Indication and timing of trough plasma-voriconazole (VCZ)-concentration (t-PVC) measurement during VCZ treatment is a debated issue. Patterns of t-PVC were prospectively evaluated in pediatric (50 courses) and adult (95 courses) hematologic patients. Efficacy patterns were defined: adequate, t-PVC always ≥1 mcg/ml; borderline, at least one t-PVC measurement <1 mcg/ml but median value of the measurements ≥1 mcg/ml; inadequate, median value of the measurements <1 mcg/ml. Toxicity patterns were defined: favorable, t-PVC always ≤5 mcg/ml; borderline, one or more t-PVC measurements >5 mcg/ml but median value of the measurements ≤5 mcg/ml; unfavorable, median value of the measurements >5 mcg/ml. In children and adults the mean t-PVCs were higher during intravenous treatments. The t-PVC efficacy pattern was adequate, borderline and inadequate in 48%, 12%, and 40% of courses, respectively, in children, and in 66.3%, 16.8%, and 16.8% of courses, respectively, in adults. Adequate efficacy pattern was more frequent in children with body weight above the median (≥25 kg) (OR 4.8; P = .011) and in adults with active hematological disease receiving intravenous therapy (OR 3.93; P = .006). Favorable toxicity pattern was more frequent in children receiving VCZ daily dosage below the median (<14 mg/kg) (OR 4.18; P = .027) and in adults with body weight below the median (<68 kg) (OR 0.22; P = .004). T-PVC measurement is generally needed, however, a non t-PVC guided approach may be considered in heavier adults receiving intravenous VCZ. The risk of supratherapeutic levels does not seem an absolute indication for t-PVC monitoring.