OBJECTIVE Using multidetector computed tomography, we compared the utility of a spilt bolus technique and the impact of performing delayed imaging compared with immediate imaging on tumor enhancement. METHODS Thirty-seven patients with masses of the extracranial head and neck were imaged. An initial contrast bolus (50 cm) was administered over 30 seconds (1.67 cm/s) and an initial (immediate) imaging acquisition followed. Ninety seconds after completion of the initial bolus, a second bolus of 75 cm (2.5 cm/s) was administered and a second (delayed) acquisition obtained concurrently. Mean tumor enhancement in immediate and delayed imaging was calculated and analyzed using the paired t test. RESULTS Mean enhancement was 68.3 +/- 21.0 Hounsfield units (HU) and 91.4 +/- 27.4 HU (n = 37) for immediate and delayed imaging groups, respectively. The mean difference between the initial and delayed enhancement was 23.1 HU (P < 0.01). CONCLUSIONS The delayed imaging technique used resulted in a statistically significant increase in enhancement relative to immediate imaging.