Indomethacin: its role in the management of intractable intracranial pressure (ICP) after severe head injury

Intracranial hypertension that fails to respond to first linemedical and surgical treatment after head injury is associatedwith a 92% mortality overall.¹ The addition of barbiturateswill result in a good or moderate neurological outcome in 35%of patients,² but those with hyperaemia do significantly worse.However barbiturates may not be a logical choice for those patientswhose intracranial hypertension is secondary to hyperaemia,and may be associated with significant complications. In thesepatients cerebral vasoconstrictors such as indomethacin maybe more appropriate and possibly associated with fewer unwantedeffects. In one study six out of 10 patients who received indomethacinfor intracranial hypertension unresponsive to barbiturates survived.³However it is unclear how many were hyperaemic since jugularvenous saturation (Sj_O2) was not monitored. We report our experiencewith indomethacin in 10 severely head-injured patients. Our protocol aims to maintain a target cerebral perfusion pressure(CPP) and ICP through the application of sedation, diuretics,CSF drainage, mild hypothermia, muscle relaxation and controlof arterial carbon dioxide (Pa_CO2). If the ICP remains elevatedthen Sj_O2 is monitored. The combination of raised ICP and Sj_O2is taken to indicate hyperaemia (absolute or relative). In thesecircumstances the patient is hyperventilated to a Pa_CO2 of 28mmHg and if necessary an intravenous infusion of thiopentonecommenced. We used indomethacin infusions in 10 patients fulfillingthese criteria of hyperaemia. In seven patients the hyperaemiawas confirmed as absolute by demonstrating a raised middle cerebralartery velocity (MCAV) with transcranial Doppler. The mean ageof the patients was 21.2 yr (range 8–55). Indomethacinwas infused for a mean of 3.8 days (1–11) at a rate of3–11 mg h^–1. The effect of indomethacin on mean(SD) ICP, CPP and Sj_O2 is shown in Table 4. At 6-month follow up there were seven survivors (three goodrecovery, three moderate recovery, one severely disabled). Threepatients died with intractable ICP and septic shock. Two ofthese patients had associated renal failure. There were no episodesof gastrointestinal bleeding. Two of the three patients whodied did not have MCAV measured, and therefore indomethacinmay not have been strictly indicated. These results achievedin this subgroup of head injured patients is much better thanthat expected, and matches the outcome achieved in the overallICU head injury population. Indomethacin may have a role inthe management of raised ICP associated with hyperaemia aftersevere head injury. We recommend however that it should onlybe used with monitoring of both Sj_O2 and MCAV.