Format

Send to

Choose Destination
BMC Neurol. 2015 May 15;15:83. doi: 10.1186/s12883-015-0338-9.

Accuracy and cutoff values of delayed heart to mediastinum ratio with (123)I-metaiodobenzylguanidine cardiac scintigraphy for Lewy body disease diagnoses.

Author information

1
Department of Neurology, University Hospital of Caen, Avenue Côte de Nacre, 14033, Caen, Basse Normandie, France. guillaumelamotte14@gmail.com.
2
Department of Statistics and Clinical Research, University Hospital of Caen, Caen, France. morello-r@chu-caen.fr.
3
Department of Nuclear Medicine, University Hospital of Caen, Caen, EA 4650, France. adrien.lebasnier@gmail.com.
4
Department of Nuclear Medicine, University Hospital of Caen, Caen, EA 4650, France. agostini-de@chu-caen.fr.
5
Department of Neurology, University Hospital of Caen, Avenue Côte de Nacre, 14033, Caen, Basse Normandie, France. defer-g@chu-caen.fr.
6
Unité INSERM U 919, Sérine protéase et physiopathologie de l'Unité Neurovasculaire, GIP Cycéron, Université Caen Basse-Normandie, Caen, France. defer-g@chu-caen.fr.

Abstract

BACKGROUND:

Different studies have found diminished cardiac metaiodobenzylguanidine (MIBG) uptake in Lewy body (LB) related conditions (Parkinson's disease (PD) and Lewy body dementia (LBD)). However, delayed heart/mediastinum (d-H/M) ratio diagnostic cutoff points are debated in parkinsonian syndromes.

METHODS:

We performed a monocentric retrospective analysis on 62 consecutive parkinsonian patients who underwent an (123)I-MIBG scintigraphy, brain imaging and dopaminergic imaging using (123)I-Ioflupane single photon emission computed tomography (SPECT) from 2009 to 2013. The optimal d-H/M ratio was determined from a Receiver Operating Characteristic (ROC) curve and the sensitivity (Se), specificity (Sp) and likelihood ratios (LR) were calculated. 42 patients were diagnosed with LB diseases (20 PD, 22 LBD) and 20 patients with other diseases.

RESULTS:

(123)I-MIBG scintigraphy helped to distinguish PD (p < 0.001) and LBD (p = 0.03) from other diseases. The optimal d-H/M ratio was 1.48 (0.85 area under the ROC curve). Se and Sp were 83.3 %, and 85 % respectively with positive and negative LR of 5.5 and 0.2 respectively. Patients with LBD had a lower d-H/M ratio than patients with PD (result not statistically significant) and a cutoff point at 1.2 could help to differentiate the two diseases. We did not find any correlation between the d-H/M ratio and clinical or (123)I-Ioflupane SPECT data.

CONCLUSION:

According to our population, the d-H/M ratio at 1.48 led to the best performance diagnosis with good Se, Sp and accuracy. In addition, a d-H/M ratio cutoff at 1.2 could help to differentiate PD from LBD.

PMID:
25971430
PMCID:
PMC4448316
DOI:
10.1186/s12883-015-0338-9
[Indexed for MEDLINE]
Free PMC Article

Supplemental Content

Full text links

Icon for BioMed Central Icon for PubMed Central
Loading ...
Support Center