Premium
Performance of screening algorithms in systemic sclerosis‐related pulmonary arterial hypertension: a systematic review
Author(s) -
Thakkar V.,
Stevens W. M.,
Moore O. A.,
Nikpour M.
Publication year - 2013
Publication title -
internal medicine journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.596
H-Index - 70
eISSN - 1445-5994
pISSN - 1444-0903
DOI - 10.1111/imj.12167
Subject(s) - medicine , algorithm , predictive value , population , pulmonary hypertension , medline , predictive value of tests , intensive care medicine , environmental health , computer science , political science , law
Background Pulmonary arterial hypertension ( PAH ) is a major cause of mortality in systemic sclerosis ( SSc ). There is emerging evidence that screening may enable the earlier detection and treatment of SSc ‐ PAH , and thereby improve survival. Aims We undertook a systematic review to evaluate the performance of current screening algorithms in SSc ‐ PAH . Methods We searched the Medline and EMBASE databases to 31 M arch 2012. We selected studies if they applied a screening algorithm to consecutively enrolled SSc patients not known to have PAH ; SSc ‐ PAH had to be confirmed on right heart catheterisation ( RHC ). The performance of each screening algorithm and the methodological quality of each study was evaluated. Results Nine studies met the inclusion criteria with a total intent‐to‐screen population of 3504 participants. In studies of patients with prevalent disease, the positive predictive value ( PPV ) of screening for PAH was 20.4–87.0%. In studies of patients with incident disease, the PPV of screening for PAH was 20.0–30.7%. The PPV of algorithms using echocardiography alone, or in combination with other tests, was comparable. No study enabled an accurate determination of negative predictive value, sensitivity or specificity of the screening algorithm as only patients who screened positive underwent confirmatory testing with RHC . The optimal timing and frequency of repeat screening is unknown. Conclusion The low to moderate PPV of current screening algorithms, coupled with the inability to determine accurately the negative predictive value, sensitivity and specificity, suggests that there is a need to validate further these algorithms before making recommendations regarding screening for SSc ‐ PAH .