , cardiology , pulmonary hypertension , pulmonary artery , pulse oximetry , confidence interval , pulmonary function testing , population , diffusing capacity , lung , anesthesia , lung function , environmental health
Objective To construct a readily applicable formula for selecting patients with systemic sclerosis (SSc) for right‐sided heart catheterization (RHC) based on the results of their pulmonary function tests (PFTs). Methods The diagnostic value of PFT variables was quantified in 386 patients with SSc against data obtained from RHC. Results We derived the following formula using data from 257 patients: predicted mPAP = 136 – Sp O 2 – 0.25 × DL CO % predicted, where mPAP is the mean pulmonary artery pressure, Sp O 2 is the oxygen saturation as measured by pulse oximetry, and DL CO is the diffusing capacity for carbon monoxide. We validated the formula in the remaining 129 SSc patients. The area under the curve was 0.75 (95% confidence interval [95% CI] 0.67, 0.84). Using a predicted threshold of 25 mm Hg, the sensitivity was 90.1% (95% CI 82, 96) and the specificity was 29.2% (95% CI 17, 44). When used as a screening procedure in a typical scleroderma patient population, it is projected that those with an mPAP below 25 mm Hg are unlikely to have pulmonary hypertension (PH; prevalence 4.4%), those with a predicted mPAP of 25–35 mm Hg are at average risk of having PH (prevalence of 11.3%), and those with a formula‐predicted mPAP above 35 mm Hg are likely to have PH (prevalence of 62.9%), thus justifying RHC. In patients with equivocal findings on echocardiography, a high formula‐predicted mPAP is strongly associated with the presence of PH. Conclusion We derived and validated an easily applied formula for determining pulmonary function in patients with SSc that identifies subgroups with a low, average, or high prevalence of PH. It provides information that is complementary to echocardiography and that should improve the selection of patients for RHC.