A Retrospective Analysis of Intracranial Pressure Monitoring and Outcomes in Adults after Severe Traumatic Brain Injury at Kaiser Permanente Trauma Centers



 

Kaveh Barami, MDPhD1; Jessica Pemberton, MSN2; Amit Banerjee, MD3; Jason London, MDPhD4; William Bandy, MD2

Perm J 2021;25:20.293

https://doi.org/10.7812/TPP/20.293
E-pub: 05/19/2021

Background: The role of intracranial pressure (ICP) monitoring in improving outcomes after severe traumatic brain injury especially at level II trauma centers remains controversial. A retrospective analysis was undertaken to assess the impact of ICP monitoring on mortality and long-term functional outcome in adults after severe traumatic brain injury at level II trauma centers.

Methods: The data were extracted from the Kaiser Permanente trauma database. Inclusion criteria were adults (≥ 18 years) with severe traumatic brain injury (Glasgow Coma Scale score, < 9) admitted to 2 level II trauma centers in Northern California from 2014 to 2019.

Results: Of 199 patients, 58 (29.1%) underwent ICP monitoring. The monitored subgroup was significantly younger (< 65 years), had lower Glasgow Coma Scale scores (3–5), underwent cranial procedures (craniotomy or decompressive craniectomy) more often, and had greater injury severity scores (≥ 15). Despite monitored patients being more severely injured, there was no significant difference in mortality or 6-month favorable outcomes between monitored and nonmonitored patients, including patients who underwent cranial procedures. Increased monitoring frequency and reduction in overall mortality was seen throughout the study period yet with a parallel reduction in both groups.

Conclusion: ICP monitoring may not impact in-patient mortality or long-term outcomes at level II trauma centers. Improved outcomes may be more related to identifying patients who may benefit from ICP-guided therapy rather than simply increasing the overall use of it. Last, our pattern of care and outcomes are comparable to level I trauma centers and our findings may serve as a benchmark for future studies.

INTRODUCTION

Traumatic brain injury (TBI) is one of the leading causes of death and disability worldwide, affecting approximately 10 million people annually.1,2 In the US it contributes to nearly one- third of injury-related deaths and is a major cause of severe long-term disability requiring 24-hour care.3 High intracranial pressure (ICP) (defined as ICP > 20 mmHg) is associated with increased mortality and disability in TBI.4,5 Although the association between high ICPs and poorer outcomes is generally accepted, the utility of ICP monitoring remains controversial. Various studies have shown that ICP-monitored patients have a lower risk of mortality,6-10 and an increased frequency of good recovery and favorable outcome.11 In contrast, other studies have failed to see a survival benefit12-14; several reports have actually demonstrated an association between ICP monitoring and increased mortality and complications.12,15,16 Moreover, a large multicenter randomized controlled study published in 2012 demonstrated that there was no difference in survival comparing ICP monitoring vs care based on clinical exam and imaging.17 As a result, despite recommendations by the Brain Trauma Foundation guidelines,18 in some states there has been a significant decline in the use of ICP monitoring since 2014.19 Specifically, Brain Trauma Foundation guidelines recommend monitoring ICP in salvageable patients with a Glasgow Coma Scale (GCS) score of 3 to 8 and abnormal computed tomography (CT) or normal CT if 2 or more of the following are noted: admission age more than 40 years, unilateral or bilateral posturing or systolic blood pressure (SBP) < 90 mmHg.18 ICP monitoring has also been implicated in discrepancies in performance between level I and level II trauma centers. For example, in 2012, Barmparas et al20 reported decreased overall frequency of ICP-guided therapy of severe TBI patients admitted to designated level II trauma centers was associated with increased mortality compared to level I centers.

Patterns of care and outcomes at level II trauma centers are crucial to optimizing planning and triaging strategies for severe TBI patients in our current health-care environment. According to a meta-analysis study, studies published after 2012 have indicated a lower mortality in patients undergoing monitoring.14 However, there has been no study since the report by Barmparas et al20 that has looked at ICP monitoring and mortality specifically at level II trauma centers. No study has ever reported long-term outcomes at level II trauma centers or performed a temporal analysis to determine whether changes in frequency of ICP monitoring per year correlate with annual changes in mortality.

Therefore, the aims of this study were to assess the impact of monitoring on outcomes, including mortality and 6-month functional outcome, in patients with severe TBI at level II trauma centers. We also sought to determine whether mortality trends correlated with yearly frequency of ICP monitoring.

METHODS

Data Source and Study Population

Trauma databases from 2 level II trauma centers located in northern California and covered by the same group of neurosurgeons, affiliated with Kaiser Permanente, the largest integrated managed care consortium in the US, were used as the data source. The databases accumulate data prospectively from the electronic medical records of all trauma patients admitted to these trauma centers according to American College of Surgeons guidelines. Patients with severe TBI, as defined by a GCS score of less than 9, who were 18 years of age or older and were admitted to the intensive care unit between January 2014 and August 2019 were included in the study. Patients who died in the emergency department were excluded. Our institutional review board approved the study; patient consent was not required.

Covariates, Analysis of ICP-monitored Patients and Outcomes

Covariates extracted from the databases included patient demographics and emergency department clinical characteristics, including mechanism of injury, vitals, injury severity score (ISS), GCS score, pupillary response (abnormal response included fixed and dilated pupil, sluggish or unreactive pupil), admission head CT finding of midline shift (defined at the foramen of Monro calculated by the difference between the biparietal diameter divided by 2 and the distance from the inner table to the septum pellucidum on the side of the shift), toxicology including alcohol, ICP monitor placement, and whether patients underwent cranial procedures. ICP monitoring included either placement of an external ventricular drain (EVD) or intraparenchymal bolt monitor. The decision to place an ICP monitor was left to the attending neurosurgeon. ICP values were recorded hourly. A subgroup analysis of the ICP-monitored patients included type of monitor placed, timing of monitor placement (initially within the first 24 hours of admission and prior to any cranial procedures vs concurrent with the cranial procedure or delayed fashion). We also looked at conversion rates—that is, patients who underwent initial monitoring yet later required cranial procedures to control their ICPs resulting from the failure of medical management. The primary outcome was in-hospital mortality; cause of death was determined for each patient. The secondary outcome was the 6-month score, where death = 1, persistent vegetative state = 2, severe disability = 3, moderate disability = 4, and good recovery = 5. GCS scores 4 or 5 were considered a favorable outcome; GCS scores of 1 to 3 were considered unfavorable.

ICP Treatment Protocol

All patients with severe TBI were intubated, admitted to the intensive care unit, and underwent initial treatment for ICP as follows: head-of-bed elevation to 30°, maintaining the head and neck in a neutral position; adequate sedation, analgesics, and/or neuromuscular paralysis using train-of-4 monitoring; monitoring mean arterial blood pressure to avoid hypotension; maintaining SBP > 90 mmHg, mean arterial blood pressure > 70 mmHg, normothermia goal < 38.1°C, mechanical ventilation, saturated oxygen goal > 90%, and partial pressure of oxygen > 60 mmHg; pharmacological prophylaxis for early posttraumatic seizures; and vasopressors when necessary to obtain an SBP > 90 mmHg or mean arterial blood pressure > 70 mmHg. In monitored patients, our goal was also to maintain cerebral perfusion pressure at 70 mmHg. Patients with abnormal head CT received hyperosmolar therapy with either mannitol (intermittent boluses of 0.25–1 g/kg body weight) or hypertonic saline boluses of 3% sodium chloride to maintain sodium ≥ 145 mEq/L (maximum, 160 mEq/L for hypertonic saline or 320 mOsm/L for mannitol). Extra interventions were done for ICPs > 22 mmHg and included administering additional doses of mannitol or hypertonic saline to achieve sodium levels or osmolarity closer to the maximum limits, elevating the head of the bed more than 30°, increasing paralysis/sedation and/or increasing cerebrospinal fluid drainage (if an EVD was in place).

Statistical Analysis

The chi-square test was used to compare characteristics between survivors and nonsurvivors as well as differences between patients treated with and without the use of ICP monitors. The means are expressed ± standard deviation. Multivariable logistic regression models predicting in-hospital mortality for the whole sample were used to evaluate the association between ICP monitoring status and in-hospital mortality, controlling for age, gender, hypotension on day 1, ISS, cranial procedure, initial GCS score, pupillary status on day 1, and midline shift > 5 mm on admission CT. The odds ratios, 95% confidence intervals, and p values of the covariates were reported as well as the area under the receiver operating characteristic curve. Statistical significance was defined as p < 0.05. Correlation coefficients (r) were calculated to quantify the linear relationship between frequency of monitoring per year and overall mortality. To evaluate the normality of our data, we performed the skewness and Kurtosis tests on all noncategorical data sets (age, blood pressure, GCS score, and ISS). All values were within the –3.29 and +3.29 range (for n < 300), suggesting normal distribution. Data were analyzed with R version 3.6.2 (R Foundation for Statistical Computing, Vienna, Austria).

RESULTS

Characteristics of Study Population

Between 2014 and 2019, after applying inclusion criteria, 199 adult patients, age 18 years or older with severe TBI treated at 2 level II trauma centers were included in the study. Table 1 shows the characteristics of the studied population. The most common mechanism of injury was vehicular accidents (42%), followed by falls (31%). Mean age was 47.1 ± 20.2 years. Approximately 80% of the population were younger than 65 years old. The mean GCS score was 5.04 ± 2.01, and roughly equal numbers had a GCS score of 3 to 5 (53.8%) and 6 to 8 (46.2%). Hypotension on admission was present in nearly one-third of the population, and approximately 40% had midline shift on admission CT. Approximately 65% had abnormal pupillary responses. The mean ISS was 27.1 ± 9.8, and approximately 90% of the population had an ISS ≥ 15, which is considered to be severe. Frequency of ICP monitoring was nearly 30%. Nearly 45% of the population had positive toxicology, including alcohol. Approximately 30% underwent cranial procedures. Nearly all deaths (95%) occurred within the first 2 weeks of admission. Overall in-hospital mortality was 40.7%.

Table 1. Baseline characteristics of adult (age, ≥18 y) study population

Characteristic n (%)
No. of patients 199
Age, y
 Median ± SD 47.1 ± 20.2
 Range 18–96
 ≥ 65 43 (21.6)
 < 65 156 (78.4)
Gender
 Male 151 (75.9)
 Female 48 (24.1)
Initial GCS score, pts
 Mean ± SD 5.04 ± 2.01
 6–8 92 (46.2)
 3–5 107 (53.8)
Systolic BP < 90 mmHg, day 1
 Yes 60 (30.2)
 No 139 (69.8)
Midline shift ≥ 5 mm
 Yes 76 (38.2)
 No 123 (61.8)
Abnormal pupillary response
 Yes 131 (65.8)
 No 68 (34.2)
ISS, pts
 Mean ± SD 27.13 ± 9.8
 Range 1–57
 ≥ 15 176 (88.4)
 < 15 23 (11.6)
ICP monitoring
 Yes 58 (29.1)
 No 141 (70.9)
Craniotomy/craniectomy
 Yes 61 (30.7)
 No 138 (69.3)
Toxicology
 Positive 90 (45.2)
 Negative 109 (54.8)
Outcome
 Alive 118 (59.3)
 Dead 81 (40.7)
 6-Mo favorable outcome, Glasgow outcome score of 4 or 5 82 (41.2)
 6-Mo unfavorable outcome, Glasgow outcome score of 1–3 117 (58.8)

BP = blood pressure; GCS = Glasgow Coma Scale; ICP = intracranial pressure; SD = standard deviation.

Differences between Survivors and Nonsurvivors

Table 2 shows the differences between survivors and nonsurvivors. There was a significant increase in mortality in patients 65 years and older, admission GCS score 3 to 5, SBP < 90 mmHg on day 1, midline shift ≥ 5 mm, abnormal pupillary response, and ISS ≥ 15. There was no significant difference in mortality between patients with ICP monitoring and those without monitoring.

Table 2. Differences between survivors and nonsurvivors

Variable Alive, n (%) Dead, n (%) p Value
No. of patients 118 (59.3) 81 (40.7)  
Age, y
 ≥ 65 13 (11) 30 (37) < 0.001
 < 65 105 (89) 51 (63)  
Gender
 Male 93 (78.8) 58 (71.6) 0.24
 Female 25 (21.2) 51 (63)  
GCS score, pts
 6–8 71 (60.2) 21 (25.9) < 0.001
 3–5 47 (39.8) 60 (74.1)  
Systolic BP < 90 mmHg
 Yes 6 (5.1) 17 (21) < 0.001
 No 112 (94.9) 64 (79)  
Midline shift ≥ 5 mm
 Yes 31 (26.3) 46 (56.8) < 0.001
 No 87 (73.7) 35 (43.2)  
Abnormal pupillary response
 Yes 32 (27.1) 55 (67.9) < 0.001
 No 86 (72.9) 26 (32.1)  
ISS, pts
 ≥ 15 96 (81.4) 80 (98.8) < 0.001
 < 15 22 (8.6) 1 (1.2)  
ICP monitor
 Yes 36 (30.5) 22 (27.2) 0.61
 No 82 (69.5) 59 (72.8)  
Craniotomy/craniectomy
 Yes 39 (33.1) 22 (27.2) 0.38
 No 79 (66.9) 59 (72.8)  
Toxicology
 Positive 75 (63.6) 42 (51.9) 0.1
 Negative 43 (36.4) 39 (48.1)  
 Monitored patient with cranial procedure 19 (57.6) 14 (42.4) 0.26
 Nonmonitored patient with cranial procedure 20 (71.4) 8 (28.6)  

BP = blood pressure; ICP = intracranial pressure; ISS = injury severity score.

Independent Predictors of Mortality

Logistic regression analyses showed that ICP monitoring was not a statistically significant predictor of mortality (adjusted odds ratio, 1.95; 95% confidence interval, 0.62–6.16; p = 0.25) after controlling for 8 covariates. Independent significant risk factors included age ≥ 65 years, admission GCS score of 3 to 5, pupillary abnormality, and midline shift ≥ 5 mm. Logistic regression analyses had high discrimination, with the area under the receiver operating characteristic curve = 0.86. Results are shown in Table 3.

Table 3. Logistic regression analyses predicting mortality for all patients

Predictor variable Adjusted OR (95% CI) p Value
Age, y
 ≥ 65 0.23 (0.08–0.68) 0.006
 < 65 Reference  
Gender
 Male 1.95 (0.73–5.23) 0.18
 Female Reference  
Systolic BP, mmHg
 < 90 0.28 (0.06–1.19) 0.08
 ≥ 90 Reference  
GCS score, pts
 3–5 0.37 (0.14–0.98) 0.04
 6–8 Reference  
ISS, pts
 ≥ 15 0.32 (0.03–3.06) 0.27
 < 15 Reference  
Craniotomy/craniectomy
 Yes 1.1 (0.31–3.87) 0.88
 No Reference  
ICP monitor
 Yes 1.95 (0.62–6.16) 0.25
 No Reference  
Pupillary abnormality
 Yes 0.15 (0.06–0.39) < 0.001
 No Reference  
Midline shift ≥ 5 mm
 Yes 0.28 (0.1–0.8) 0.02
 No Reference  

BP = blood pressure; CI = confidence interval; GCS = Glasgow Coma Scale; ICP = intracranial pressure; ISS = injury severity score; OR = odds ratio.

Differences between Patients with ICP Monitoring and No Monitoring

Table 4 shows the characteristics of our study population based on ICP monitoring status. Patients who had ICP monitoring were significantly younger (93.1% vs 72.3%, p = 0.0012), had initial GCS scores of 3 to 5 (65.5% vs 48.9%, p = 0.033), underwent cranial procedures significantly more frequently (56.9% vs 19.9%, p < 0.000001), and had higher ISSs (98.3% vs 84.4%, p = 0.005). Otherwise, there were no significant differences in terms of gender, day 1 hypotension, midline shift, pupillary response, or outcome between monitored and nonmonitored patients. There was no significant difference in mortality between monitored patients who underwent cranial procedures vs nonmonitored patients (42.4% vs 28.6%, p = 0.26). There was no significant difference in 6-month favorable outcomes between monitored vs nonmonitored patients (36.2% vs 43.3%, p = 0.36), nor between these groups if they underwent cranial procedures (36.3% vs 46.4%, p = 0.43).

Table 4. Characteristics of study population based on intracranial pressure (ICP) monitoring status

Characteristic No ICP monitoring, n (%) ICP monitoring, n (%) p Value
No. of patients 141 (70.9) 58 (29.1)  
Age, y
 Mean ± SD 50.4 ± 20.8 39.2 ± 16.4  
 Range 18–96 20–81  
 ≥ 65 39 (27.7) 4 (6.9) 0.001
 < 65 102 (72.3) 54 (93.1)  
Gender
 Male 108 (76.6) 43 (74.1) 0.71
 Female 33 (23.4) 15 (25.9)  
Initial GCS score, pts
 Mean ± SD 5.2 ± 2.0    
 6–8 72 (51.1) 20 (34.5) 0.03
 3–5 69 (48.9) 38 (65.5)  
Systolic BP < 90 mmHg, day 1
 Yes 18 (12.8) 5 (8.6) 0.41
 No 123 (87.2) 53 (91.4)  
Midline shift ≥ 5 mm
 Yes 53 (37.6) 24 (41.4) 0.62
 No 88 (62.4) 34 (58.6)  
Abnormal pupillary response
 Yes 56 (39.7) 31 (53.4) 0.08
 No 85 (60.3) 27 (46.6)  
ISS, pts
 Mean ± SD 25 ± 8.9 32.4 ± 10.1  
 ≥ 15 119 (84.4) 57 (98.3) 0.005
 < 15 22 (15.6) 1 (1.7)  
Craniotomy/craniectomy
 Yes 28 (19.9) 33 (56.9) < 0.001
 No 113 (80.1) 25 (43.1)  
Outcome
 Alive 82 (58.2) 36 (62.1) 0.61
 Dead 59 (41.8) 22 (37.9)  
 6-Mo favorable outcome, Glasgow outcome score of 4 or 5 61 (43.3) 21 (36.2) 0.36
 6-Mo unfavorable outcome, Glasgow outcome score of 1–3 80 (56.7) 37 (63.8)  
 With cranial procedures, 6-mo favorable outcome, Glasgow outcome score of 4 or 5 13 (46.4) 12 (36.3) 0.43
 With cranial procedures, 6-mo unfavorable outcome, Glasgow outcome score of 1–3 15 (53.6) 21 (63.6)  

BP = blood pressure; GCS = Glasgow Coma Scale; ISS = injury severity score; SD = standard deviation.

Relationship of Frequency of ICP Monitoring per Year and Overall Mortality Trends

Mortality based on frequency of monitoring per year is shown in Table 5 and Figure 1. As frequency of monitoring increased from 20% in 2014 to 41.7% in 2019, overall mortality was reduced from 42.9% to 25% (r = –0.84). Further analysis was also performed to assess mortality rates of monitored and nonmonitored patients per year. The correlation between decline in overall mortality and mortality trends of the monitored patients compared to nonmonitored patients were nearly identical (r = –0.82 vs –0.83, respectively).

Table 5. Frequency of intracranial pressure (ICP) monitoring per year and mortality trends

Year Frequency monitoring/y, % Overall mortality, % ICP-monitored mortality, % Nonmonitored mortality, %
2014 20 42.9 33.3 45.5
2015 26.7 40.6 50 37.5
2016 28.3 45.0 47.1 44.2
2017 19 43.9 50 42.4
2018 42.5 35.0 23.5 43.5
2019 41.7 25.0 20 28.6

Correlation coefficients: Frequency monitoring/y and overall mortality, –0.84; frequency monitoring/y and ICP-monitored mortality, –0.82; frequency monitoring/y and nonmonitored mortality, –0.83.

tpj20293f1 copy

Figure 1. Graph showing monitoring frequency per year and annual overall, monitored and nonmonitored mortality trends.

Analysis of 58 ICP-monitored Patients

Of the 22 patients who died in the ICP-monitored group, most (n = 16, 72.7%) had uncontrollable ICP values—that is, ICPs greater than a threshold value of 36 mmHg sustained over 10 minutes. Sixty percent of nonmonitored patients had refractory intracranial hypertension as determined by clinical examination. The remainder of deaths were the result of associated brainstem injuries, anoxic brain injury, pulmonary embolism, and cardiac or respiratory issues.

With regard to cerebrospinal fluid drainage in patients with EVDs, we left the ventricular drain open continuously in patients with high ICP values. We did not determine whether patients with bolts needed more interventions to control ICP compared to patients with EVDs. Patients with ICP values < 22 mmHg (n = 32, 55%) did not require additional interventions and their monitors were removed after a few days (on average, 1–3 days).

There were nearly twice as many EVDs inserted vs bolts placed (n = 40 vs 18) and there was no statistically significant difference in mortality between the 2 groups (42.5% vs 27.8%, p = 0.29). With regard to timing of insertion of the monitors, most monitors (58.6%) were placed initially (within 24 hours of admission prior to any cranial procedure), the remainder (41.4%) were placed concurrent with the cranial procedure or in a delayed fashion (at least 24 hours after the cranial procedure). In all cases, monitors were placed within the first 72 hours of admission. There was no significant difference in mortality between the initial monitoring group vs the concurrent/delayed group (32.4% vs 45.8%, p = 0.30). The conversion rate—that is, patients in the initial monitoring group who were refractory to medical management who subsequently underwent decompressive craniectomy—was 13.8% (n = 8), half of whom died. These results are summarized in Table 6.

Table 6. Characteristics of 58 intracranial pressure (ICP)-monitored patients

Characteristic n (%) p Value
Mortality of patients with ICP monitoring 22 (37.9)  
Death resulting from uncontrollable ICP 16 (72.7)  
Death resulting from non-ICP causes 6 (27.3)  
Conversions 8 (13.8)  
Mortality of patients with conversion 4 (6.9)  
Type of monitor
 External ventricular drain 40 (69)  
 Bolt 18 (31)  
 Mortality of patients with external ventricular drain 17 (29.3) 0.29
 Mortality of patients with bolt 5 (8.6)  
Timing of monitor placement
 Initial 34 (58.6)  
 Concurrent with surgery or delayed 24 (41.4)  
ICP profile
 < 22 mmHg 32 (55.2)  
 22–36 mmHg 10 (17.2)  
 > 36 mmHg 16 (27.6)  

DISCUSSION

The main findings of this study include 1) there was a predilection to monitor younger patients with more severe injuries based on GCS scores and ISSs; 2) there was no significant difference in mortality or long-term functional outcomes between monitored and nonmonitored patients, including patients who underwent cranial procedures; 3) ICP monitoring was not a significant predictor of mortality; and 4) during the study period, there was an increased frequency of monitoring and a parallel reduction in mortality in ICP monitored and nonmonitored groups.

ICP has been established as a main predictor of mortality and outcome after severe TBI since Lundberg’s initial report in 1960.21-24 Despite this, the role of ICP monitoring in guiding management of intracranial hypertension early after trauma remains controversial. Many reports suggest monitoring is associated with decreased mortality,6-10 superior functional outcome,11 faster recovery,25 reduced intensive cat unit stay,25 reduced radiation exposure,25 reduced brain-specific treatments,25 and reduced costs.26 Yet others have concluded that monitoring has no beneficial effect,13,14,17 is associated with poor outcomes,16 and may even increase mortality,15,16 hospital resource utilization,12 and complications.12 Another area where the role of ICP monitoring remains controversial is the apparent discrepancy in performance between level I and level II centers. Several reports have shown increased mortality of patients with severe TBI at level II centers,20,27-29 and that this may be related to decreased frequency of ICP monitoring.20

Our study differs from the other studies focusing on level II performance in several ways. First, instead of extracting data from national or statewide trauma databases, which have inherent heterogeneity in terms of treating physicians and treatment protocols, we were able to compare mortality and long-term outcomes between groups of patients treated at the same trauma centers and by the same neurosurgeons and ICP treatment protocols. Second, in addition to mortality and long-term outcomes, we also looked at different types and timing of ICP monitoring as well as causes of death. Third, and last, a temporal analysis was done to determine whether frequency of monitoring per year affected mortality rates.

Outcomes after Severe TBI at Level II Trauma Centers

We found there was no significant difference in mortality or long-term functional outcomes between monitored and nonmonitored patients, including patients who underwent cranial procedures despite the fact that monitored patient were injured more severely. We also found that ICP monitoring was not a significant predictor of mortality. There are several plausible explanations for these findings: 1) the effect of the empiric ICP treatment protocol on all severe TBI patients regardless of their ICP monitoring status; 2) despite monitored patients being more injured, the negative effects of their injuries might have been mitigated by their younger age; 3) efficient access to physical/occupational therapy and early intensive neurorehabilitation programs in our integrated health-care system; and 4) ICP monitoring might have actually had a positive impact on the outcomes. We are unable to distinguish between these possibilities or combinations of them. Our findings are consistent with several reports assessing the impact of ICP monitoring on long-term outcomes demonstrating that monitoring status was not associated with a statistically significant difference in overall unfavorable13 or favorable outcome10

Frequency of ICP Monitoring and Mortality at Level II Centers

Previous reports have suggested inferior outcomes, including survival after severe TBI at level II centers when compared to level I centers.20,27-29 To date, there is only 1 published study that has looked specifically at frequency of ICP monitoring at level II centers.20 In this 2012 report by Barmparas et al,20 data were extracted from the national trauma databank from 2007 to 2008 comparing level I and level II centers. There was a significant difference in ICP monitoring frequency (11.0% vs 14.7%) as well as mortality (32.3% vs 30.8%) comparing level II to level I centers. The authors concluded that decreased use of ICP monitoring at level II centers was associated with increased mortality. The overall frequency of ICP monitoring in the present study was 29.1%, with an overall mortality of 40.7%. For the 6-year study period, the frequency of monitoring doubled from 20% to 41.7%. This correlated to a nearly identical decrease in both ICP monitored and nonmonitored patient mortality (25%–35% in the last 2 years of the study). This suggests that the reduced mortality in the monitored group may not have been related to frequency of ICP monitoring. This is in contrast to the conclusions of Barmparas et al20 and suggests that merely increasing the use of ICP monitoring may not reduce mortality.

It is also unclear whether there is a difference in overall mortality between level I and level II trauma centers. Although various studies have shown increased mortality at level II centers,20,27-29 other investigations have failed to demonstrate this.30,31 Comparative studies have reported overall mortality rates ranging as low as 13.9% vs 9.6% (level II vs level I)28 to as high as 32.3% vs 30.8% (level II vs level I).20 Our overall in-hospital mortality rate was 40.7% during the study period spanning 6 years. When mortality was analyzed for each year, our overall mortality during the last year of the study was reduced to 25% (from 42.9% in the first year of the study). This shows that survival may fluctuate throughout the course of a study period.

Role of Cranial Procedures, and Type and Timing of ICP Monitoring on Mortality

We found that there was an approximately 15% difference between mortality of monitored patients undergoing cranial procedures vs nonmonitored patients (42.4% vs 28.6%), yet this difference was not statistically significant (p = 0.26). Moreover, there was no statistically significant difference in functional outcomes between monitored patients who underwent cranial procedures vs nonmonitored patients undergoing cranial procedures. The increased mortality was most likely a result of monitored patients undergoing cranial procedures being more injured. We found that ICP monitoring after cranial procedures was useful in guiding therapy after cranial procedures, which is consistent with the report by Picetti et al32 on the beneficial effects of ICP monitoring after primary decompressive craniectomy.

We also found there was no significant difference in mortality between timing (initial vs concurrent with cranial procedure) or type of monitoring device (EVD vs bolt). Regarding the latter finding, this is consistent with the systematic review and meta-analysis published by Volovici et al,33 who looked at EVDs and bolt monitors and found no significant difference in mortality between the 2, although EVDs had more complication rates. On the other hand, Liu et al,34 in an observational prospective study, showed that EVDs had an advantage over bolts in managing refractory intracranial hypertension and were associated with improved survival.

Patients Who May Benefit from ICP Monitoring

This study showed there was a predilection to place monitors in younger and more injured patients based on GCS score and ISS. This is consistent with multiple studies that have concluded that much of the decision to perform invasive monitoring is made on an individual basis and reflects the neurosurgeon’s complex risk assessment of which patient stands to gain the most benefit from it, rather than strict adherence to Brain Trauma Foundation guidelines 10,35,36

We found that the majority of our monitored patients had ICPs < 22 mmHg and did not require extra interventions beyond the baseline treatment of ICP. This suggests that a select group of patients benefited from ICP-guided therapy. The fact that a minority of patients with severe TBI generated high ICPs is consistent with several studies that have looked at the relationship between ICP and outcomes.37,38 There is also mounting evidence that critical ICP thresholds cannot be generalized,39,40 and that subgroups such as females and older people may have lower ICP thresholds for favorable outcome.41 There is an emerging concept that key processes implicated in the pathophysiology of TBI, including cerebral pressure autoregulation, ICP, and cerebral perfusion pressure, currently seem inadequately defined and deserve further investigation.42 This has led to a paradigm shift in the management of TBI, consisting of maintaining normal partial pressure oxygen and carbon dioxide, temperature, volume status, and serum glucose in addition to titrating ICP and cerebral perfusion pressure to individual patients.42 Taken together, it may be that merely increasing the frequency of monitoring may not improve survival, yet a select group of patients with severe TBI may benefit from ICP-guided therapy and that, in the future, a stratification scheme might be applicable.

Limitations

Our study has certain limitations. There are inherent problems with a retrospective design, and patients were not randomized to receive ICP monitoring or not. The study population was limited and possibly underpowered to detect significant differences, yet still comparable to other published works. Rationale for decisions to place ICP monitors and reasons why there were changes in frequency of monitoring and mortality per year were not investigated. Lack of a standard protocol to monitor patients, a predilection of monitoring more severely injured patients or, conversely, to forgo monitoring in patients who may have more benign CT scans or were not salvageable may have resulted in discrepancy between groups. There was reliance on accurate documentation in the electronic medical records in this study that could not be controlled. Last, propensity scoring was not done because omitted variable bias might have still been an issue. ICP monitoring biases were taken into account in logistic regression analyses, which had high discrimination (area under the receiver operating characteristic curve = 0.86). Despite these limitations, we provide new information regarding ICP monitoring and TBI-related survival and long-term outcomes at level II trauma centers.

CONCLUSION

There was a predilection to monitor younger and more severely injured patients in this study. There were no significant differences in mortality or long-term outcomes between monitored vs nonmonitored patients, including patients who underwent cranial procedures. Merely increasing the overall use of ICP monitoring may not improve survival. Future studies are needed to determine whether identifying a select group of patients who may benefit from ICP-guided therapy will improve outcomes. The findings of this study may serve as a benchmark for performance at level II trauma centers.

Disclosure Statement

The author(s) have no conflicts of interest to disclose.

Financial Support

No funding was provided in support of this study.

Author Affiliations

1Department of Neurosurgery, Kaiser Vacaville, Vacaville, CA

2Department of Trauma Surgery, Kaiser Vacaville, Vacaville, CA

3Department of Neurosurgery, Kaiser Sacramento, Sacramento, CA

4Department of Trauma Surgery, Kaiser South Sacramento, Sacramento, CA

Corresponding Author

Kaveh Barami, MD, PhD (kbarami@yahoo.com)

Author Contributions

All authors contributed to study design, data collection, data analysis, and manuscript preparation.

References

1. Hyder AA, Wunderlich CA, Puvanachandra P, Gururaj G, Kobusingye OC. The impact of traumatic brain injuries: A global perspective. Neurorehabilitation 2007;22(5):341–53. DOI: https://doi.org/10.3233/nre-2007-22502, PMID:18162698.

2. Faul M, Coronado V. Epidemiology of traumatic brain injury. Handb Clin Neurol 2015 Jan;127:3–13. DOI: https://doi.org/10.1016/B978-0-444-52892-6.00001-5, PMID:25702206.

3. Centers for Disease Control and Prevention. TBI-related deaths. Accessed January 22, 2020. www.cdc.gov/traumaticbraininjury/data/tbi-deaths.html

4. Miller JD, Becker DP, Ward JD, et al. Significance of intracranial hypertension in severe head injury. J Neurosurg 1977 Oct;47:503–16. DOI: https://doi.org/10.3171/jns.1977.47.4.0503

5. Balestreri M, Czosnyka M, Hutchinson P, et al. Impact of intracranial pressure and cerebral perfusion pressure on severe disability and mortality after head injury. Neurocritical Care 2006 Feb;4(1):8–13. DOI: https://doi.org/10.1385/NCC:4:1:008, PMID:16498188.

6. Al Saiegh F, Philipp L, Mouchtouris N, et al. Comparison of outcomes of severe traumatic brain injury in 36,929 patients treated with or without intracranial pressure monitoring in a mature trauma system. World Neurosurg 2020 Apr;136:e535–41. DOI: https://doi.org/10.1016/j.wneu.2020.01.070, PMID:31954892.

7. Farahvar A, Gerber LM, Chiu YL, Carney N, Härtl R, Ghajar J. Increased mortality in patients with severe traumatic brain injury treated without intracranial pressure monitoring. J Neurosurg 2012 Oct;117(4):729–34. DOI: https://doi.org/10.3171/2012.7.JNS111816, PMID:22900846.

8. MacLaughlin BW, Plurad DS, Sheppard W, et al. The impact of intracranial pressure monitoring on mortality after severe traumatic brain injury. Am J Surg 2015 Dec;210(6):1082–7. DOI: https://doi.org/10.1016/j.amjsurg.2015.08.007, PMID:26482513.

9. Rønning P, Helseth E, Skaga N-O, Stavem K, Langmoen IA. The effect of ICP monitoring in severe traumatic brain injury: A propensity score-weighted and adjusted regression approach. J Neurosurg 2019 Dec;131(6):1896–904. DOI: https://doi.org/10.3171/2018.7.JNS18270

10. Yuan Q, Wu X, Cheng H, et al. Is intracranial pressure monitoring of patients with diffuse traumatic brain injury valuable? An observational multicenter study. Neurosurgery 2016 Mar;78(3):361–9. DOI: https://doi.org/10.1227/NEU.0000000000001050, PMID:26891376.

11. Rahmanian A, Haghnegahdar A, Rahmanian A, Ghaffarpasand F. Effects of intracranial pressure monitoring on outcome of patients with severe traumatic brain injury: Results of a historical cohort study. Bull Emerg Trauma 2014 Oct;2(4):151–5, PMID:27162888.

12. Aiolfi A, Benjamin E, Khor D, Inaba K, Lam L, Demetriades D. Brain trauma foundation guidelines for intracranial pressure monitoring: Compliance and effect on outcome. World J Surg 2017 Jun;41(6):1543–9. DOI: https://doi.org/10.1007/s00268-017-3898-6, PMID:28188356.

13. Su SH, Wang F, Hai J, et al. The effects of intracranial pressure monitoring in patients with traumatic brain injury. PLoS One 2014 Feb;9(2):e87432. DOI: https://doi.org/10.1371/journal.pone.0087432, PMID:24586276.

14. Yuan Q, Wu X, Sun Y, et al. Impact of intracranial pressure monitoring on mortality in patients with traumatic brain injury: A systematic review and meta-analysis. J Neurosurg 2015 Mar;122(3):574–87. DOI: https://doi.org/10.3171/2014.10.JNS1460

15. Ahl R, Sarani B, Sjolin G, Mohseni S. The association of intracranial pressure monitoring and mortality: A propensity score-matched cohort of isolated severe blunt traumatic brain injury. J Emerg Trauma Shock 2019 Jan–Mar;12(1):18–22. DOI: https://doi.org/10.4103/JETS.JETS_59_18

16. Piccinini A, Lewis M, Benjamin E, Aiolfi A, Inaba K. Intracranial pressure monitoring in severe traumatic brain injuries: A closer look at level 1 trauma centers in the United States. Injury 2017 Sep;48(9):1944–50. DOI: https://doi.org/10.1016/j.injury.2017.04.033

17. Chesnut RM, Temkin N, Carney N, et al. A trial of intracranial-pressure monitoring in traumatic brain injury. N Engl J Med 2012 Dec;367:2471–81. DOI: https://doi.org/10.1056/NEJMoa1207363

18. Carney N, Totten AM, O’Reilly C, et al. Guidelines for the management of severe traumatic brain injury, fourth edition. Neurosurgery 2017 Jan;80(1), 6–15. DOI: https://doi.org/10.1227/NEU.0000000000001432

19. Mouchtouris N, Turpin J, Chalouhi N, et al. Statewide trends in intracranial pressure monitor use in 36,915 patients with severe traumatic brain injury in a mature trauma system over the past 18 years. World Neurosurg 2019 Jun;130:e166–71. DOI: https://doi.org/10.1016/j.wneu.2019.06.025

20. Barmparas G, Singer M, Ley E, et al. Decreased intracranial pressure monitor use at level II trauma centers is associated with increased mortality. Am Surg 2012 Oct;78(10):1166–71.

21. Lundberg N. Continuous recording and control of ventricular fluid pressure in neurosurgical practice. Acta Psychiatr Scand Suppl 1960 36:1–193.

22. Alberico AM, Ward JD, Choi SC, Marmarou A, Young HF. Outcome after severe head injury: Relationship to mass lesions, diffuse injury, and ICP course in pediatric and adult patients. J Neurosurg 1987 Nov;67:648–56. DOI: https://doi.org/10.3171/jns.1987.67.5.0648

23. Falcão AL, Dantas Filho VP, Sardinha LA, et al. Highlighting intracranial pressure monitoring in patients with severe acute brain trauma. Arq Neuropsiquiatr 1995 Sep;53(3-A):390–4. DOI: https://doi.org/10.1590/s0004-282x1995000300004

24. Miller JD, Becker DP, Ward JD, Sullivan HG, Adams WE, Rosner MJ. Significance of intracranial hypertension in severe head injury. J Neurosurg 1977 Oct;47:503–16. DOI: https://doi.org/10.3171/jns.1977.47.4.0503

25. Vora TK, Karunakaran S, Kumar A, et al. Intracranial pressure monitoring in diffuse brain injury: Why the developing world needs it more? Acta Neurochir (Wien) 2018 Jun;160(6):1291–9. DOI: https://doi.org/10.1007/s00701-018-3538-4

26. Faul M, Wald MM, Rutland-Brown W, Sullivent EE, Sattin RW. Using a cost–benefit analysis to estimate outcomes of a clinical treatment guideline: Testing the Brain Trauma Foundation guidelines for the treatment of severe traumatic brain injury. J Trauma 2007 Dec;63:1271–8. DOI: https://doi.org/10.1097/TA.0b013e3181493080

27. Chalouhi N, Mouchtouris N, Saiegh FA, et al. Comparison of outcomes in level I vs level II trauma centers in patients undergoing craniotomy or craniectomy for severe traumatic brain injury. Neurosurgery 2020 Jan;86(1):107–11. DOI: https://doi.org/10.1093/neuros/nyy634

28. DuBose JJ, Browder T, Inaba K, Teixeira PGR, Chan LS, Demetriades D. Effect of trauma center designation on outcome in patients with severe traumatic brain injury. Arch Surg 2008 Dec;143(12):1213–7. DOI: https://doi.org/10.1001/archsurg.143.12.1213

29. Demetriades D, Martin M, Salim A, et al. Relationship between American College of Surgeons trauma center designation and mortality in patients with severe trauma (injury severity score > 15). J Am Coll Surg 2006 Feb;202(2):212–5. DOI: https://doi.org/10.1016/j.jamcollsurg.2005.09.027

30. Rogers FB, Osler T, Lee JC, et al. A mature trauma system: There is no difference in outcome (survival) between level I and level II trauma centers. J Trauma 2011 Jun;70(6):1354–7. DOI: https://doi.org/10.1097/TA.0b013e3182183789

31. Alkhoury F, Courtney J. Outcomes after severe head injury: A national trauma data bank-based comparison of level I and level II trauma centers. Am Surg 2011 Mar;77(3):277–80.

32. Picetti E, Caspani ML, Iaccarino C, et al. Intracranial pressure monitoring after primary decompressive craniectomy in traumatic brain injury: A clinical study. Acta Neurochir (Wien) 2017 Apr;159(4):615–22. DOI: https://doi.org/10.1007/s00701-017-3118-z

33. Volovici V, Huijben JA, Ercole A, et al. Ventricular drainage catheters versus intracranial parenchymal catheters for intracranial pressure monitoring-based management of traumatic brain injury: A systematic review and meta-analysis. J Neurotrauma 2019 Apr;36(7):988–95. DOI: https://doi.org/10.1089/neu.2018.6086

34. Liu H, Wang W, Cheng F, et al. External ventricular drains versus intraparenchymal intracranial pressure monitors in traumatic brain injury: A prospective observational study. World Neurosurg 2015 May;83(5):794–800. DOI: https://doi.org/10.1016/j.wneu.2014.12.040

35. Dawes AJ, Sacks GD, Cryer HG, et al. Intracranial pressure monitoring and inpatient mortality in severe traumatic brain injury: A propensity score-matched analysis. J Trauma Acute Care Surg 2015 Mar;78(3):492–502; discussion 501–2. DOI: https://doi.org/10.1097/TA.0000000000000559

36. Lele A, Kannan N, Vavilala MS, et al. Patients who benefit from intracranial pressure monitoring without cerebrospinal fluid drainage after severe traumatic brain injury. Neurosurgery 2019 Aug;85(2):231–9. DOI: https://doi.org/10.1093/neuros/nyy247

37. Farahvar A, Gerber LM, Chiu Y, et al. Response to intracranial hypertension treatment as a predictor of death in patients with severe traumatic brain injury. J Neurosurg 2011 May;114(5):1471–8. DOI: https://doi.org/10.3171/2010.11.JNS101116

38. Stocchetti N, Zanaboni C, Colombo A, et al. Refractory intracranial hypertension and “second-tier” therapies in traumatic brain injury. Intensive Care Med 2008 Mar;34(3):461–7. DOI: https://doi.org/10.1007/s00134-007-0948-9

39. Güiza F, Depreitere B, Piper I, et al. Visualizing the pressure and time burden of intracranial hypertension in adult and paediatric traumatic brain injury. Intensive Care Med 2015 Jun;41(6):1067–76. DOI: https://doi.org/10.1007/s00134-015-3806-1

40. Kahraman S, Dutton RP, Hu P, et al. Automated measurement of “pressure times time dose” of intracranial hypertension best predicts outcome after severe traumatic brain injury. J Trauma 2010 Jul;69(1):110–8. DOI: https://doi.org/10.1097/TA.0b013e3181c99853

41. Sorrentino E, Diedler J, Kasprowicz M, et al. Critical thresholds for cerebrovascular reactivity after traumatic brain injury. Neurocritical Care 2012 Apr;16 (2):258–66. DOI: https://doi.org/10.1007/s12028-011-9630-8

42. Chesnut RM. What is wrong with the tenets underpinning current management of severe traumatic brain injury? Ann N Y Acad Sci 2015 May;1345:74–82. DOI: https://doi.org/10.1111/nyas.12482

Keywords: intracranial pressure monitoring, Kaiser Permanente, level II trauma center, mortality, severe traumatic brain injury

ETOC

Click here to join the eTOC list or text ETOC to 22828. You will receive an email notice with the Table of Contents of The Permanente Journal.

CIRCULATION

2 million page views of TPJ articles in PubMed from a broad international readership.

Indexing

Indexed in MEDLINE, PubMed Central, EMBASE, EBSCO Academic Search Complete, CrossRef, and SciVerse/Scopus.


                                             

 

 

ISSN 1552-5775 Copyright © 2021 thepermanentejournal.org

All Rights Reserved