The measuring of substantia nigra hyperechogenicity in an Italian cohort of Parkinson disease patients: a case/control study (NOBIS Study).

ABSTRACT

Transcranial sonography (TCS) shows an increased echogenic area of the substantia nigra (SN) in patients with Parkinson disease (PD). It has been increasingly used in the diagnosis of PD and its differentiation from atypical parkinsonian syndromes. Here, we studied the diagnostic accuracy of SN TCS in Italian patients. In this blinded cross-sectional study (NOBIS study), two expert neuro-sonologists performed TCS in 25 PD patients and 29 age- and sex-matched controls. The study participants were completely hidden to the TCS investigators using large drapery. One month later, the SN TCS recordings were re-read by the initial investigator, and cross-read by the second reader. Diagnostic accuracy was estimated on the first reading, intra-reader reliability on re-reading, and inter-reader reliability on cross-readings. The mean SN echogenic area was larger in the patients (0.24 cm2) than in the controls (0.15 cm2; Mann-Whitney test, p < 0.001). SN measures did not differ between right and left, or between ipsilateral and contralateral to the clinically more affected side. There was no correlation between SN echogenicity and PD severity or duration. High intra-reader (concordance correlation coefficient 0.93) and inter-reader (0.98) agreement of SN measurements was found. The diagnostic accuracy for the detection of PD was high (area under receiver-operating characteristic curve 0.91; 95% CI 0.83-1.00) with an optimum cut-off value for SN echogenic area of 0.18 cm2 with the device used here (specificity 0.83-0.90; sensitivity 0.72-0.92). This study supports the use of SN TCS in the diagnostic workup of PD if performed by trained readers.