Table 1. The upper and lower tables present mean kinetic estimates of the voxel clusters identified by parametric mapping analysis.

Lt CDH	Controls(n = 16)	Detoxified Alcoholics(n = 12)	B (withACQ)	B (cigarettes/day)
K	0.0170±0.0032	0.0153±0.0039	n.s.	n.s.
kloss	0.0078±0.0023	0.0123±0.0039*** (+58%)	n.s.	n.s.
Vd	2.36±0.77	1.36±0.54** (−43%)	n.s.	n.s.
Lt VST	Controls (n = 16)	Detoxified Alcoholics (n = 12)	B (with ACQ)	B (cigarettes /day)
K	0.0235±0.0045	0.0213±0.0045	n.s.	n.s.
kloss	0.0082±0.0022	0.0071±0.0018	0.773† (R: 0.725)	n.s.
Vd	3.00±0.88	3.30±1.17	−0.835†† (R: 0.782)	n.s.

The cluster of left medial caudate head (Lt CDH) was identified in the contrast between the alcoholic and healthy groups, whereas the left ventral striatum cluster (Lt VST) was identified by voxel-wise multivariate regression analysis with the individual ACQ scores of the patient group, partialling out the effects of cigarette consumption (numbers of cigarettes smoked per day as a nuisance covariate). The kinetic parameters are the blood-brain FDOPA clearance corrected for loss of trapped metabolites (K, ml g⁻¹ min⁻¹), the fractional rate constant for elimination of [¹⁸F]-fluorodopamine formed in brain (k_loss, min⁻¹), and steady-state storage of FDOPA and its decarboxylated metabolites (V_d, ml g⁻¹), as calculated using the linear solution of Equation 8 in [17]. Each estimate is the mean ± SD of 16 normal controls or 12 patients. P values were calculated using GLM. Significance of difference between alcoholic and control groups: (**) P<0.005; (***) P<0.001. Standardized partial regression coefficients (β) and multiple correlation coefficients (R) are presented for significant correlation between kinetic results in patients and ACQ scores: (^†) P<0.05; (^††) P<0.005. Results uncorrected for multiple comparisons.