Premium
Response to Richard Schreiber: Diagnosing Pulmonary Embolism in Frail Older Adults Out of the Hospital
Author(s) -
Schouten Henrike J.,
Geersing GeertJan,
Oudega Ruud,
Delden Johannes J. M.,
Moons Karel G. M.,
Koek Huiberdina L.
Publication year - 2015
Publication title -
journal of the american geriatrics society
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.992
H-Index - 232
eISSN - 1532-5415
pISSN - 0002-8614
DOI - 10.1111/jgs.13419
Subject(s) - medicine , d dimer , pulmonary embolism , confidence interval , test (biology) , point of care testing , point of care , pediatrics , gerontology , surgery , pathology , paleontology , biology
dimer tests, there would have been at least five more participants with confirmed PE. Recalculating the total number of participants with PE with a low score but an abnormal D-dimer brings the total to 25 of 98 (24.5%). The newly recalculated false-negative rate is 0, and the sensitivity of a Wells score of 4 or less combined with a negative bedside D-dimer rises from 92.2% to 100%. The specificity of an unlikely Wells score but with a positive D-dimer does not change, although the negative predictive value rises from 94.1% to 100%, and the positive predictive value rises from 50.8% to 52.9%. The study protocol called for clinicians to “refrain from referral in all other individuals (unlikely risk: ≤4 points).” Why then did these five individuals have further tests? Furthermore, we do not know whether there were more individuals with low scores and high D-dimers who had PE, nor do we know if all individuals with high scores had PEs. (Only 62% were tested, contrary to the study protocol.) These are not so much “irregularities” as they are protocol violations. All studies suffer from some deviations, but when as many as 40% of participants are not followed per protocol, it is difficult to draw reliable conclusions. Based on these deficiencies in the methodology and follow-up, the only valid conclusions are that the selection of a bedside D-dimer test must be based on vigorous evaluation of its testing characteristics in the individuals on whom it will be used, one must follow a rigorous testing algorithm to ensure that individuals with this potentially fatal disease are detected and that those who do not have the disease are not treated with potentially life-threatening treatment, and clinicians must make vigorous efforts to follow up with these individuals. The authors assert that theirs is the first study that focuses on older adults, but there was one study on elderly adults seen in emergency departments and inpatients in 2012 and another in outpatients in March 2014.