Use of indomethacin in brain-injured patients with cerebral perfusion ...

J Neurosurg 83:627–630, 1995

Use of indomethacin in brain-injured patients with cerebral perfusion pressure impairment: preliminary report ALBERTO A. BIESTRO, M.D., RICARDO A. ALBERTI, M.D., ANA E. SOCA, M.D., MARIO CANCELA, M.D., CORINA B. PUPPO, M.D., AND BERNARDO BOROVICH, M.D. Intensive Care Unit and Department of Neurosurgery, Hospital de Clínicas, Faculty of Medicine, Montevideo, Uruguay U The effect of indomethacin, a cyclooxygenase inhibitor, was studied in the treatment of 10 patients with head injury and one patient with spontaneous subarachnoid hemorrhage, each of whom presented with high intracranial pressure (ICP) (34.4 6 13.1 mm Hg) and cerebral perfusion pressure (CPP) impairment (67.0 6 15.4 mm Hg), which did not improve with standard therapy using mannitol, hyperventilation, and barbiturates. The patients had Glasgow Coma Scale scores of 8 or less. Recordings were made of the patients’ ICP and mean arterial blood pressure from the nurse’s end-hour recording at the bedside, as well as of their CPP, rectal temperature, and standard therapy regimens. The authors assessed the effects of an indomethacin bolus (50 mg in 20 minutes) on ICP and CPP; an indomethacin infusion (21.5 6 11 mg/hour over 30 6 9 hours) on ICP, CPP, rectal temperature, and standard therapy regimens (matching the values before and during infusion in a similar time interval); and discontinuation of indomethacin treatment on ICP, CPP, and rectal temperature. The indomethacin bolus was very effective in lowering ICP (p , 0.0005) and improving CPP (p , 0.006). The indomethacin infusion decreased ICP (p , 0.02), but did not improve CPP and rectal temperature. The effects of standard therapy regimens before and during indomethacin infusion showed no significant changes, except in three patients in whom mannitol reestablished its action on ICP and CPP. Sudden discontinuation of indomethacin treatment was followed by significant ICP rebound. The authors suggest that indomethacin may be considered one of the frontline agents for raised ICP and CPP impairment.

KEY WORDS • head injury • brain injury • intracranial pressure • cerebral perfusion pressure • indomethacin

LEVATED intracranial pressure may be a fatal complication of severe head injuries and other diseases of the central nervous system.5 The mortality rate associated with head injury or expanding intracranial lesions is roughly proportional to the level of intracranial pressure (ICP), reaching 100% when ICP is maintained at 60 mm Hg.7 Elevated ICP is the primary cause of death in approximately 50% of patients with severe head injuries.8 Aggressive treatment of intracranial hypertension has been shown to reduce the overall mortality rate in patients with these injuries.11 Controlled hyperventilation, osmotherapy, and barbiturates are standard therapy for intracranial hypertension; however, these treatments are not consistently successful. On the basis of preliminary clinical reports about the effectiveness of indomethacin, a prostaglandin inhibitor, in lowering ICP,3,10 we decided to evaluate its influence on ICP and cerebral perfusion pressure (CPP) in a group of patients who exhibited poor response to standard therapy.

E

J. Neurosurg. / Volume 83 / October, 1995

Clinical Material and Methods Patient Population Our study focused on 11 patients, eight males and three females, ranging in age from 10 to 39 years (mean 24 years). These patients were admitted to the Intensive Care Unit (ICU) of the University Hospital de Clínicas, Montevideo, Uruguay, over a period of 21 months from January 1991 to September 1992. Ten of the patients had head injuries and one had a spontaneous subarachnoid hemorrhage complicated with an intracerebral hematoma. The patients’ Glasgow Coma Scale2 scores at admission ranged from 3 to 11 (mean 6). The mean delay between injury and admission to the ICU was 32 hours. Computerized tomography (CT) of the head was performed at admission in all cases. The head injuries sustained by the patients included: one gunshot wound with brain swelling, five isolated brain contusions, one diffuse axonal injury with brain swelling, and three extracerebral hematomas 627

A. A. Biestro, et al. TABLE 1 Clinical characteristics of 11 brain-injured patients* Case Age (yrs), No. Sex

Injury

1 2 3 4 5 6 7 8 9 10 11

CHT CHT CHT CHT CHT CHT SSAH CHT CHT GS CHT

21, F 39, M 25, M 19, M 22, M 24, F 24, M 34, M 10, M 21, F 34, M

GCS Score

6 7 3 4 5 11 8 4 5 7 4

CT Findings

Surgical Treatment

Outcome

yes no yes yes yes yes yes no yes no no

death alive death alive death death alive alive alive alive alive

EH + HS UC + BS SH + UC UC + BS UC + BS UC + BS SSAH + IH UC + BS EH + HS GS + BS DAI + BS

* Abbreviations: GCS = Glasgow Coma Scale; CT = computerized tomography; CHT = closed head trauma; EH = extradural hematoma; HS = hemispheric swelling; UC = unilateral contusion; BS = brain swelling; SH = subdural hematoma; SSAH = spontaneous subarachnoid hemorrhage; IH = intracranial hematoma; GS = gunshot wound; DAI = diffuse axonal injury.

(two extradural and one acute subdural) with postoperative hemispheric brain swelling. Compressed basal cisterns were appreciated in all of the cases and seven patients underwent surgery for mass lesion. Table 1 provides a summary of the clinical data, CT findings, and outcomes of the 11 patients. Treatment and Record Keeping

Intracranial pressure and mean arterial blood pressure (MABP) measurements were obtained from the nurse’s bedside “end-hour” notes according to Traumatic Coma Data Bank methodology,6 and rectal temperature was recorded throughout the treatment. The ICP was monitored with a Richmond subdural screw in nine cases and with a subdural catheter in five cases of patients operated on for mass lesions. Three of the patients were monitored by TABLE 2 Effects of indomethacin bolus and indomethacin infusion on ICP and CPP in 11 patients* Indomethacin Bolus (50 mg, 20 min)

Indomethacin Infusion (21.5 6 11 mg/hr, 30 6 9 hrs)

Case No.

ICP (mm Hg)

CPP (mm Hg)

ICP (mm Hg)

CPP (mm Hg)

1 2 3 4 5 6 7 8 9 10 11

31 ➝ 7 30 ➝ 19 45 ➝ 28 28 ➝ 8 25 ➝ 12 52 ➝ 22 48 ➝ 21 26 ➝ 17 NA NA 25 ➝ 11

67 ➝ 72 65 ➝ 67 84 ➝ 96 68 ➝ 73 61 ➝ 73 86 ➝ 96 38 ➝ 59 NA NA NA 65 ➝ 74

29 ➝ 14 28 ➝ 15 41 ➝ 29 16 ➝ 12 23 ➝ 14 32 ➝ 35 41 ➝ 36 27 ➝ 19 49 ➝ 38 15 ➝ 16 21 ➝ 25

60 ➝ 57 69 ➝ 72 70 ➝ 66 72 ➝ 68 60 ➝ 66 54 ➝ 62 41 ➝ 58 NA 60 ➝ 63 81 ➝ 84 66 ➝ 60

* Number preceding the arrow indicates pressure before indomethacin administration; number following the arrow indicates pressure during administration. ICP = intracranial pressure; CPP = cerebral perfusion pressure; NA = not available.

628

FIG. 1. Bar graphs showing the effect of indomethacin bolus and infusion on intracranial pressure (ICP) and cerebral perfusion pressure (CPP). A and B: Decrease in ICP and increase in CPP occur as a result of an indomethacin bolus given over 20 minutes and recorded 1 hour later. C and D: Decrease in ICP and an insignificant increase in CPP occur as a result of an indomethacin infusion administered over 30 hours.

both methods: a subdural Richmond screw initially and a subdural catheter after surgery. Standard therapy was initiated to maintain ICP under 20 mm Hg and CPP above 70 to 80 mm Hg. This therapy consisted of hyperventilation reaching a mean PaCO2 of 24.8 6 7.1 mm Hg; mannitol at a dosage of 0.25 g/kg each time ICP surpassed 20 mm Hg; and barbiturates (thiopental) administered as a bolus or infusion only when ICP could not be maintained under 20 mm Hg with hyperventilation and mannitol. Thiopental infusion (2–3 g/day) was only given when the original bolus of 0.2 to 0.3 g was effective. If the 0.2- to 0.3-g bolus was slightly effective, it was repeated intermittently. Indomethacin treatment was only begun after the previous therapy proved ineffective. At this time the patients had a mean ICP of 34.4 6 13.1 mm Hg. Throughout the treatment we assessed the effects of 1) an indomethacin bolus (50 mg, 20 minutes) on ICP and CPP; 2) an indomethacin infusion (21.5 6 11.4 mg/hour, 30 6 9 hours) on ICP, CPP, rectal temperature, and additional standard therapy regimens (matching values before and during infusion in a similar time interval); and 3) discontinuation of indomethacin on ICP, CPP, and rectal temperature evaluated 24 hours after interruption of treatment. Statistical Analysis

All values are given as the mean 6 standard deviation. We used the t-test for paired data; a calculated difference of probability less than 0.05 was considered to be statistically significant. Results Table 2 lists the effects of the indomethacin bolus and infusion on ICP and CPP in each patient. Table 3 and Figs. 1 and 2 summarize the results. In response to the indomethacin bolus, ICP decreased J. Neurosurg. / Volume 83 / October, 1995

Indomethacin and patients with severe brain injury TABLE 3 Changes in patient characteristics and standard therapy caused by indomethacin treatment in 11 patients* Indomethacin Administration

indomethacin bolus significance indomethacin infusion significance discontinuation of indomethacin significance

Intracranial Pressure (mm Hg)

Cerebral Perfusion Pressure (mm Hg)

Rectal Temperature (Celsius)

Hyperventilation (mm Hg)

Mannitol (g)

Barbiturate (g)

218.0 p , 0.0005 26.35 p , 0.02 +10.46 p , 0.003

+9.43 p , 0.006 +2.32 NS 27.02 NS

NA

NA

NA

NA

NC NS +0.64 p , 0.007

+1.1 NS NA

229 NS NA

+0.3 NS NA

* NA = not applicable; NC = no change; NS = not significant.

from 34.4 6 13.1 mm Hg to 16.4 6 7.2 mm Hg (p , 0.0005) (Fig. 1A) and CPP increased from 67.0 6 15.4 mm Hg to 76.4 6 14.5 mm Hg (p , 0.0006) (Fig. 1B). With indomethacin infusion ICP decreased from 29.5 6 9.9 mm Hg to 23.1 6 9.7 mm Hg (p , 0.02) (Fig. 1C) and CPP increased from 63.3 6 9.8 mm Hg to 65.6 6 8.5 mm Hg (Fig. 1D). These results were not significant (p = 0.4) and no changes in rectal temperature were recorded. Standard therapy was maintained before and throughout the trial of indomethacin. Hyperventilation changed from a mean PaCO2 of 24.8 mm Hg prior to indomethacin infusion to 25.9 mm Hg during the infusion (p = 0.6). Administration of mannitol varied from 114 g before indomethacin infusion to 85 g during the infusion (p = 0.1). Thiopental was given at a mean dose of 0.74 g before the indomethacin infusion was begun; during indomethacin infusion the dose of thiopental was raised to 1.05 g, a dif-

FIG. 2. Bar graphs depicting the rebound effect of intracranial pressure (ICP) (A) and cerebral perfusion pressure (CPP) (B) shown at 24 hours after discontinuation of indomethacin infusion.


ference not considered significant. Two patients (Cases 7 and 9) received a 2-g dose per day of thiopental infusion both before and during indomethacin. The other patients received thiopental as an intermittent bolus before and during indomethacin infusion. It should be noted that in three cases mannitol reestablished its effects during indomethacin infusion. One example (Case 1) is shown in Fig. 3. Adverse effects of indomethacin were not observed. After administration of indomethacin was discontinued, ICP increased from 21.3 to 31.8 mm Hg, a change that was significant (p , 0.003). Cerebral perfusion pressure decreased from 71.4 to 64.4 mm Hg (Fig. 2); however, this change was not statistically significant (p , 0.08). Rectal temperature increased significantly from 37.28 to 37.98C (p , 0.007). Discussion On the basis of our data and a review of the literature, we believe that the action of indomethacin is mediated through the following three mechanisms. First, like Jensen and colleagues,3 we found the indomethacin bolus to exert a very effective action on lowering ICP and improving CPP. However, we used in-

FIG. 3. Graph showing the effect on intracranial pressure of indomethacin bolus and infusion and the reestablishment of mannitol action during indomethacin infusion in Case 1. Numerals on x-axis indicate time in hours.

629

A. A. Biestro, et al. domethacin infusion over a longer period than these authors and found that its effect did not fade away rapidly in time. This fact, together with the risks of rebound that were observed as a consequence of indomethacin discontinuation, led us to believe that the vasoactive effect of indomethacin is paramount and that this effectiveness continues for a longer period of time than previously considered.3 Second, it has been asserted that standard therapy for increased ICP is only effective if some degree of autoregulation is maintained or if, at least, CO2 reactivity is preserved.12 Although the standard therapy observed before and during indomethacin infusion did not appear to effect significant changes, the fact that in three cases mannitol reestablished its action on ICP and CPP during indomethacin infusion was consistent with the view that indomethacin could have a beneficial effect on cerebral autoregulation.9 Third, it is also probable that the beneficial effects of indomethacin infusion could be due to nonvascular actions on baseline ICP in pathological conditions such as ischemic and vasogenic edema.1,4,14 Decreased production of cerebrospinal fluid following endothelin potentiation has been recently demonstrated.13 Although we did not record cerebral temperature, we did find that the effect of indomethacin could not be attributed to reduction in rectal temperature. In this study we found indomethacin to be a safe drug with good hemodynamic tolerance and no sedative sideeffects. Although we did not measure cerebral blood flow and cerebral metabolic rate for oxygen, we did not observe areas of ischemia or infarction on follow-up CT scans. We believe that indomethacin should not only be considered a last option for treatment but that it can be considered as one of the frontline therapeutic agents for raised ICP and impaired CPP. We must caution, however, against a sudden interruption of indomethacin infusion because of a real risk of ICP rebound. References 1. Deluga KS, Plötz FB, Betz AL: Effect of indomethacin on edema following single and repetitive cerebral ischemia in the gerbil. Stroke 22:1259–1264, 1991

630

2. Jennett B, Teasdale G, Braakman R, et al: Prognosis of patients with severe head injury. Neurosurgery 4:283–289, 1979 3. Jensen K, Öhrström J, Cold GE, et al: The effects of indomethacin on intracranial pressure, cerebral blood flow and cerebral metabolism in patients with severe head injury and intracranial hypertension. Acta Neurochir 108:116–121, 1991 4. Kim HJ, Levasseur JE, Patterson JL Jr, et al: Effect of indomethacin pretreatment on acute mortality in experimental brain injury. J Neurosurg 71:565–572, 1989 5. Lobato RD, Cordobes F, Rivas JJ, et al: Outcome from severe head injury related to the type of intracranial lesion. A computerized tomography study. J Neurosurg 59:762–774, 1983 6. Marmarou A, Ward JD, Young HF, et al: NINDS Traumatic Coma Data Bank: intracranial pressure monitoring methodology. J Neurosurg (Suppl) 75:S21–S27, 1991 7. Marshall LF, Smith RW, Shapiro HM: The outcome with aggressive treatment in severe head injuries. Part I: The significance of intracranial pressure monitoring. J Neurosurg 50: 20–25, 1979 8. Miller JD, Becker DP, Ward JD, et al: Significance of intracranial hypertension in severe head injury. J Neurosurg 47: 503–516, 1977 9. Norins NA, Wendelberger K, Hoffman RG, et al: Effects of indomethacin on myogenic contractile activation and responses to changes in O2 and CO2 in isolated feline cerebral arteries. J Cereb Blood Flow Metab 12:866–872, 1992 10. Pickard JD: Role of prostaglandine and arachidonic acid derivatives in the coupling of cerebral blood flow to cerebral metabolism. J Cereb Blood Flow Metab 1:361–384, 1981 11. Saul TG, Ducker TB: Effect of intracranial pressure monitoring and aggressive treatment on mortality in severe head injury. J Neurosurg 56:498–503, 1982 12. Schalén W, Messeter K, Nordström CH: Cerebral vasoreactivity and the prediction of outcome in severe traumatic brain lesions. Acta Anaesthesiol Scand 35:113–122, 1991 13. Schalk KA, Faraci FM, Heistad DD: Effect of endothelin on production of cerebrospinal fluid in rabbits. Stroke 23: 560–563, 1992 14. Yen MH, Le SH: Effects of cyclooxygenase and lipoxygenase inhibitors on cerebral edema induced by freezing lesions in rats. Eur J Pharmacol 144:369–373, 1987 Manuscript received July 11, 1994. Accepted in final form May 12, 1995. This work was supported in part by Laboratorios Gerardo Ramón y Cia., Montevideo, Uruguay. Address reprint requests to: Alberto A. Biestro, M.D., Avenida Italia 3190/803, 11600 Montevideo, Uruguay.
