Strength of Recommendations: Weak.
Quality of Evidence: Low, primarily from Class III studies and indirect evidence.

Adult

1. Patients with suspected severe traumatic brain injury (TBI) should be monitored in the prehospital setting for hypoxemia (<90% arterial hemoglobin oxygen saturation) or hypotension (<90 mmHg systolic blood pressure [SBP]).
2. Percentage of blood oxygen saturation should be measured continuously in the field with a pulse oximeter.
3. Systolic (SBP) and diastolic blood pressure (DBP) should be measured using the most accurate method available under the circumstances.
4. Oxygenation and blood pressure should be measured as often as possible, and should be monitored continuously if possible.

Pediatrics

1. Pediatric patients with suspected severe TBI should be monitored in the prehospital setting for hypotension. Pediatric hypotension is defined as follows:

   Age	SBP
   0 to 28 days	<60 mmHg
   1-12 months	<70 mmHg
   1-10 years	<70 + 2 X age in years
   >10 years	<90 mmHg

2. Percentage of blood oxygen saturation should be measured continuously in the field with a pulse oximeter using an appropriate pediatric sensor.
3. SBP and DBP should be measured using an appropriately-sized pediatric cuff. When a blood pressure is difficult to obtain because of the child's age or body habitus, documentation of mental status, quality of peripheral pulses, and capillary refill time can be used as surrogate measures.
4. Oxygenation and blood pressure should be measured as often as possible, and should be monitored continuously if possible.

In severe TBI, secondary insults occur frequently and exert a profound negative influence on outcome. This influence appears to differ markedly from that of hypoxemic or hypotensive episodes of similar magnitude occurring in trauma patients who do not have neurologic involvement. Therefore, it is important to determine if evidence exists to support prehospital threshold values for oxygenation and blood pressure.

For this update, Medline was searched from 1996 through July 2006 using the search strategy for this question (see Appendix B), and results were supplemented with literature recommended by peers or identified from reference lists. For adult studies, of 28 potentially relevant publications, 4 were added to the existing table and used as evidence for this question. For pediatric studies, of 53 potentially relevant publications, one new study was included as evidence (see Evidence Tables).

Adult

1. Patients with suspected severe traumatic brain injury (TBI) should be monitored in the prehospital setting for hypoxemia (<90% arterial hemoglobin oxygen saturation) or hypotension (<90 mmHg systolic blood pressure [SBP]).
2. Percentage of blood oxygen saturation should be measured continuously in the field with a pulse oximeter.
3. Systolic (SBP) and diastolic blood pressure (DBP) should be measured using the most accurate method available under the circumstances.
4. Oxygenation and blood pressure should be measured as often as possible, and should be monitored continuously if possible.

Foundation. The deleterious influence of hypotension and hypoxemia on the outcome of patients with severe TBI was analyzed from a large (717 patients), prospectively collected dataset from the Traumatic Coma Data Bank (TCDB).2,13 The TCDB study demonstrated that prehospital hypotension (defined as a single observation of <90 mmHg SBP) and hypoxemia (defined as apnea, cyanosis, or a hemoglobin oxygen saturation <90% in the field or as a PaO2 <60 mmHg by arterial blood gas analysis) were among the five most powerful predictors of outcome. These clinical findings were statistically independent of other major predictors, such as age, admission Glasgow Coma Scale (GCS) score, admission GCS motor score, intracranial diagnosis, and pupillary status. A single episode of hypotension was associated with a doubling of mortality and an increased morbidity when compared with a matched group of patients without hypotension. The TCDB study defined hypotension and hypoxemia as a single reported incident that met the definition of each and did not require a protracted duration for secondary insult.

A smaller Class III study from Australia corroborated the above findings; particularly with respect to the effects of hypotension on outcome.6 The clinical predictors of mortality derived fromthis study were identical. In both studies, the two predictors with the potential for being altered through clinical manipulations are hypotension and hypoxemia. These data are similar to those in other retrospectively analyzed inhospital reports in adults.8-11,14-16,18,20,25

The incidence of hypotension in patients with TBI upon first contact in the field was reported in a recent study from Australia to be 16% (48/296).7 Similarly, in a report from San Diego, initial hypotension (systolic BP <90mmHG) was encountered in 19% of patients.17

A study from Italy21 of 50 patients with TBI transported by helicopter revealed that 55% had an oxygen saturation < 90% measured at the scene prior to intubation. Both hypoxemia and hypotension had significant negative impacts on outcome. Of the 28 patients who were hypoxemic, 13 did not have associated hypotension. There was a significant association between arterial desaturation and poor outcome (p < 0.005):

The importance of frequent or continuous monitoring of oxygen saturation was recently documented in a study from San Diego of patients with suspected TBI who were undergoing rapid sequence intubation (RSI) in the pre-hospital setting.3 Each of the 59 study patients was matched to three historical non-intubated control patients. During the second half of the study, patients were monitored with a device which continuously recorded oxygen saturation (as well as endtidal CO2 [EtCO₂]) at 8 second intervals, thus allowing investigators to document the occurrence, timing, and duration of hypoxic episodes. Compared to historical controls, after controlling for other confounding factors, the authors reported that profound oxygen desaturations (SpO2 <70%) during intubation and any oxygen desaturation (SpO2 <90%) were associated with higher mortality (O.R. 3.89, 95% C.I. 1.12-13.52 and 3.86, 95% C.I. 1.18-12.61 respectively). Subgroup analysis from this study revealed that 31/54 (57%) patients had transient hypoxemia during the RSI attempt.4 In addition, the desaturation period lasted approximately 2.33 minutes.

Related Discussion. In-hospital studies provide important data thatmaybe extrapolated to and representative of issues found in the prehospital setting. In a study of 107 patients with moderate or severe TBI (GCS <13), the authors attempted to evaluate the effect of hypotension and hypoxia on the functional neurologic outcome of these patients by specifically quantifying the degree and duration of the secondary insult.1 Any occurrence of hypotension was associated with an increased risk of 30 day in-hospital mortality (O.R. 3.39) and based upon the quantification analysis, moderate and high-dose of hypotension were associated with progressively greater death rates (O.R. 3.14 and 12.55 respectively) and a strong trend towards poorer functional neurologic outcomes. In this study, there was no effect on patient outcome from hypoxia.

Another study involving a convenience sample of 50 patients with TBI who were undergoing transfer from an initial receiving hospital to a regional neurosurgical referral center revealed that upon arrival at the referral center, 6% of patients were hypoxic (O2 saturation <95%) and 16% were hypotensive (systolic BP < 90 mmHg for adults and adjusted for age in children).5 The authors did not evaluate for effect on patient outcome. This study does demonstrate, however, that even after presumed resuscitation and stabilization, secondary insults are not uncommon and must be always be considered.

The value of 90-mmHg systolic pressure to delineate the threshold for hypotension arose arbitrarily, and is more a statistical than a physiologic parameter. In considering the evidence concerning the influence of cerebral perfusion pressure (CPP) on outcome, it is possible that systolic pressures significantly >90 mmHg would be desirable during the prehospital and resuscitation phase, but no studies have been performed to corroborate this. The importance of meanarterial pressure (MAP), as opposed to systolic pressure, should also be stressed, not only because of its role in calculating [CPP] = [MAP - intracranial pressure (ICP)], but because the lack of a consistent relationship between the systolic and mean pressures makes calculations based on systolic values unreliable. It may be valuable to maintain mean arterial pressures considerably above those represented by systolic pressures of 90 mmHg throughout the patient's course.

No Class I study has directly addressed the efficacy of preventing or correcting early hypotension in the prehospital setting to improve outcome. However, a subgroup of severe TBI patients was subjected to post hoc analysis in a randomized, placebo controlled, multicenter trial that compared the efficacy of administering 250 mL of hypertonic saline versus normal saline as the initial resuscitation fluid in hypotensive trauma patients.23 In that trial, the hypertonic saline group had improved blood pressure responses, decreased overall fluid requirements, and associated improvements in survival. The investigators retrospectively reviewed the records of the subgroup of patients with severe TBI and found that this group had statistically significant improvement in survival to discharge. Although this was a post hoc analysis of Class II data, it suggests that elevating the blood pressure in severe TBI patients with hypotension improves outcome.

Pediatrics

1. Pediatric patients with suspected severe TBI should be monitored in the prehospital setting for hypotension. Pediatric hypotension is defined as follows:

   Age	SBP
   0 to 28 days	<60 mmHg
   1-12 months	<70 mmHg
   1-10 years	<70 + 2 X age in years
   >10 years	<90 mmHg

2. Percentage of blood oxygen saturation should be measured continuously in the field with a pulse oximeter using an appropriate pediatric sensor.
3. SBP and DBP should be measured using an appropriately-sized pediatric cuff. When a blood pressure is difficult to obtain because of the child's age or body habitus, documentation of mental status, quality of peripheral pulses, and capillary refill time can be used as surrogate measures.
4. Oxygenation and blood pressure should be measured as often as possible, and should be monitored continuously if possible.

Foundation. The deleterious influence of hypotension and hypoxemia on the outcome of children with severe TBI is similar to that seen in adults. There are a very limited number of Class III pediatric studies and most were hospital-based data. There is no Class I or II evidence that addresses the value of either prehospital assessment or intervention on the outcome of severe TBI in children.

Pigula and colleagues prospectively evaluated the effect of hypotension (SBP <90 mmHg) and hypoxia (PaO2 < 60 mmHg) on outcome in 170 patients with a GCS < 8.19 In the pediatric group age < 17 years they noted that the overall mortality rate in children was better than adults (29% vs. 48%). Children with both hypoxia and hypotension had a higher mortality rate (67%) compared to normotensive children without hypoxia (16%).

Another retrospective study examined the effect of hypotension on outcome of 72 children with a GCS of 6-8.12 Hypotension was defined hypotension as SBP<5th percentile for age lasting for longer than 5 minutes. Eighty-nine percent of the patients were intubated in the field and the remaining 11% were intubated in the emergency department. There were 62 hypotensive episodes and 9 patients were hypoxic. The majority of these episodes occurred in the emergency department or intensive care unit. Patients with poor outcome had more hypotensive episodes compared to those with good outcome.

Related Discussion. One retrospective study of ICU patients noted survival improved 19-fold in children age 0-17 years with GCS <8 when the maximum SBP was >135 mmHg.24 Vavilala and colleagues examined the association between age adjusted SBP percentile and outcome after severe TBI. Using the age-adjusted SBP values published by the BTF in 2000 the author reported that a SBP less than the 75th percentile for age was associated with poor outcome.22

Clinical trials are needed in the following areas:

1. Do prospective data correlate magnitude and duration of hypotensive and hypoxic episodes to outcome?
2. Is mean arterial pressure a more accurate indicator of hypotension than systolic blood pressure?
3. How accurate are devices that measure systolic, diastolic, and mean blood pressures during transportation?
4. A similar assessment to that outlined in (2) above is needed for arterial oxygen saturation.
5. Prospective studies on the above four points are needed for the pediatric population.

The two major areas needing investigation are (1) the critical values for duration and magnitude of hypotensive and hypoxemic episodes and how they affect neurological outcome, and (2) the optimal resuscitation protocol (fluid type, route of administration, etc.) for resuscitating the patient with severe TBI. The former question is not a subject for a controlled trial for ethical reasons, and therefore is best undertaken using a prospective study with the precise collection of prehospital blood pressure and oxygenation data, which is then correlated with outcome. The latter question can be studied in prospective, randomized investigations, several of which are presently underway.

The pediatric TBI population has needs similar to that of adults. Specifically research in the following areas is needed:

1. The correlation between prehospital SBP and pulse oximeter measurement and outcome.
2. The impact of prehospital intervention on outcome (fluid resuscitation and maintaining airway/ventilation).
3. The reliability of other hemodynamic stability parameters, such as peripheral perfusion, capillary refill time, mental status, and the quality of peripheral pulses as prehospital surrogates to obtaining blood pressure in young children.