j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 4 ) 1 e6

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.JournalofSurgicalResearch.com

Intracranial pressure monitor in patients with traumatic brain injury Andrew Tang, MD,* Viraj Pandit, MD, Vernard Fennell, MD, Trevor Jones, MD, Bellal Joseph, MD, Terence O’Keeffe, MB ChB, Randall S. Friese, MD, and Peter Rhee, MD Division of Trauma, Emergency Surgery, Critical Care, and Burns, Department of Surgery, University of Arizona, Tucson, Arizona

article info

abstract

Article history:

Background: Brain Trauma Foundation (BTF) guidelines recommend intracranial pressure (ICP)

Received 20 June 2014

monitoring for traumatic brain injury (TBI) patients with a Glasgow Coma Scale score of 8 or less

Received in revised form

with an abnormal head computed tomography, or a normal head computed tomography scan

22 October 2014

with systolic blood pressure 90 mm Hg, posturing, or in patients of age 40. The benefits of

Accepted 12 November 2014

these guidelines on outcome remain unproven. We hypothesized that adherence to BTF

Available online xxx

guidelines for ICP monitoring does not improve outcomes in patients with TBI. Methods: All TBI patients with an admission Glasgow Coma Scale 8 admitted to our level I

Keywords:

trauma center over a 3-y period were identified. Adherence to the individual components of

Traumatic brain injury

our institutional TBI Bundle (ICP monitoring, SpO2 95%, PaCO2 30e39 mm Hg, systolic blood

Intracranial pressure monitoring

pressure 90 mm Hg, cerebral perfusion pressure 60 mm Hg, ICP 25 mm Hg, and temper-

Brain trauma foundation

ature 36 Ce37 C) was assessed. Patients were stratified into two groups as follows: patients

Mortality after traumatic

with ICP monitoring (ICP) and patients without ICP monitoring (no-ICP). Outcome measures

brain injury

were survival and discharge disposition. Multivariate regression analysis was performed. Results: We identified 2618 TBI patients, 261 of whom met the BTF criteria for ICP monitoring. After excluding those with nonsurvivable injuries (n ¼ 67), 194 patients were available for analysis. The two groups were similar in demographics and severity of head injury. Survival rate was higher in the no-ICP group compared with that in the ICP group (98% versus 76%, P < 0.004). Non-monitored patients were discharged with higher levels of function per discharge location (28% home versus 4% home; P < 0.001). Patients without ICP monitoring were 1.21 times more likely to survive compared with that of patients with ICP monitoring (odds ratio: 1.21, 95% confidence interval [1.1e1.9], P ¼ 0.01). In the ICP group, the overall compliance rate to the ICP and cerebral perfusion pressure goals as required by the BTF guidelines was poor. Conclusions: Our data suggest that there is a subset of patients meeting BTF criteria for ICP monitoring that do well without ICP monitoring. This finding should provoke reevaluation of the indication and utility of ICP monitoring in TBI patients. ª 2014 Elsevier Inc. All rights reserved.

Poster presentation at the annual meeting of the American Association for the Surgery of Trauma; September 2013; San Francisco; California. * Corresponding author. Division of Trauma, Critical Care, and Emergency Surgery, Department of Surgery, University of Arizona, 1501 N. Campbell Ave, Room 5411, P.O. Box 245063, Tucson, AZ 85724. Tel.: þ1 520 626 5056; fax: þ1 520 626 5016. E-mail address: [email protected] (A. Tang). 0022-4804/$ e see front matter ª 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jss.2014.11.017

2

1.

j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 4 ) 1 e6

Introduction

Traumatic brain injury (TBI) continues to remain as the leading cause of death and disability among trauma patients [1e4]. According to the Center for Disease Control and Prevention, over 1.7 million individuals sustain TBI annually, resulting in over 53,000 deaths and 500,000 patients developing permanent neurologic damage [3,4]. The primary physical insult is compounded by secondary injury due to expansion of the initial intracranial hemorrhage, disruption of the cerebral blood flow auto-regulation system, brain swelling, and tissue hypoxia [5e7]. Combination of these mechanisms results in elevated intracranial pressures (ICP), which further contributes to adverse outcomes in TBI patients [5e7]. The management of elevated ICP varies greatly in clinical practice with conflicting reports assessing the utility of ICP monitoring on the functional outcomes and survival in patients with TBI [8e14]. According to the Brain Trauma Foundation (BTF) guidelines, ICP monitoring should be performed in all TBI patients with a Glasgow Coma Scale (GCS) score of 8 with an abnormal head computed tomography (CT), or a normal head CT scan with systolic blood pressure (SBP) 90 mm Hg, posturing, or age 40 y [15]. Additionally, the goals prescribed by the BTF guidelines require maintenance of ICP 25 mm Hg and cerebral perfusion pressure (CPP) 60 mm Hg. Based on these guidelines, a vast majority of patients with severe TBI meet the criteria for ICP monitoring. However, only a subset of these patients receive ICP monitoring based on institutional guidelines. A recent prospective multicenter controlled trial demonstrated no difference in outcomes in patients managed with ICP monitoring versus patients managed with an established protocol of neuroimaging and clinical examination [8]. Despite the increasing body of knowledge reexamining the utility of ICP monitoring, there continues to be a paucity of consistent data on ICP monitoring’s impact on patient outcome. The aim of this study was to evaluate outcomes in patients meeting BTF guidelines for ICP monitoring who received ICP monitoring compared with those who did not. We hypothesized that adherence to BTF guidelines for ICP monitoring does not improve outcome in patients with TBI.

2.

Methods

After approval from the Institutional Review Board of the University of Arizona, College of Medicine, we performed a 3year (2010e2012) retrospective analysis of all patients with TBI who presented to our level 1 trauma center. We included all patients who met the BTF guidelines for placement of an ICP monitor [15]. Patients transferred from other institutions and patients with nonsalvageable brain injury were excluded. Our inclusion criteria based on the BTF guidelines for placement of ICP monitor were as follows: 1. GCS score 8 on presentation and an abnormal head CT scan

2. GCS score 8 with a normal head CT and two or more of the features noted on admission: age 40 y, SBP 90 mm Hg, or unilateral or bilateral motor posturing. A single violation from the BTF guidelines at any time point was considered a lack of compliance. Abnormal head CT scan was defined as presence of intracranial hematomas, contusions, swelling, herniation, compressed basal cisterns, or skull fractures. We reviewed patients’ electronic medical records and recorded the following data points: demographics (age, gender, race, and ethnicity), mechanism of injury, vitals on presentation, which included GCS score, SBP, heart rate (RR), temperature (Temp), neurologic examination on presentation, intoxication details, history of loss of consciousness, abnormal posturing, initial and repeat head CT scan findings, GCS score as independently assessed by the trauma and neurosurgery teams, neurosurgical intervention, hospital and intensive care unit (ICU) length of stay, discharge disposition, and in-hospital mortality. We obtained the injury parameters, which included Injury Severity Score and head-Abbreviated Injury Scale from the trauma registry. In patients who received ICP monitoring, we also recorded the ICP, CPP, SBP, temperature, oxygen saturation, and arterial partial pressure of oxygen for the entire duration the ICP monitor was in place. A single investigator reviewed the initial CT scan findings for the type and the size of the intracranial hemorrhage and presence of skull fracture. We defined abnormal neurologic examination as abnormal pupillary reflex and/or focal neurologic deficits in the absence of intoxication. The ICP monitors were placed by the neurosurgeons at our institution. The decisions to place ICP monitors were at the discretion of the attending neurosurgeon. Patients were categorized into two groups as follows: patients with an ICP monitor (ICP) and patients without an ICP monitor (No-ICP). We then compared the demographics and outcomes between the two groups. Our primary outcome measures were hospital length of stay and in-hospital mortality. Our secondary outcome measures were adherence to the individual components of our institutional TBI bundle. Our institutional guidelines included SpO2 95%, PaCO2 30e39 mm Hg, SBP 90 mm Hg, CPP 60 mm Hg, ICP 25 mm Hg, and temperature 36 Ce37 C. Data are reported as mean  standard deviation for continuous descriptive variables, median (range) for ordinal descriptive variables, and as proportions for categorical variables. We performed ManneWhitney U and Student t-test to explore for differences in the two groups (with ICP and without ICP monitors) for continuous variables. We used chisquare test to identify the differences in outcomes between the two groups for categorical variables. Univariate analysis was used to assess for factors associated with survival. Variables with a significant (P  0.2) association per our univariate analysis were then used in a multivariate logistic regression model to identify factors that were independently associated with survival. On multivariate logistic regression analysis, variables were considered significant at P  0.05. All statistical analyses were performed using Statistical Package for Social Sciences (SPSS, version 20; IBM, Inc., Armonk, NY).

j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 4 ) 1 e6

3.

Results

A total of 2618 TBI patients were reviewed, of which 261 met the BTF criteria for ICP monitoring. After excluding those with nonsurvivable injuries (n ¼ 67), 194 patients (ICP ¼ 71, No-ICP ¼ 123) were included for analysis. Figure illustrates the details of the study population. The mean age was 41.2  15.4 y, 82% were males, median GCS was 3 (3e5), and median head-Abbreviated Injury Scale was 4 (4e5). There were no differences in the demographics, injury severity, and presentation vital parameters between patients with and without ICP monitoring. There was no difference in the admission serum sodium level (ICP 132.4  14.2 versus noICP 130.9  16.8, P ¼ 0.52) or the serum sodium levels at 24 h (ICP 138.1  18.2 versus no-ICP 139.5  11.6, P ¼ 0.51) and 48 h (ICP 135.7  9.8 versus no-ICP 134.6  13.1, P ¼ 0.53) time points. Table 1 highlights the demographics of the study population. Table 2 illustrates the intracranial findings on initial head CT scan of the study population. The pattern of intracranial

3

hemorrhage was similar among the cohort of patients with and without ICP monitoring. Subarachnoid hemorrhage was the most common form of intracranial hemorrhage followed by intraparenchymal and subdural hemorrhage. Epidural hemorrhage was the least common in both groups. A total of 40% (n ¼ 74) patients had multiple types of intracranial hemorrhage. The need for surgical intervention did not differ among patients with or without ICP monitoring (P ¼ 0.1), neither were the hospital or ICU lengths of stays. However, patients without an ICP monitor had a higher survival rate (P ¼ 0.004) compared with that of the patients with an ICP monitor. Patients without an ICP monitor were more likely to be discharged with higher levels of function based on their discharge location (28% home versus 4% home; P < 0.001). Table 3 highlights the outcomes of the study population. Results of univariate and multivariate analysis for survival is shown in Table 4. The use of no ICP monitoring was independently associated (P ¼ 0.01) with survival in patient with TBI meeting BTF criteria for ICP monitor placement.

Figure e Details of the study population.

4

j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 4 ) 1 e6

Table 1 e Demographics.

Table 3 e Outcomes.

Demographics

ICP No-ICP monitoring monitoring (n ¼ 71) (n ¼ 123)

Age, y, (mean  SD) Males, % GCS on admission, median (range) SBP on admission, (mean  SD) LOC, % ISS, median (range) Head AIS, median (range) Alcohol, % GCS by neurosurgeon (mean  SD)

P

38.8  15.1 79 3 (3e5)

41.6  16 84 3 (3e5)

0.3 0.3 0.7

143.8  33.7 70 26 (16e28) 4 (4e5) 62 5 (4e6)

146.4  29.3 66 26 (14e28) 4 (4e5) 64 5 (4e6)

0.6 0.4 0.08 0.09 0.5 0.3

Outcomes

ICP monitoring (n ¼ 71)

Hospital LOS 16.4  15.1 ICU LOS 8.9  7.8 Ventilation days 7.6  6.8 Craniectomy, n (%) 13 (18) Survival, n (%) 54 (76) Discharge disposition, n (%) Transferred 10 (14) Discharged home 3 (4) Discharged rehab 30 (42) Discharged hospice 3 (4) Discharged SNiF 8 (9)

No-ICP monitoring (n ¼ 123)

P

15.8  11.2 8.1  7.1 7.1  6.6 36 (29) 120 (98)

0.75 0.46 0.61 0.09 0.0001

4 34 68 3 14

(3) (28) (55) (2) (11)

0.005 0.001 0.08 0.4 0.9

AIS ¼ Abbreviated Injury Score; ISS ¼ Injury Severity Score; LOC ¼ loss of consciousness; SD ¼ standard deviation.

LOS ¼ length of Stay; Rehab ¼ rehabilitation center; SNiF ¼ skilled nursing facility.

The overall compliance to the institutional TBI bundle was poor. Among patients with ICP monitoring, only 64% patients maintained the ICP 25 mm Hg, and 67% achieved CPP 60 mm Hg at all times. Among all patients, normothermia and normocarbia were poorly achieved goals. There was no statistical difference for adherence to any of the components of the institutional TBI bundle between patients with and without ICP monitoring. Table 5 demonstrates the compliance with the components of the institutional TBI bundle.

redefining the BTF guidelines for ICP monitoring in patients with severe TBI is warranted. The reported impact of ICP monitoring on outcomes in patients with severe TBI is variable [8e14]. Studies have advocated the use of ICP monitors in severe TBI patients because of the potential survival benefit [9e14]. Lane et al. in a retrospective review supported the use of ICP monitors citing the information gained from ICP monitoring may have allowed for more informed management decisions, thus resulting in better survival. However, the authors also acknowledged the likelihood that ICP monitoring may be a surrogate for the overall aggressiveness of neurointensive care, and not necessarily singly responsible for better outcome [12]. Talving et al. [11] in a prospective study of severe blunt TBI patients with GCS 8 demonstrated significantly higher in-hospital survival rate in patients who underwent

4.

Discussion

The impact of ICP monitoring on outcomes in patients with severe TBI remains unclear [8e14]. In our cohort of patients meeting BTF criteria for ICP monitoring, patients without ICP monitors had a higher survival rate and were more likely to be discharged home compared with that of patients managed with ICP monitors. Additionally, in patients with ICP monitors, the overall compliance rate to the BTF guidelines for ICP and CPP parameters was poor. This study questions the role of ICP monitoring in patients with severe TBI. We believe that

Table 4 e Univariate and multivariate analysis for survival. Variables

Univariate analysis Multivariate analysis OR (95% CI)

Table 2 e Findings on initial CT. Findings on initial CT head Number of RHCT Skull fracture, % ICH, % SDH EDH SAH IVH IPH

ICP monitoring (n ¼ 71)

No-ICP monitoring (n ¼ 123)

P

2.9  0.8 63

3.1  1.2 64

0.21 0.5

57 20 66 22 61

44 9 61 13 49

0.08 0.06 0.8 0.1 0.08

EDH ¼ epidural hemorrhage; ICH ¼ intracranial hemorrhage; IPH ¼ intraparenchymal hemorrhage; IVH ¼ intraventricular hemorrhage; RHCT ¼ repeat head CT scan; SAH ¼ subarachnoid hemorrhage; SDH ¼ subdural hemorrhage.

Age Male Alcohol intoxication Head-AIS GCS score Hypotension (SBP 90 mm Hg) Skull fracture SDH EDH No-ICP monitoring

P

0.86 (0.56e0.98) 0.04 1.3 (0.78e2.7) 0.3 1.1 (0.42e5.1) 0.9

OR (95% CI) 0.91 (0.78e1.1) d d

P 0.13 d d

0.63 (0.58e0.85) 0.001 0.71 (0.68e0.85) 0.69 (0.61e0.79) 0.001 0.8 (0.75e0.93) 0.83 (0.76e0.95) 0.02 0.91 (0.74e1.2)

0.01 0.03 0.3

0.92 0.82 0.89 1.3

0.5 0.1 0.6 0.01

(0.83e1.1) (0.69e0.91) (0.81e1.1) (1.25e2.5)

0.1 0.04 0.09 0.001

0.95 0.96 0.98 1.21

(0.86e1.2) (0.79e1.1) (0.84e1.6) (1.1e1.9)

AIS ¼ Abbreviated Injury Scale; CI ¼ confidence interval; EDH ¼ epidural hemorrhage; OR ¼ odds ratio; SDH ¼ subdural hemorrhage. Significant P values (0.2 for univariate and

Intracranial pressure monitor in patients with traumatic brain injury.

Brain Trauma Foundation (BTF) guidelines recommend intracranial pressure (ICP) monitoring for traumatic brain injury (TBI) patients with a Glasgow Com...
390KB Sizes 2 Downloads 11 Views