OBJECTIVE It is generally accepted that the intracranial compartment behaves as a unicameral space in which intracranial pressure (ICP) is uniformly distributed. However, this concept has been challenged many times. Although there is general agreement on the existence of craniospinal and suprainfratentorial gradients, the existence of interhemispheric gradients is still a matter of debate. The object of this study was to reexamine the issue of interhemispheric supratentorial ICP gradients in patients with head injuries and the clinical significance of these gradients in their management. METHODS The authors present the results of a prospective study conducted in 50 head-injured patients to determine the clinical significance of supratentorial ICP gradients. In each case a concurrent bilateral frontal intraparenchymatous device was implanted within the 6-hour window after computerized tomography (CT) scanning. According to CT criteria, each patient was categorized into one of three different groups: 1) diffuse lesions, in which no unilaterally measured volumes greater than 25 ml were present and the midline shift was 3 mm or less; 2) Focal A, in which added hemispheric volumes were greater than 25 ml and midline shift was 3 mm or less; and 3) Focal B, in which all patients with a midline shift greater than 3 mm were included. From the results of the entire group the authors were able to distinguish four different patterns of supratentorial ICP. In Pattern I, the intracranial compartment behaved as a true unicameral space with similar mean ICPs and pulse amplitudes in both hemispheres; in Pattern II, different mean ICPs and amplitudes were observed although ICP increases or decreases were congruent; and in Pattern III, patients with different mean ICPs, different ICP amplitudes, and no congruent increases or decreases of ICP were included. All (15 cases) but one patient with a diffuse lesion presented with ICP Pattern I. Fifteen patients with focal lesions showed a Type II pattern, whereas only one patient presented with a Type III pattern. In 10 patients, of whom all but one presented with a focal lesion, transient gradients that disappeared in less than 4 hours were also observed. CONCLUSIONS In many patients with focal lesions, clinically important interhemispheric ICP gradients exist. In this subset, transient gradients that disappear with time are frequently observed and may indicate an increase in the size of the lesion. The clinical relevance of such gradients is discussed and guidelines for adequately monitoring ICP are suggested to optimize head injury management and to avoid suboptimal or even harmful care in patients with mass lesions.