Combining markers increases specificity, reduces sensitivity for PD diagnosis
medwireNews: Using both enlarged substantia nigra hyperechogenicity (SN+) and hyposmia for diagnosing Parkinson’s disease (PD) greatly enhances the specificity, research shows, but lowers the sensitivity compared with either marker on its own.
Researcher Christine Klein (University of Luebeck, Germany) and colleagues explain that although SN+ and hyposmia individually have good sensitivity and specificity for PD diagnosis, so far they have failed to yield reliable predictions, prompting them to investigate the two markers together.
The majority of participants, aged 50 to 79 years, were recruited from a population-based cohort. In all, 106 and 73 had clinically diagnosed PD and possible prediagnostic PD (ppPD), respectively, while 253 were diseased controls, defined as individuals with motor impairments atypical of PD, and 283 were healthy controls.
SN+ or hyposmia, as assessed by transcranial sonography and the Brief Smell Identification Test, respectively, were present in 93.6%, 61.8%, 33.0% and 32.6% of participants in the PD, ppPD, diseased and healthy control groups, respectively. And the concomitant presence of both markers was observed in 51.1% of PD patients and 7.3% of ppPD participants versus 2.1% of diseased controls and 1.3% of healthy controls.
SN+ alone had a sensitivity of 76.6% and a specificity of 86.5% for PD diagnosis, while for hyposmia, the sensitivity and specificity were 68.1% and 74.9%, respectively. When both SN+ and hyposmia were taken together, the sensitivity was lower than for either marker singly, at 51.1%, but the specificity increased to 97.7%.
The team also notes that the combination of SN+ and hyposmia had a positive predictive value (PPV) of 17.6%, which was “strikingly” more than three times the PPV of either marker alone, at 5.2% for SN+ and 2.5% for hyposmia.
“Both SN+ and hyposmia offer good enrichment towards PD and ppPD, are stable against other diseases, and the combination of markers highly increases specificity”, Klein et al write in Movement Disorders.
They caution, however, that using combinations of markers can increase the number of false–negative results, highlighting that 48% of PD and 94% of ppPD cases in their cohort would have been missed if positive results for both tests had been a requirement of diagnosis.
The researchers therefore conclude that “for diagnostic purposes, additional markers need to be identified, including those of disease progression.”
medwireNews is an independent clinical news service provided by Springer Healthcare Limited. © Springer Healthcare Ltd; 2015