Intracranial Pressure Monitoring in the Management ofPenetrating Brain Injury

J Trauma. 2001;51:S12–S15.

I. RECOMMENDATIONSA. Standards

The available data are not sufficient to support a treatmentStandard for the use of intracranial pressure (ICP) monitoring inthe management of acute penetrating brain injury (PBI).

B. GuidelinesThe available data are not sufficient to support a treat-

ment Guideline for this topic.

C. OptionsEarly ICP monitoring is recommended when the clini-

cian is unable to assess the neurologic examination accu-rately; the need to evacuate a mass lesion is unclear; orimaging studies (e.g., computed tomographic scan) suggestelevated intracranial pressure. In the absence of studies spe-cific to managing intracranial hypertension in PBI patients,we recommend that the clinician follow the recommendationsin Guidelines for the Management of Severe Head Injury1 andthe enlarged revision Guidelines for the Management of Se-vere Traumatic Brain Injury.2

II. OVERVIEWICP monitoring has been established as an important

variable in the prognosis and management of severe nonpen-etrating traumatic brain injury (TBI) because intracranialhypertension is clearly associated with worse recovery andeffective control of elevated ICP appears to improveoutcome.1 In contrast to nonpenetrating TBI, few studies onICP in PBI have been performed, and the majority of thosestudies do not provide ICP-related data.3–19 Few studies onintracranial hypertension after PBI in the military have beenpublished, and there is a dearth of civilian literature on use ofICP monitoring for patients with PBI. Because all availablestudies are observational and without controls, this section ofthe Guideline can make no recommendations stronger thanOptions (see definitions above). However, some aspects ofICP management discussed in the literature on nonpenetrat-ing TBI can be generalized and applied logically to PBI, eventhough all information is derived from Class III data.

III. PROCESSA MEDLINE search from January 1966 to January 2000

using the search terms wounds, gunshot, and brain injuries orhead injuries, limited to human subjects, identified 382 articlesfrom peer-reviewed journals. Eighty-eight articles were rejected

on the basis of clearly irrelevant titles; an additional 33 articleswere then identified from the bibliographies of reviewed articles.The primary selection process, therefore, identified 327 articlesfor further review. Two independent reviewers read the abstractsof all 327 articles and selected 65 for inclusion. Articles wererejected on the basis of relevance to the topic (e.g., non-Englishlanguage, case reports, irrelevance to project, atypical mecha-nisms of injury, and series of less than 10 subjects with no otherunique reasons for inclusion). The 65 articles were then read indetail, and more articles were considered from their bibliogra-phies. Of these 65 articles, 52 were directly relevant to the topicsof surgical management, monitoring intracranial pressure, andtreating cerebrospinal fluid leaks. A total of seven articles wereappropriate for this topic and were analyzed in detail for thissection

IV. SCIENTIFIC FOUNDATIONStudies that include analyses of data on ICP in PBI reveal

that intracranial hypertension is common after PBI.6,13,15,16 Lil-lard reported that in a series of 18 patients with PBI who diedand had a postmortem examination, all had evidence of in-creased intracranial pressure with temporal lobe and/or cerebel-lar tonsillar herniation. He also recorded ICP in 12 patients andfound “significant” (degree not specified) intracranial hyperten-sion in 11 (92%).13 Miner et al.14 monitored ICP in 18 of 33children with gunshot wounds to the brain who were deemedsalvageable but not so neurologically intact as to preclude mon-itoring. Of the 12 children with ICP values uncontrollably above40 mm Hg, 8 died (75%), 2 were left with severe deficits, and 2were left with moderate deficits. In contrast, all patients havingICPs controllable at � 20 mm Hg had good outcomes or mod-erate disabilities at 6 months after injury. Nagib et al.15 recordedICP in 13 civilian patients with PBI and found that the majorityof those with intracranial hypertension that responded to treat-ment survived, whereas all those in whom intracranial hyper-tension was refractory died. They also found that outcome wassuperior in those who never manifested intracranial hyperten-sion. Crockard6 recorded ICP in 20 patients very early after PBIand found a very high mortality associated with elevated ICPand a much better outcome in those in whom intracranial hy-pertension was never manifest. He also found a group with verylow ICP associated with profound shock, but when ICP rose tonormal after resuscitation, it presaged survival, whereas intra-cranial hypertension after resuscitation was uniformly associatedwith mortality. His analysis of the relationship between ICP andblood pressure suggested that pressure autoregulation was de-

The Journal of TRAUMA� Injury, Infection, and Critical Care

S12 August Supplement 2001

fective after PBI, and that this commonly resulted in intracranialhypertension associated with respiratory difficulties, such ascoughing or respiratory distress. He stressed the importance offull-volume resuscitation and aggressive airway and respiratorymanagement, including endotracheal intubation and mechanicalventilation.

Unfortunately, in all of these reports, the percentage oftotal patients who received ICP monitoring was small and notrandomized, so their results cannot be generalized. Neverthe-less, all reports indicate that intracranial hypertension is com-mon after PBI and that response to treatment may be associ-ated with improved outcome in some patients.

Siccardi et al.19 published the only study in which ICPmonitoring was not described favorably. In their report, re-cording of ICP “was performed in several of the surgicalcases and in a few of the conservatively treated cases.” Theystated that “there was no significant difference in the outcomebetween patients who were and who were not submitted topostoperative measurement of intracranial pressure.” Unfor-tunately, they presented no data to support this statement, didnot describe the characteristics of the patients undergoing ICPmonitoring, and did not describe either their method formanaging intracranial hypertension or its results.

The physiology of intracranial hypertension after PBI hasnot been well studied. Postmortem studies of PBI victims reveala preponderance of evidence for intracranial hypertension. Kirk-patrick and DiMaio found sulcal flattening and signs of tran-stentorial herniation in 39 of 42 PBI patients at necropsy.20

Some of the marked elevation in ICP deduced from this evi-dence may have been associated with the cavitation associatedwith the missile wound. However, it appeared that cerebraledema can appear extremely rapidly and contribute to intracra-nial hypertension in patients with treatable PBI. As noted above,Crockard’s study found that intracranial hypertension occurredfrequently among patients in whom ICP monitoring was insti-tuted in under 20 minutes after arrival at a medical facility. Itappears that some patients with intracranial hypertension can betreated and survive with acceptable morbidity.6,14–16 However,the literature does not clearly relate successful treatment ofintracranial hypertension after PBI to an improved outcome.

In addition, no solid information is available to guide man-agement strategies for intracranial hypertension (ICH) after PBI.It cannot be assumed that the intracranial pathology or its evo-lution over time is the same as in nonpenetrating TBI. It has beensuggested that failure of cerebral pressure autoregulation is acommon source of ICP elevations after PBI.6 Hyperventilationhas been considered a mainstay of treatment and even in pro-phylaxis of post-PBI intracranial hypertension.4,6–8,12,14,16,20,21

The degree of hyperventilation reported in the literature for bothof these indications has often been extreme (i.e., PaCO2 levels of20–25 mm Hg).7,16,21 In light of the current concern aboutinduced hypoperfusion caused by hyperventilation, however, therisk/benefit ratio of hyperventilation needs to be compared withthe other measures available for controlling ICH.1 Absent theevidence of a clear relationship between hyperventilation in PBI

and subsequent hypoperfusion, our present poor level of under-standing of the pathophysiologic alterations after PBI is clearlyan area for further study.

One practical use for ICP monitoring is to follow patientswith depressed levels of consciousness as an early sign ofprogressive intracranial pathology that may require repeatedimaging and surgical decompression.6,15,16 The utility of ICPmonitoring as a surrogate for a more sensitive clinical indi-cator of neurologic deterioration is probably the most logicalparallel between nonpenetrating TBI and PBI.

In cases where ICP is monitored and intracranial hyper-tension is present, treatment has involved the same measuresused in nonpenetrating TBI (e.g., hyperventilation, mannitol,cerebrospinal fluid drainage, and high-dose barbiturates).Prophylaxis against intracranial hypertension, an aspect ofcare frequently addressed in the limited literature, uses thesame agents. In this context, the potential toxicity of suchagents compared with the risks of monitoring ICP becomesan important consideration. This same risk/benefit analysiswas carried out for nonpenetrating TBI in Guidelines for theManagement of Severe Head Injury.1 That evidence-baseddocument concludes that the risks of ICP monitoring arefrequently outweighed by the potential toxicity of anti-ICHagents that may not be necessary.1 For example, severalprospective, randomized, controlled trials showed that high-dose barbiturates were not useful as prophylaxis for prevent-ing ICH in nonpenetrating TBI. Furthermore, although high-dose barbiturates are effective in lowering ICP elevationsrefractory to less aggressive treatments in these same pa-tients, they should not be used in patients without provenintracranial hypertension and in whom less hazardous treat-ment measures have not been tried.22–24 The unclear paral-lelism between nonpenetrating TBI and PBI precludes directapplication of the analysis contained in Guidelines for theManagement of Severe Brain Injury to PBI. Nevertheless, therisk/benefit question for “empiric” treatment versus ICP mon-itoring may be pertinent to patients with PBI and suggeststhat empiric treatment of increased ICP is not beneficial.

V. SUMMARYThe role of ICP monitoring and its application in PBI

have been incompletely studied. In the available literature,intracranial hypertension appears to be common after PBIand, when present, is predictive of less favorable outcome. Italso appears that systemic hypertension can exacerbate, andpotentially be solely responsible for, ICP elevation after PBI.Unfortunately, few data reveal whether or how successfulmanagement of intracranial hypertension improves outcomesin patients with PBI. Most of the few available Class IIIreports suggest that patients whose ICP can be maintainedbelow 20 mm Hg have more favorable recovery than patientswith refractory intracranial hypertension. There is some ClassIII evidence that indicates that ICP monitoring can be usefulto help detect deterioration in patients with depressed levelsof consciousness.

ICP Monitoring in the Management of Penetrating Brain Injury

Volume 51 • Number 2 S13

Unfortunately, almost no data are available on theoptimal method for treating post-PBI intracranial hyper-tension. Few data are available on the efficacy of individ-ual methods of management, and there are no data on theirrelative efficacy. Without such evidence, it seems appro-priate to defer to what is known about treating intracranialhypertension in nonpenetrating TBI regarding the absoluteand relative efficacies of individual treatment methods, asoutlined in Guidelines for the Management of Severe Trau-matic Brain Injury.2

VI. KEY ISSUES FOR FUTURE INVESTIGATIONSOur knowledge regarding the role of intracranial hyperten-

sion in determining outcome in PBI is more primitive than fornonpenetrating TBI. Nevertheless, many of the issues aresimilar. The nature of the pathophysiologic processes in-volved, their similarity to those occurring in nonpenetrat-ing TBI, and the optimal methods for detecting, measuring,and treating them are yet to be determined. Formal studies

are needed to assess the prognostic significance of ICPelevations after PBI. How to identify those patients at riskfor intracranial hypertension and what level of ICH re-quires treatment also need to be defined. Finally, whethermanaging intracranial hypertension improves outcome inPBI and how this management interacts with the manipu-lation of other physiologic parameters, such as cerebralperfusion pressure and cerebral oxygen delivery, are topicsfor future study.

Because there is so little research on the above issues,there is no clear evidence that any particular diagnostic ortherapeutic approach is superior to any others in PBI. Assuch, the ethical constraints that have prevented definitiveresearch into the utility of ICP monitoring and managementin improving outcome in patients with nonpenetrating TBIwill not a priori apply to PBI. For this reason, it shouldremain possible to perform a prospective, randomized, con-trolled trial on the role of ICP management in improving PBI.

VII. Evidentiary Table: ICP Monitoring in Penetrating Brain Injury

Authors, Year Description of Study DataClass Conclusions

Crockard,19756

Civilian (n � 20). Intracranial pressure wasmonitored within 20 min in patients withgunshot wounds to the head in Belfast.

III Intracranial hypertension was frequent in patients in whom ICPmonitoring was instituted in less than 20 min. Reported veryhigh mortality associated with elevated ICP and much betteroutcome among patients who never manifested intracranialhypertension. Concluded that patients with intracranialhypertension can be treated and survive with acceptablemorbidity. ICP monitoring useful as an early sign of progressiveintracranial pathology and as an indication for repeated imagingand surgical decompression. Thought that failure of cerebralpressure autoregulation is a common source of ICP elevationsafter PBI.

Kirkpatrickand DiMaio,197820

Civilian (n � 42). Postmortem anatomic studyof fatal gunshot wounds to the brain.

III Found sulcal flattening and signs of transtentorial herniation in 39 of42 PBI patients at necropsy.

Lillard, 197813 Civilian (n � 83). The report reviewed clinicalcourse and conservative surgical approachto victims with civilian PBI seen over a 5-yperiod.

III Recorded ICP in 12 patients and found “significant” intracranialhypertension in 11 (92%).

Miner et al.,199014

Civilian (n � 33). Retrospective study ofoutcome from gunshot wounds to the brainin a pediatric population (8 mo through 5 y)in an urban setting.

III Monitored ICP in of 33 children with gunshot wounds to the brain thatwere deemed neither unsalvageable nor too intact to preclude theneed for monitoring. 75% mortality in 12 with ICP valuesuncontrollably � 40 mm Hg, with two left with severe deficits, andtwo with moderate deficits. All patients with ICPs controllable at � 20mm Hg had good outcomes or moderate disabilities at 6 mo afterinjury.

Nagib et al.,198615

Civilian (n � 55). Over a 5-y period, victimsof cranial gunshot wounds to the headwere managed and factors related tooutcome including GCS score,fragmentation, and suicide wereevaluated.

III Recorded ICP in 13 patients and found that the majority of thosewith intracranial hypertension that responded to treatmentsurvived, whereas all in whom intracranial hypertension wasrefractory died. Also found superior outcome in those who nevermanifested intracranial hypertension.

Sarnaik et al.,198916

Civilian (n � 14). The effect of aggressivecontrol of intracranial pressure on long-term outcome of children with penetratinggunshot wound of the brain was evaluated.

III Used ICP monitoring of severely injured children to decide whetherto operate (debridement, hematoma removal). Thought that thesurvivors in this study had sufficient recovery of physical,behavioral, and intellectual functions to support ICP/CPP-basedmanagement in comatose pediatric victims of gunshot wounds tothe brain.

The Journal of TRAUMA� Injury, Infection, and Critical Care

S14 August Supplement 2001

Given the growing importance of PBI in today’s society, suchresearch seems highly indicated.

References1. Guidelines for the management of severe head injury. Brain Trauma

Foundation, American Association of Neurological Surgeons, JointSection on Neurotrauma and Critical Care. J Neurotrauma. 1996;13:641–734.

2. Guidelines for the management of severe traumatic brain injury.Brain Trauma Foundation, American Association of NeurologicalSurgeons, Joint Section on Neurotrauma and Critical Care.J Neurotrauma. 2000;17:451–627.

3. Brandvold B, Levi L, Feinsod M, George ED. Penetratingcraniocerebral injuries in the Israeli involvement in the Lebaneseconflict, 1982-1985: analysis of a less aggressive surgical approach.J Neurosurg. 1990;72:15–21.

4. Byrnes DP, Crockard HA, Gordon DS, Gleadhill CA. Penetratingcraniocerebral missile injuries in the civil disturbances in NorthernIreland. Br J Surg. 1974;61:169–176.

5. Crockard HA. Bullet injuries of the brain. Ann R Coll Surg Engl.1974;55:111–123.

6. Crockard HA. Early intracranial pressure studies in gunshot woundsof the brain. J Trauma. 1975;15:339–347.

7. Gordon DS. Missile wounds of the head and spine. BMJ. 1975;1:614–616.

8. Graham TW, Williams FC Jr, Harrington T, Spetzler RF. Civiliangunshot wounds to the head: a prospective study. Neurosurgery.1990;27:696–700.

9. Helling TS, McNabney WK, Whittaker CK, Schultz CC, Watkins M.The role of early surgical intervention in civilian gunshot wounds tothe head. J Trauma. 1992;32:398–400.

10. Kaufman HH, Levy ML, Stone JL, et al. Patients with GlasgowComa Scale scores 3, 4, 5 after gunshot wounds to the brain.Neurosurg Clin North Am. 1995;6:701–714.

11. Kaufman HH, Makela ME, Lee KF, Haid RW Jr, Gildenberg PL.Gunshot wounds to the head: a perspective. Neurosurgery. 1986;18:689–695.

12. Levi L, Borovich B, Guilburd JN, et al. Wartime neurosurgicalexperience in Lebanon, 1982-85, II: penetrating craniocerebralinjuries. Isr J Med Sci. 1990;26:555–558.

13. Lillard PL. Five years experience with penetrating craniocerebralgunshot wounds. Surg Neurol. 1978;9:79–83.

14. Miner ME, Ewing-Cobbs L, Kopaniky DR, Cabrera J, Kaufmann P.The results of treatment of gunshot wounds to the brain in children.Neurosurgery. 1990;26:20–25.

15. Nagib MG, Rockswold GL, Sherman RS, Lagaard MW. Civiliangunshot wounds to the brain: prognosis and management.Neurosurgery. 1986;18:533–537.

16. Sarnaik AP, Kopec J, Moylan P, Alvarez D, Canady A. Role ofaggressive intracranial pressure control in management of pediatriccraniocerebral gunshot wounds with unfavorable features. J Trauma.1989;29:1434–1437.

17. Sherman WD, Apuzzo MLJ, Heiden JS, Petersons VT, Weiss MH.Gunshot wounds to the brain: a civilian experience. West J Med.1980;132:99–105.

18. Shoung HM, Sichez JP, Pertuiset B. The early prognosis ofcraniocerebral gunshot wounds in civilian practice as an aid to thechoice of treatment. Acta Neurochir (Wien). 1985;74:27–30.

19. Siccardi D, Cavaliere R, Pau A, Lubinu F, Turtas S, Viale GL.Penetrating craniocerebral missile injuries in civilians: a retrospectiveanalysis of 314 cases. Surg Neurol. 1991;35:455–460.

20. Kirkpatrick JB, DiMaio V. Civilian gunshot wounds of the brain.J Neurosurg. 1978;49:185–198.

21. Raimondi AJ, Samuelson GH. Craniocerebral gunshot wounds incivilian practice. J Neurosurg. 1970;32:647–653.

22. Eisenberg H, Frankowki R, Contant CF, Marshall LF, WalkerMD. High dose barbiturate control of elevated intracranialpressure in patients with severe head injury. J Neurosurg. 1988;69:15–23.

23. Schwartz ML, Tator CH, Rowed DW, Reid SR, Meguro K, AndrewsDF. The University of Toronto head injury treatment study: aprospective, randomized comparison of pentobarbital and mannitol.Can J Neurol Sci. 1984;11:434–440.

24. Ward JD, Becker DP, Miller JD, et al. Failure of prophylacticbarbiturate coma in the treatment of severe head injury. J Neurosurg.1985;62:383–388.

VII. Continued

Authors, Year Description of Study DataClass Conclusions

Siccardi et al.,199119

Civilian (n � 314). Studied the outcome ofpatients with gunshot wounds to the brainpresenting to hospital dead or alive.Operated on all patients withpostresuscitation GCS scores � 3.

III Stated that “there was no significant difference in the outcomebetween patients who were and who were not submitted topostoperative measurement of intracranial pressure.” Authorspresent no data in support of this statement.

GCS, Glasgow Coma Scale; CPP, cerebral perfusion pressure.

ICP Monitoring in the Management of Penetrating Brain Injury

Volume 51 • Number 2 S15