In Reply:-The comment by Cold el al. to our article "Cerebral Vasoconstriction by Indomethacin in Intracranial Hypertension: addresses important problems within neuro-intensive care: Should post-traumatic brain swelling be treated with pharmacologic vasoconstriction? Is so, what drug and dosage should be chosen? In our study of induced intracranial hypertension in piglets, we found that indomethacin reduced intracranial pressure (ICP) by constriction of pre-capillary resistance vessels only when cerebral blood flow (CBF) was reduced to a level where progressive changes in CacO2, vpH, and electroencephalography occurred, indicating ischemia. We concluded that our data did not support the use of indomethacin in patients with intracranial hypertension due to head injury. In their comment, Cold el al. refer to clinical publications not available at that time. Some of these are still in press, but we had the opportunity to review the manuscripts.
The study by Biestro et al. included 10 patients with severe head injuries and raised ICP. The patients were given an indomethacin bolus (50 mg) followed by a continuous infusion (21.5 +/- 11.4 mg/h). The authors measure significant decreases in ICP and increase in cerebral perfusion pressure (CPP) during the treatment, but no data were given regarding CBF, CavO2, vpH, electroencephalography, or other physiologic parameters. Four of the ten head-injured patients were reported dead, but no data regarding the cause or the time of death were presented.
In the study by Dahl et al., physiologic parameters that included ICP, CPP, CBF, and CavO2were studied in 14 patients with head injuries (mean ICP 14.8 mmHg before treatment). The study compared the effect of a bolus indomethacin (30 mg intravenously) with that of induced hypocapnia. Indomethacin was found to decrease ICP and CBF and increase CavO2significantly. It is remarked by the authors that: "Although ischaemia evaluated from CBF, SvjO3, and LOI (Lactate/oxygen index) were indicated in individual cases after both hyperventilation and indomethacin, it was not reflected in outcome." However, the increase in ICP before treatment was very modest, and "patients in poor clinical condition indicating transtentorial herniation were excluded." There is no indication in the study that a bolus injection of indomethacin had any beneficial effect outcome.
In their letter, Cold et al. claim that, during continuous intravenous indomethacin treatment, "the ICP-reducing effect is sustained and the increase in AVDO2, the jugular venous lactate and EEG are transitory" and that they "find a low mortality of 10% in indomethacin treated severe head injury." We have been unable to find the data supporting these statements. In the report by Jensen et al. indomethacin was administered intravenously throughout a 7-h period. During this period, the decrease in global CBF and the increase in CavO sub 2 remained. At 7 h after of infusion, ICP seemed to have returned close to pretreatment level (28 +/- 3 mmHg vs. 25 +/- 11 mmHg). There is a lack of information regarding physiologic parameters in patients with severe head injuries when indomethacin infusion is continued beyond of a focal mass expanding lesion in pigs cannot rightly be transferred to the pathophysiological events of severe human head injury." Although this is obviously true, it seems to us that the clinical experiences by Biestro et al. and Dhal el al. are in good agreement with our experimental study: indomethacin reduces increased ICP, but at the expense of decreased CBF and impending cerebral ischemia. We agree with Cold et al. that indomethacin is contraindicated in patients suffering from cardiac schemia, renal insufficiency, gastric bleedings, and coagulation disorders. Unfortunately, patients with severe head trauma may, quite often, experience multi-organ failure, which may limit the clinical use of indomethacin even further.
In our opinion, cerebral vasoconstriction is only one of the components, though important in the treatment of posttraumatic brain swelling. Our treatment concepts was published recently. The concepts aims at a reduction of ICP through decreases CBF and increased absorption of cerebral interstitial fluid. The latter is accomplished through reduction of intracapillary pressure by a combination of moderate precapillary vasoconstriction (DHE) and reduction of mean arterial blood pressure. We are currently reviewing our clinical results since 1989 that show mortality in patients with closed head injuries has been almost eliminated, with no increase in the number of patients surviving in a vegetative state (Eker et al., in preparing).
Although Cold et al. and our group seem to disagree partly regarding the clinical use of indomethacin, we share the view that pharmacologic vasoconstriction shows promise for the treatment of increased ICP. Probably, the optimal drug should combine a pronounced constriction of cerebral venous capacitance vessels and a moderate constriction of cerebral precapillary resistance vessels with no or insignificant effects in other vascular territories.
Fritiof Nilsson, M.D., Staff anesthesiologist, Department of Anesthesia, Malmo University Hospital.
Sven Bjorkman, Ph.D., Associate Professor Pharmaceutical Chemistry, Hospital pharmacy, Malmo University Hospital.
Ingmar Rosen, M.D., Ph.D., Professor Clinical Neurophysiology, University Hospital, Lund.
Kenneth Messeter, M.D., Ph.D., Associate Professor Anesthesiology, Head of Department, Department of Anesthesia, Malmo University Hospital, S-20502 Malmo, Sweden.
Carl-Henrik Nordstrom, M.D., Ph.D., Associate Professor Neurosurgery, Department of Neurosurgery, University Hospital, Lund, S-22185 Lund, Sweden.
(Accepted for publication September 12, 1996.)