Geriatric Trauma, Triage of

Authors

EAST Practice Management Guidelines Work Group

David G. Jacobs, MD, Committee Co-Chair, Carolinas Medical Center, Charlotte, NC
Brian Ray Plaisier, MD, Committee Co-Chair, Bronson Hospital, Kalamazoo, MI
Philip S. Barie, MD, NYH—Cornell Medical, Center New York, NY
Jeffrey S. Hammond, MD, Robert Wood Johnson Medical School, New Brunswick, NJ
Michele R. Holevar, MD, Mt. Sinai Hospital Chicago, IL
Karlene E. Sinclair, MD, Morehouse School of Medicine, Atlanta, GA
Thomas M. Scalea, MD, University of Maryland Medical Center, Baltimore, MD
Wendy Wahl, MD, University of Michigan Health System, Ann Arbor, MI

Introduction

Advanced age is a well-recognized risk factor for adverse outcomes following trauma.  A substantial body of literature, much of it cited within this document, demonstrates increased morbidity and mortality in geriatric trauma patients compared to their younger counterparts. Whether this outcome difference is due to the decreased physiologic reserve that accompanies aging, a higher incidence of pre-existing medical conditions in the geriatric patient, or to other factors yet to be identified, remains unclear. It is clear, however, that good outcomes can be achieved in this patient population when appropriately aggressive trauma care is directed towards geriatric patients with survivable injuries. Implicit in the above statement is the need to identify, as soon as possible following injury, those patients who will benefit from aggressive resuscitation, timely injury management, and post-trauma rehabilitation. It is equally important, however, to limit these intensive and expensive treatment modalities to patients whose injuries are not only survivable, but are compatible with an acceptable quality of life.

Our purpose in developing this guideline was to provide the trauma practitioner with some evidence-based recommendations that could be used to guide decision-making in the care of the geriatric trauma patient. We began this process by first developing a series of questions, the answers to which we hoped could be supported by the existing scientific literature. The initial set of questions were as follows:

1. Is age itself a marker of increased morbidity/mortality? If so, what age should be used?
2. Is age instead a surrogate for increased pre-existing conditions (PEC’s)? If so, which pre-morbid conditions are particularly predictive of poor outcomes?
3. Should age itself be a criterion for triage from the field directly to a trauma center, regardless of Glasgow Coma Scale (GCS) score, trauma score (TS), etc.? If so, what age should be used?
4. Do trauma centers have better outcomes with geriatric trauma than non-trauma centers?
5. Are there specific injuries, scores [Injury Severity Score (ISS), TS, GCS, etc], or PEC/age combinations in geriatric trauma patients that are so unlikely to be survivable that a non-aggressive approach from the outset could be justified?
6. What resuscitation end-points should be used for the geriatric trauma patient?
7. Should all geriatric trauma patients receive invasive hemodynamic monitoring? If so, what specific types of monitoring should be used? If not, which geriatric patients benefit from invasive monitoring?
8. Are there specific types of therapies that should be employed routinely in trauma patients (e.g. ß- blockade, nitroglycerin infusions, etc.)?
9. How are outcomes measured in geriatric trauma? Which specific outcome measures should be used?

Unfortunately, after examining the available literature, it is clear that evidence-based responses to all of the questions raised above are not possible. As the accompanying evidentiary tables demonstrate, there are a few, if any, prospective, randomized, controlled trials which definitively address any of the above issues. Secondly, there is a lack of uniformity as to a specific age criterion for geriatric trauma. As shown in the accompanying evidentiary table, geriatric trauma is variously defined in the literature as age greater than or equal to 55, 60, 65, 70, 75, and even 80 years of age. There is even literature support for increased mortality from trauma beginning at age 45! Furthermore, since age is a continuous variable, and not a dichotomous one, adverse outcomes associated with geriatric trauma are likely to increase in a continuous fashion which age as opposed to a stepwise leap as a given patient reaches a specific age. Third, there is no consise definintion of a geriatric trauma patient. In some studies, all patients over a given age are included, whereas in others, patients with penetrating injuries, burns, and those with minor injuries, such as slip-and-falls, are excluded.  Some studies include all patients regardless of hemodynamic instability or injury severity, while others impose strict entrance criteria or exclude patients who do not survive for a predetermined period of time following admission. Such lack of uniformity with regards to inclusion criteria makes it difficult to compare outcomes across different patient populations. Finally, much of the literature concerning geriatric trauma is relatively “old”, that is, published more than 10 years ago.  Given the significant improvements in patient care which have occurred over the past 10 to 20 years, recommendations based upon outcomes achieved more than 10 years ago may not be applicable to today’s geriatric trauma patient.

Despite the above-mentioned shortcomings, our committee still felt that it was important to summarize the available literature and make evidence-based recommendations where satisfactory evidence did exist.  In light of the nine questions raised above, three broad areas of focus emerged within this guideline: Issues of Geriatric Trauma Triage, Issues of Geriatric Trauma Resuscitation, and Issues of Outcome Measurement in Geriatric Trauma.  Although there was considerable overlap among these three areas, each issue has been addressed separately within this guideline and, accordingly, three separate “sub-guidelines,” each with its own recommendations, evidentiary table, and areas for future research, comprise this practice management guideline for geriatric trauma. It is hoped that the information provided within these three sub-guidelines, will provide evidence-based support for the difficult decisions which are required to achieve optimal outcomes in this difficult, but ever increasing, patient group.

I. Statement of the problem

The process of triage, as it relates to the geriatric trauma patient, is an attempt to provide the patient with the appropriate intensity of medical resources, taking into account the severity of illness, the cost and availability of medical resources, the prognosis for functional survival and, if known, the expressed desires of the patient. For the geriatric trauma patient this process begins in the pre-hospital phase of care where decisions must be made regarding the appropriate patient destination, trauma center versus non-trauma center.  In the resuscitative phase of trauma care, triage decisions regarding patient destination must again be made, specifically whether patient circumstances dictate provision of intensive care resources or whether standard trauma inpatient care will suffice. Throughout the hospital phase of care, the patients must be “triaged” towards or away from operative procedures, invasive and expensive critical care therapies, and powerful, yet potentially dangerous pharmacologic treatment options, decisions which, again, must be based upon the likelihood of achieving a good, long-term outcome for the patient.  An increasingly common circumstance, particularly in the geriatric trauma patient, involves the decision to withdraw, or perhaps not even institute, an aggressive course of treatment, when the clinical circumstances are incompatible with a quality of life which all parties concerned would deem acceptable. Fundamental to all of these “triage” decisions, is the ability to predict with reasonable accuracy, what a particular patient’s outcome might be depending on which “triage” decision is made. In order to be of any value to the trauma practitioner, and ultimately to the patient and his or her family, the clinical variables upon which these predictions are to be made must be easy to obtain, reliable, and available to the trauma practitioner within a relatively short period of time following injury. The task, therefore, of this particular subcommittee was to determine whether there existed adequate support in the scientific literature to develop recommendations regarding 1) appropriate criteria for triage of the geriatric trauma patient to trauma centers, 2) the clinical variables which would be useful in predicting the need for intensive care resources for the geriatric trauma patient, and 3) those clinical circumstances where a non-aggressive approach from the outset could be justified.

II. Process

An initial computerized search was undertaken using Medline with citations published between the years of 1966 and 1999. Using the search words “geriatric”,  “trauma”, “elderly”, and “injury”, and by limiting the search to citations dealing with human subjects and published in the English language, well over 2,300 citations were identified. From this number were then excluded letters to the editor, case reports, reviews, and a large number of articles dealing with minor injury mechanisms, particularly hip fractures from slip-and-falls. An additional cause for exclusion of references was publication prior to 1975 as it was felt that the trauma care provided at this time was so different compared to current trauma care that recommendations based upon data from this earlier time period would not be valid. The abstracts of the remaining citations were each reviewed, and those articles that did not address prognostic variables or other issues pertinent to the triage of the geriatric trauma patient were further excluded. This yielded a total of 32 articles that comprised the initial evidentiary table. The bibliographies of these 32 articles were then further reviewed and additional 13 articles meeting the above-mentioned criteria were added for a total of 45 references within the evidentiary table. Each reference was then reviewed by three trauma surgeons, and consensus reached regarding appropriate classification of each reference according to the Canadian and United States Preventive Task Force.
 
Criteria for achieving a specific classification and the number of articles for each class are shown below:

Class I: Prospective randomized controlled trials - the gold standard of clinical trials.  Some may be poorly designed, have inadequate numbers, or suffer from other methodological inadequacies. (0 references)

Class II: Clinical Studies in which data was collected prospectively, and retrospective analyses that were based on clearly reliable data. Types of studies so classified include observational studies, cohort studies, prevalence studies and case control studies. (2 references)

Class III: Studies based on retrospectively collected data. Evidence used in this class indicate clinical series, database or registry review, large series of case reviews, and expert opinion (44 references) (The total number of classified references exceeds the total number of references by one because one two-part study was classified as both a Class II and Class III reference.)

III. Recommendations

A. Level I

There is insufficient Class I and Class II data to support any standards regarding triage of geriatric trauma patients.

B. Level II

1. Advanced patient age should lower the threshold for field triage directly to a trauma center.

C. Level III

1. All other factors being equal, advanced patient age, in and of itself, is not predictive of poor outcomes following trauma, and therefore should NOT be used as the sole criterion for denying or limiting care in this patient population.
2. The presence of pre-existing medical conditions (PEC’s) in elderly trauma patients adversely affects outcome. However this effect becomes progressively less pronounced with advancing age.
3. In patients 65 years of age and older, a GCS < 8 is associated with a dismal prognosis. If  substantial improvement in GCS is not realized within 72 hours of injury, consideration should be given to limiting further aggressive therapeutic interventions.
4. Post-injury complications in the elderly trauma patient negatively impact survival and contribute to longer lengths of stay in survivors and nonsurvivors compared to younger trauma patients. Specific therapies designed to prevent and/or reduce the occurrence of complications (particularly iatrogenic complications) should lead to optimal outcomes in this patient population.
5. With the exception of patients who are moribund on arrival, an initial aggressive approach should be pursued with the elderly trauma patient, as the majority will return home, and up to 85% will return to independent function.
6. In patients 55 years of age and older, an admission base deficit < -6 is associated with a 66% mortality. Patients in this category may benefit from in-patient triage to a high-acuity nursing unit.
7. In patients 65 years of age and older, a Trauma Score < 7 is associated with a 100 % mortality. Consideration should be given to limiting aggressive therapeutic interventions.
8. In patients 65 years of age and older, an admission respiratory rate < 10 is associated with a 100 % mortality. Consideration should be given to limiting aggressive therapeutic interventions.
9. Compared to younger trauma patients, patients 55 years of age and older are at considerably increased risk for undertriage to trauma centers even when these older patients satisfy appropriate triage criteria. The factors responsible for this phenomenon must be identified and strategies developed to counteract it.

IV. Scientific foundation

Triage is the process whereby the patient’s medical needs are matched with the available medical resources. For the geriatric trauma patient, the process begins in the pre-hospital arena, where pre-hospital providers must decide on the basis of relatively scant clinical information whether a patient should bypass the local hospital in favor of a trauma center. The American College of Surgeons Committee on Trauma (ACS-COT), among other medical organizations, in its manual, “Optimal Resources for the Care of the Trauma Patient” has published a set of triage criteria to aid pre-hospital providers in identifying appropriate patients for direct transport to trauma centers^[1]. Within this document, it is suggested that patients greater than age 55 should be “considered” for direct transport to a trauma center, apparently without regard to the severity of injury. This recommendation is based upon a substantial medical literature that demonstrates significantly worse outcomes for geriatric trauma patients compared to their non-geriatric counterparts. One of the earliest studies to look at the influence of age on outcome from major trauma was the Major Trauma Outcome Study, sponsored by the ACS-COT.  Data from 3,833 patients 65 years and older was compared to that of 42,944 patients less than 65 years of age.  Mortality rose sharply between age 45 and 55 and doubled at age 75 years. This age-dependent survival decrement occurred at all ISS values, for all mechanisms of injury, and for all body region.^[2] Numerous other studies have supported the findings that the effect of trauma on the elderly is more serious than that on younger patients.^[3-9]

Given these findings, some authors have suggested triaging elderly trauma victims to trauma centers at a much lower threshold than similarly injured younger patients, in order to minimize mortality and morbidity.^[10] Support for this recommendation can be found in a study by Smith et al, documenting fewer complications for elderly femur fracture patients treated at trauma centers Vs non-trauma centers.^[11]   In spite of these poorer outcomes, trauma patients 55 years of age and older are frequently triaged to non-trauma hospitals even when they satisfy well-defined anatomic or physiologic criteria. Compliance with physiologic criteria appears to be especially troublesome.^[12] In two unrelated studies, undertriage in patients over the age of 55 was twice that of younger patients,^{[12] [13]} while a similar study demonstrated even worse results for patients over the age of 65.^[14]

The factors responsible for the increased morbidity and mortality seen in geriatric trauma are not entirely clear. It has been suggested that it is not patient age per se, but the high incidence of pre-existing medical conditions in the geriatric patient that accounts for the difference.  Others have suggested that the elderly, simply by virtue of being more frail, sustain a greater degree of injury in response to a given impact, compared to their younger counterparts. The existing medical literature was therefore reviewed in an attempt to identify clinical factors that might be used to triage geriatric trauma patients to either aggressive Vs non-aggressive treatment strategies.

Age and outcome

It is difficult to find consensus in the existing literature regarding the relationship between the patient age and outcome. Many of the reasons for this failure have been mentioned above, and include differences in the age definition of geriatric trauma, and differences in inclusion criteria for the various studies.  In addition to these two factors, there is a lack of uniformity regarding the length of follow-up required to define a poor outcome.  This has been variably defined as death within 24 or 48 hours of injury, death prior to ICU or hospital discharge, and even death/vegetative outcome at three or four years post injury. Furthermore, there are wide variances in the statistical methods used to explore the relationship between age and outcome. Many authors have documented a statistically significant difference between the mean age of geriatric survivors compared with the mean age of geriatric non-survivors, and thus have concluded that age is significantly associated with poor outcome. Other authors have applied logistic regression analysis to their data set to determine which particular factors are predictive of adverse outcomes. Given the wide variation in inclusion criteria, outcome variables and statistical methods present within the existing literature, the conflicting results regarding age and outcome are not surprising. 

Certainly, the largest data set examined to date is that published by Morris et al in 1990. These authors examined 199,737 trauma admissions, aged 15 and older, to acute care hospitals in the state of California during 1986. Using logistic regression techniques, the Injury Severity Score (ISS) was found to be the best predictor of mortality in trauma patients, but age, gender, and pre-existing medical conditions (PEC’S) were also found to be independent predictive factors of mortality.  Mortality was defined as in- hospital death. The authors also found that while the mortality from minor injury (ISS <9) begins to increase beyond the age of 65, the mortality for moderate injuries (ISS 9-24) begins to increase at 45 years of age.^[4] This increase in trauma mortality beginning at age 45 had been confirmed by other investigators as well.^{[2] [10] [15]}. Several authors have examined the relationship between in-hospital mortality and age, with differing conclusions. Pelicane et al demonstrated a statistically significant difference in age between elderly non-survivors and elderly survivors in a series of 374 geriatric trauma patients, defined as age greater than or equal to 65 years. Five of the deaths in this series occurred in the Emergency Department, a subset of patients that has been excluded from the analyses in other series. Burn patients, however, were excluded from this series.^[16] In a similar study, performed by Osler, of 100 geriatric trauma patients 65 years or older, no significant differences in age was found between elderly survivors and non-survivors.^[9] Despite the fact that patients who expired prior to transfer to the operating room or to the ICU were excluded in Osler’s series, mortality in this series was more than twice that in Pelicane’s series. Perhaps this is explained by the lower mean Trauma Score (TS) in Pelicane’s series relative to that of Osler (13 versus 15.4). Pelicane’s series contains nearly four times as many patients which raises, the possibility of a type 2 statistical error with regards to Osler’s inability to demonstrate a statistical difference between the ages of geriatric trauma survivors and non-survivors.  A large and more recent study of 448 patients, 65 years and older, employed a logistic regression analysis and demonstrated age to be significantly predictive of both early (<24 hours) and late (>24 hours) mortality.^[17] In this analysis, survival was used as the outcome variable, with “geriatric status” (age greater than or equal to 65) entered into the logistic regression equation. In so doing, “geriatric status” was associated with a 2.46-fold increased likelihood of early mortality and a 4.64-fold increased risk of late mortality. However, an even larger study yet, consisting of 852 patients, reported on by Knudson et al, using stepwise discriminant analysis, did not find age to be predictive of in-hospital death. The authors reported a 1.33-fold increased risk of death associated with age status greater than 75 years, just barely missing statistical significance with a p-value of 0.06.  Interestingly enough, however, the age of 75 years was entered into the discriminant analysis, not the age of 65 or greater which was the authors’ original age definition for entrance into the study. Perhaps statistical significance would have been demonstrated had age 65 or greater been used in the discriminant analysis.^[18]

Two studies specifically examined the relationship between age and in-hospital mortality for geriatric trauma patients admitted to the Intensive Care Unit. Neither found any association between age and outcome. In a small series of 39 patients requiring intensive care unit admission and placement of pulmonary and radial artery catheters, Horst et al reported no significant difference in age between elderly survivors and non-survivals.  As would be expected, overall mortality (31%) was high in this intensive care population of patients greater than 60 years of age. Logistic regression analysis was not performed in this study, probably due to the overall low number of patients.^[19]  A more recent study by Shabot and colleagues examined two subsets of geriatric trauma patients, those between the ages of 65 and 74, and those 75 years and older. Outcomes in these 99 geriatric trauma patients were then compared to 940 “younger” patients between the ages of 13 and 64, all of who were admitted to a surgical intensive care unit (SICU). SICU mortality was then examined by comparing survivors with non-survivors, regardless of age. As would be expected, there was no significant difference in age between non-survivors and survivors (39.0 years versus 34.8 years), likely due to the tenfold larger number of patients seen in the “younger” patient group.^[20]

Finally, several studies have examined the relationship of age to more long-term outcomes, although no clear consensus is evident. DeMaria et al studied a group of 82 trauma patients over the age of 65 years. Patients with penetrating injury and isolated orthopedic injury were excluded, as were patients sustaining thermal injury. Survival was defined at six months post-injury. Not only were non-survivors older, but they also demonstrated higher ISS’s and more complications.  Based on these findings, the authors developed the Geriatric Trauma Survival Score (GTSS), and then prospectively tested it on 61 patients, with 92% accuracy. Unfortunately, the GTSS, though perhaps accurate, has little triage value at the time of patient admission, as it requires information not available to the practitioner at that time.^[6] Van Der Sluis compared early and late mortality between elderly trauma patients and elderly hip fracture patients. Early mortality was higher for the trauma patients, but survival seven to eight years following injury was similar between the two groups. Logistic regression analysis was employed to identify predictors of late mortality, and demonstrated age to be a significant predictor. ^[21] van Aalst, in a study of blunt geriatric injury with a mean follow-up of almost 3 years, employed logistic regression analysis to demonstrate an association between poor outcome and age > 75.^[5]Oreskovich, however, failed to demonstrate any relationship between age and outcome at one year following injury in a group of 100 patients age 70 and greater.^[22]Broos, in two separate publications examining six-month outcome in trauma patients aged 65 and greater, did not find age to be predictive of mortality.^{[23] [24]} Inclusion criteria for each of these 3 latter studies were vaguely defined, and Broos’ 18% mortality is inexplicably low compared to other series of similarly injured patients.^{[9] [17] [25] [26]}A larger study, with a more plausible mortality, was published by Battistella et al in 1998. This study involved 279 geriatric trauma patients, which the authors defined as age greater than 75. Mean ISS in this patient group was 9.4, and associated mortality was 23%. Using logistic regression analysis, the authors found that poor outcome, defined in this study as survival less than six months following hospital discharge, was not predicted by patient age.^[27]  The issue of long-term survival and quality of life in the geriatric trauma patient is discussed more fully below.

Can the age of a geriatric patient, then, be used to predict outcome following trauma?  While age appears to have some value in mortality projections for a population of geriatric trauma patients, there is certainly no literature support for a specific age above which geriatric trauma in-hospital mortality can be predicted with any degree of confidence. It has been suggested, however, that early mortality may not be the best outcome measure in geriatric trauma, due to a high percentage of poor long-term functional survival in elderly trauma patients surviving hospital discharge.^[22]The preponderance of available literature, however, suggest more favorable long-term outcomes, with up to 85% of survivors functioning independently at home at follow-up intervals as long as six years post injury.^{[5] [27-31]} Thus, given reasonable long-term functional outcomes for geriatric trauma patients surviving hospitalization, and the inability of patient age, by itself, to predict in-hospital mortality, advanced patient age should not be used as the sole criteria for denying or limiting care in the geriatric trauma population.

Pre-existing conditions (PEC's) and outcome

If chronologic age, then, is not useful in predicting geriatric trauma survival, perhaps it is the patient’s physiologic age, or the nature and extent of pre-existing medical conditions (PEC’s), that determines outcome. Since the frequency of PEC’s does increase with age, it may be difficult to separate these two factors and their relationships to adverse outcomes in geriatric trauma. Unfortunately, due, once again, to a wide variety of age definitions for geriatric trauma, statistical methodologies, and outcome measures, the literature addressing the prognostic value of PEC’s in geriatric trauma outcome is inconclusive. The largest studies, and those with the best statistical methodology, do seem to demonstrate a significant predictive capacity of PEC’s for adverse outcomes in geriatric trauma. Morris et al, in two separate publications in 1990, examined hospital discharge data for trauma patients in California for the year 1986. Using logistic regression analysis in both studies, Morris was able to demonstrate that PEC’s were important predictive factors of mortality, independent of age. The effect of PEC’s on mortality, however, became less important in patients over the age of 65, perhaps because at this age, chronologic age becomes the predominant predictor of mortality, and the added presence of PEC’s does little to increase trauma mortality further.^{[4] [32]} Similarly, Milzman et al, in the study of nearly 8,000 trauma patients, noted a threefold increase in trauma mortality in patients with PEC’s, compared to those without. Once again, the effect of PEC’s on mortality was noted to be independent of age although, like Morris et al, these authors noted a decreasing influence of PEC’s on trauma mortality with advanced age.^[33]  A more recent study published in 1997 by Gubler et al, examined risk factors for mortality among a group of 9,424 trauma patients, aged 67 and greater, who were discharged from acute care hospitals within the state of Washington in 1987. For each trauma patient in the series, four uninjured patients, matched for age and gender, were identified from the same HCFA database. Co-morbid diagnoses (PEC’s) were identified for each patient, and a Co-Morbid Diagnosis Index Score calculated.  This score is a weighted index which takes into account not only the number, but also the severity of PEC’s.^[34] Using Cox proportional hazards regression, Gubler found that patients with PEC’s were anywhere between

2.0 and 8.4 times as likely to die within five years of injury compared to those without PEC’s, depending upon the number and severity of PEC’s.^[8] Several smaller studies, each reporting the experience of a single trauma center, and using logistic regression analysis, confirmed the value of PEC’s as predictive factors of poor outcome in geriatric trauma, although inclusion criteria, and age and outcome definitions were not uniform among these studies.^{[17] [27] [35]} Other studies have refuted these findings, but suffer from some methodological and statistical shortcomings that weaken their conclusions.^{[19] [22-24] [36] [37]}

Severity of injury scoring and outcome

A number of physiologic and anatomic “scores” have been shown to correlate with geriatric outcome. These include Trauma Score (TS), Revised Trauma Score (RTS), Glasgow Coma Scale (GCS), Acute Physiology and Chronic Health Evaluation Score (APACHE), Acute Physiologic Sore (APS), Simplified Acute Physiology Score (SAPS), Injury Severity Score (ISS), Maximal Abbreviated Injury Score (MAIS), and the Geriatric Trauma Survival Score (GTSS). In addition, although not “scores” in the typical sense, geriatric trauma outcome has also been correlated with initial blood pressure, respiratory rate, and base deficit. Although most, if not all, of these scores do correlate with geriatric outcome, from the perspective of field or emergency department triage, many of these scores have little value in that they are not derivable at the moment that these particular triage decisions need to be made. This would apply particularly to APACHE, APS, SAPS, MAIS, ISS and GTSS. These scores, however, perhaps in combination with patient age, may have some value in the prediction of lethal outcomes in geriatric trauma, and, therefore, may be valuable triage tools in the Intensive Care Unit.  These scores will therefore be discussed solely within that context below.

On the other hand, measures of physiologic derangement, whether obtained via physical examination or chemical analysis, may help to identify patients who will perhaps benefit from aggressive resuscitation strategies (and should therefore be triaged to an intensive care unit), as well as those where further resuscitated efforts are futile (thus facilitating earlier termination of resuscitation). TS (or RTS) and its components (blood pressure, respiratory rate and GCS) are the most readily obtainable, objective physiologic data available either to the pre-hospital provider or to the trauma resuscitation team in the Emergency Department. The prognostic value of each of these variables as they relate to geriatric trauma outcome will be discussed further below. (The prognostic value of GCS will be discussed within the section entitled “Outcome from Geriatric Head Injury.”)  Of those chemical analyses available in the Emergency Department, only base deficit has been subjected to sufficient scientific study, and is sufficiently relevant to geriatric trauma resuscitation that it can be included within the discussion below of potentially useful triage “scores.”

The TS assesses five physiologic functions (blood pressure, respiratory rate, respiratory effort, GCS and capillary refill), yielding a minimal score of zero and a maximal score of 16. The RTS is a simplified version of the TS, which deletes the assessment of respiratory effort and capillary refill, resulting in a range of scores between zero and eight. Several studies have demonstrated the predictive value of TS (and RTS) on geriatric trauma mortality, although a specific numerical score signifying a fatal injury has not identified by these authors.^{[17] [20]} Horst was unable to find any relationship between TS and geriatric trauma patient mortality, although the small number of patients and the rather narrow entrance criteria of the study limit the applicability of this finding.^[19]  Two other studies, however, suggest that TS may be a useful triage tool in the early stages following geriatric trauma. In a case-matched review of 100 patients age 65 and above that suffered injuries severe enough to necessitate hospitalization, no elderly patient was able to survive a TS < 9. More dramatically, no geriatric patient with a TS < 7 survived long enough to reach the hospital and be included in the study. The authors felt this to be of importance in allowing more realistic counseling of patients and their families.^[9] These findings were reinforced by a review from three trauma centers of 852 patients, age 65 and older, in which a TS < 7 was associated with a 100% mortality.^[18] These data suggest that aggressive care under these circumstances is likely to be futile, and that consideration should be given to limiting intensive therapy when a geriatric patient presents with a TS < 7. In addition to the prognostic value of TS, Knudson’s data also revealed a 100% mortality in patients 65 years and older who presented with a respiratory rate < 10.  Here, too, consideration should be given to limiting aggressive therapeutic interventions. In addition to its role in the prediction of fatal outcomes in geriatric trauma patients, TS may also have implications for intensive care unit triage. In a study of 374 patients aged 65 and older, mortality was noted to be only 5% with a TS of 15 or 16, but 25% in patients with a TS of 12 to 14, and 65% in patients with a TS <12.^[16]Thus, patients with TS between 7 and 14 may benefit from aggressive resuscitation strategies and triage to a critical care unit.

Measurement of arterial base deficit may provide useful information regarding the extent of shock and the adequacy of resuscitation in trauma patients, and may therefore be useful in early decision-making and resource allocation.  In a series of 274 elderly trauma patients, defined for the purposes of this study as age greater than or equal to 55, arterial blood gases obtained within one hour of a patient admission were correlated with ICU length of stay and mortality. Base deficits were characterized as mild (-3 to –5), moderate (-6 to –9), and severe (? –10).  As expected, elderly patients with severe base deficits had a high mortality, 80% in this series. However, geriatric trauma mortality was still markedly elevated at 60% in patients with only moderate base deficits.  Even a “normal” base deficit carried a mortality of 24%.^[7]  Thus, early determination of admission base deficit in geriatric trauma patients may facilitate early identification of “occult shock” and identify a subgroup of patients who may benefit from more intensive monitoring and resuscitation.

Injury Severity Score (ISS) is probably the most widely studied anatomic or physiologic severity of illness score yet to be correlated with geriatric trauma outcome. Most authors have found it to be a strong predictor of outcome in geriatric trauma,^{[9] [16] [17] [20] [21] [25]} while two large studies claimed that it is the best predictor of mortality in geriatric trauma.^{[4] [18]} Others, however, have failed to demonstrate any such relationship.^{[19] [22-24] [27]} Whether or not such a relationship does indeed exist, ISS is severely limited in its prognostic capability due to significant delays in obtaining sufficient data to calculate the score. It has, therefore, probably very little prognostic value in geriatric trauma, and, even then, only in patients in whom the question of futility has been raised. Despite the abundance of literature examining the relationship between ISS and geriatric trauma patient outcome, only two publications contain any ISS data that might be considered useful prognostically. Van der Sluis reported on a series of 121 trauma patients age 60 and greater, all with ISS ? 16. No patient with an ISS greater than or equal to 50 survived in this series. The authors, however, do emphasize the importance of not using the ISS to predict outcomes in individual patients.^[35] A study by Carrillo, published in 1994, reported on 94 blunt trauma victims age 65 or greater.  Mortality correlated well with APACHE II, but the combination of APACHE II and ISS performed better than APACHE II alone. All patients with APACHE II ?15 and ISS ?30 died, but this accounted for only one-third of all deaths in this series.^[28] Thus, it would appear that there is little, if any, support in the literature to justify withdrawal of care based upon any combination of age and ISS.  Likewise, for SAPS, APS, and MAIS, there is no literature support for the use of any of these scores to predict individual patient outcome following geriatric trauma.^{[3] [19] [20] [26]} Finally, mention should be made of the GTSS, the Geriatric Trauma Survival Score. This score was derived by DeMaria, based upon his experience with 82 blunt trauma patients over the age of 65 years. The formula to calculate GTSS uses patient age, ISS, and the presence of absence of cardiac and septic complications to predict patient outcome.^[6]  Given the inadvisability mentioned above of using ISS to predict individual patient outcome, as well as the fact that information regarding the presence or absence of complications will not be obtainable prior to hospital discharge, the GTSS clearly has no role in guiding decision- making, a point which the authors themselves emphasize.  Interestingly, a larger and more recent study of blunt trauma patients aged 60 and over, failed to demonstrate any relationship between the GTSS and survival.^[25]

Complications and outcome

It is generally acknowledged that when the geriatric trauma patient sustains complications during their initial hospitalization that overall outcome is adversely affected. Both DeMaria and Osler , in comparing elderly survivors with non-survivors, have noted a statistically higher incidence of cardiac and septic complications^[6] and respiratory complications^[9] in non-survivors.  Other authors, employing logistic regression statistical methodology, have identified cardiac, infectious, and pulmonary complications as independent predictors of poor outcome following geriatric trauma.^{[5] [17] [25]} In addition to the specific types of complications sustained by the geriatric trauma patient, the number of complications sustained by a given geriatric trauma patient has been identified as a risk factor for poor outcomes. Smith, in a study of 456 trauma patients aged 65 and over, reported a 5.4% mortality for those patients with no complications, 8.6% for those with one complication, and 30% for those with more than one complication.^[37]Similar results have been noted for geriatric patients sustaining traumatic brain injuries.^[15]

In spite of the well documented relationship between complications and outcome in geriatric trauma, triage decisions are rarely, if ever, affected by this information. Early triage decisions, whether in the field or in the emergency department, clearly cannot be based upon the presence or absence of complications which yet to occur. Furthermore, there is no data to suggest that any particular number, or type, of complications will allow identification of the individual geriatric trauma patient destined for an outcome so dismal that a non-aggressive course of treatment could be justified. In light of these findings, efforts should be focused on the development and implementation of strategies aimed at the prevention of complications in the geriatric trauma patient. The importance of complication prevention is highlighted in a study by Pellicane which revealed that preventable complications contributed to mortality in 32% of all deaths in this series and in 62% of deaths related to multiple organ systems failure.^[16]

Outcome from geriatric head injury

The topic of geriatric head injury has received more attention in the literature than has any other aspect of geriatric trauma Unfortunately, all of it is retrospective in nature and, therefore, suffers from many of the same methodological shortcomings discussed above for the remainder of the geriatric trauma literature. These include lack of a specific age definition for geriatric head injury, lack of standardized definitions for specific sub-populations of geriatric head injured patients, and lack of standardized outcome measures. In addition, much of the geriatric head injury literature either provides insufficient details regarding head injury management, or provides results based upon head injury management that would be considered outdated by today’s standards. Therefore, it is difficult, and perhaps even dangerous, to make meaningful recommendations regarding the triage of current day geriatric neuro-trauma patients based upon the existing literature. Despite these shortcomings, there is little question that outcomes following traumatic brain injury are much worse in geriatric patients than in their younger counterparts. Vollmer, in a study from the Traumatic Coma Databank, reported on 661 patients age 15 and older with severe brain injuries, defined as GCS less than eight. Mortality for the entire series was 38%, but was 80%  for patients greater than 55 years of age. Multi-variate analysis revealed age to be an independent and significant predictor of death and vegetative outcome, beginning at age 45.^[15] Another study examined the effect of age upon outcome in patients with acute subdural hematomas. Mortality was 18% in patients between the ages of 18 and 40, but 74% in patients greater than age 65.  Once again, advanced age was noted to be a predictive of poor outcomes.^[38]  In addition to age, a number of other factors have been examined as potential predictors of poor outcome following head injury in geriatric patients. Not surprisingly, the most extensively studied factor is that of admission GCS. Many other factors predictive of poor outcome have been examined including anatomy of the brain injury (epidural versus subdural),^[39] need for craniotomy,^{[15] [40-43]} subdural hematoma volume,^[38] midline shift,^[38,43]pupillary status,^{[40] [42] [44]} and intracranial pressure.^[15,45]  None of these factors has been examined in sufficient detail to allow us to make any recommendations regarding their potential role as triage tools in geriatric head injury. Therefore, they will not be considered further within this document.

“Low” admission GCS is clearly associated with poor outcomes in elderly head-injured patients. Reuter documented a mortality rate of 87% in elderly patients (age >60) with traumatic intracranial hemorrhage and an admission GCS < 8, though no details regarding head injury management were provided.^[46] The available scientific literature, however, does not support the use of a specific GCS that will reliably identify patients destined for poor outcomes.  Zietlow, in a study of patients age 65 and older with multi-system injury, identified a GCS ? 8 as being predictive, while Van Aalst, in a similar study, found a GCS ?  7 to be associated with death or depending living status.^{[5] [29]} Published studies limited to geriatric patients with head injuries likewise yield no consensus. Rozelle found a GCS ? 7 to be predictive of hospital mortality in patients with subdural hematomas, and Kilaru noted that this same GCS was associated with a universally poor long-term outcome.^{[43] [44]} Cagetti, however, found that a GCS ? 11 was associated with a 100% mortality, although this study involved patients 80 years of age and older.^[47]  Amacher, however, in a similar study of head-injured patients 80 years of age and older, did achieve an “excellent/good” outcome in a single patient with an admission GCS score in the 3 to 6 range.^[48]  Thus, it seems that, while “low” GCS scores are indeed associated with poor outcomes, it does not seem possible, or advisable, based upon the existing literature, to make triage decisions in head-injured geriatric patients based solely upon the admission GCS. It does seem reasonable to conclude that head-injured patients, 65 years and older, have very poor outcomes when the admission GCS is ? 7 or 8.

Other authors have examined the prognostic value of the “delayed” GCS score – that is, the GCS score determined 24 hours or more following injury.  Both Pennings and Kotwica have advocated a limited course of aggressive therapy in geriatric trauma patients with severe head injuries, although their GCS definitions of futility differ greatly. Kotwica, in a study of head-injured patients 70 years of age and older, noted a 90% mortality in patients with a GCS < 9 when craniotomy was required, and 76% when craniotomy was not required. Based on this finding in 136 patients, they recommend aggressive treatment for 24 hours only for those patients without space occupying lesions. Aggressive treatment, then, is continued only in those patients who show “significant” improvement within this time frame.^[41] Pennings, in his study of 42 patients age 60 and greater with GCS ? 5, concluded that these patients have an extremely poor prognosis, and that if they have not regained “substantial” neurologic function within 24 hours, they are unlikely to do so.^[42] Similarly, Ross reported a 100% six month mortality among patients 65 years of age and older who had a persistent GCS ? 8 at 72 hours following admission.^[45]  Even though the overall prognosis from geriatric head injury may have improved since these publications due to improvements in head injury management, it is reasonable to expect that these new therapies will exert their maximum effect in the early stages following injury. Thus, in geriatric head injury, it seems reasonable to adopt an initial course of aggressive treatment (with the possible exception of the patient who is moribund upon arrival), followed by a re-evaluation of the patient’s neurologic status at 72 hours post admission. The intensity of the subsequent care provided can then be based upon the initial response to therapy.

V. Summary

While multiple clinical and demographic factors have demonstrated an association with outcome following trauma in geriatric patients, the ability of any specific factor alone, or in combination with other factors, to predict an unacceptable outcome for any individual geriatric trauma patient is quite limited. An initial course of aggressive therapy (see following section entitled “Geriatric Resuscitation”) seems warranted in all geriatric trauma patients, regardless of age or injury severity, with the possible exception of those patients who arrive in a moribund condition. Geriatric trauma patients who do not respond to aggressive resuscitative efforts within a timely fashion are likely to have poor outcomes even with continued aggressive treatment. Modification of the intensity of treatment provided to these “non-responders” should be considered. For those geriatric trauma patients who do respond favorably to aggressive resuscitative efforts, the prognosis, not only for survival but also for return to their pre-injury level of function, is quite good, and certainly justifies the effort.

VI. Future investigations

There are no Class I data that address triage issues in geriatric trauma.  Prospective randomized controlled trials are desperately needed which address the prognostic values of age, injury severity and injury physiology upon ultimate outcome following geriatric trauma. Prior to conducting these studies, there must be agreement concerning the specific age definitions to be used for geriatric trauma, the outcomes to be measured, and the specific clinical criteria that will be used to define pre-existing medical conditions.  Furthermore, data generated in such a fashion should be subjected to rigorous and appropriate statistical analysis. Only when a substantial body of literature exists which meets these criteria, will trauma practitioners succeed in providing an appropriate level of care to the geriatric trauma patient based on that patient's predicted outcome.

VII. References

1. Committee on Trauma, American College of Surgeons: Resources for optimal care of the injured patient: 1999, Chicago, American College of Surgeons; 1998:14.
2. Champion HR, Copes WS, Buyer D, Flanagan ME, Bain L, Sacco WJ: Major trauma in geriatric patients. Am J Public Health 1989;79:1278-1282.
3. McCoy GF, Johnston RA, Duthie RB: Injury to the elderly in road traffic accidents. J Trauma 1989;29:494-497.
4. Morris JAJ, MacKenzie EJ, Damiano AM, Bass SM: Mortality in trauma patients: the interaction between host factors and severity. J Trauma 1990;30:1476-1482.
5. van Aalst JA, Morris JAJ, Yates HK, Miller RS, Bass SM: Severely injured geriatric patients return to independent living: a study of factors influencing function and independence. J Trauma 1991;31:1096-1101.
6. DeMaria EJ, Kenney PR, Merriam MA, Casanova LA, Gann DS: Survival after trauma in geriatric patients. Ann Surg 1987;206:738-743.
7. Davis JW, Kaups KL: Base deficit in the elderly: a marker of severe injury and death.  J Trauma 1998;45:873-877.
8. Gubler KD, Davis R, Koepsell T, Soderberg R, Maier RV, Rivara FP: Long-term survival of elderly trauma patients. Arch Surg 1997;132:1010-1014.
9. Osler T, Hales K, Baack B, et al: Trauma in the elderly.  Am J Surg 1988;156:537-543.
10. Finelli FC, Jonsson J, Champion HR, Morelli S, Fouty WJ: A case control study for major trauma in geriatric patients. J Trauma 1989;29:541-548.
11. Smith JSJ, Martin LF, Young WW, Macioce DP: Do trauma centers improve outcome over non-trauma centers: the evaluation of regional trauma care using discharge abstract data and patient management categories. J Trauma 1990;30:1533-1538.
12. Ma MH, MacKenzie EJ, Alcorta R, Kelen GD: Compliance with prehospital triage protocols for major trauma patients. J Trauma 1999;46:168-175.
13. Phillips S, Rond PC, Kelly SM, Swartz PD: The failure of triage criteria to identify geriatric patients with trauma: results from the Florida Trauma Triage Study.  J Trauma 1996;40:278­283.
14. Zimmer-Gembeck MJ, Southard PA, Hedges JR, et al: Triage in an established trauma system. J Trauma 1995;39:922-928.
15. Vollmer DG: Age and outcome following traumatic coma: why do older patients fare worse? J Neurosurg 1991;75:S37-S49
16. Pellicane JV, Byrne K, DeMaria EJ: Preventable complications and death from multiple organ failure among geriatric trauma victims. J Trauma 1992;33:440-444.
17. Perdue PW, Watts DD, Kaufmann CR, Trask AL: Differences in mortality between elderly and younger adult trauma patients: geriatric status increases risk of delayed death. J Trauma 1998;45:805-810.
18. Knudson MM, Lieberman J, Morris JAJ, Cushing BM, Stubbs HA: Mortality factors in geriatric blunt trauma patients.  Arch Surg 1994;129:448-453.
19. Horst HM, Obeid FN, Sorensen VJ, Bivins BA: Factors influencing survival of elderly trauma patients. Crit Care Med  1986;14:681-684.
20. Shabot MM, Johnson CL: Outcome from critical care in the "oldest old" trauma patients.  J Trauma 1995;39:254-259.
21. van der Sluis CK, Timmer HW, Eisma WH, ten Duis HJ: Outcome in elderly injured patients: injury severity versus host factors. Injury. 1997;28:588-592.
22. Oreskovich MR, Howard JD, Copass MK, Carrico CJ: Geriatric trauma: injury patterns and outcome. J Trauma 1984;24:565-572.
23. Broos PL, Stappaerts KH, Rommens PM, Louette LK, Gruwez JA: Polytrauma in patients of 65 and over. Injury patterns and outcome. Int Surg 1988;73:119-122.
24. Broos PL, D'Hoore A, Vanderschot P, Rommens PM, Stappaerts KH: Multiple trauma in elderly patients. Factors influencing outcome: importance of aggressive care. Injury. 1993;24:365-368.
25. Tornetta P, Mostafavi H, Riina J, et al: Morbidity and mortality in elderly trauma patients. J Trauma 1999;46:702-706.
26. Johnson CL, Margulies DR, Kearney TJ, Hiatt JR, Shabot MM: Trauma in the elderly: an analysis of outcomes based on age. Am Surg 1994;60:899-902.
27. Battistella FD, Din AM, Perez L: Trauma patients 75 years and older: long-term follow-up results justify aggressive management. J Trauma 1998;44:618-623.
28. Carrillo EH, Richardson JD, Malias MA, Cryer HM, Miller FB: Long term outcome of blunt trauma care in the elderly. Surg Gynecol Obstet 1993;176:559-564.
29. Zietlow SP, Capizzi PJ, Bannon MP, Farnell MB: Multisystem geriatric trauma.  J Trauma 1994;37:985-988.
30. DeMaria EJ, Kenney PR, Merriam MA, Casanova LA, Gann DS: Aggressive trauma care benefits the elderly. J Trauma 1987;27:1200-1206.
31. Day RJ, Vinen J, Hewitt-Falls E: Major trauma outcomes in the elderly.  Med J Aust. 1994;160:675-678.
32. Morris JAJ, MacKenzie EJ, Edelstein SL: The effect of preexisting conditions on mortality in trauma patients. JAMA  1990;263:1942-1946.
33. Milzman DP, Boulanger BR, Rodriguez A, Soderstrom CA, Mitchell KA, Magnant CM: Pre-existing disease in trauma patients: a predictor of fate independent of age and injury severity score. J Trauma  1992;32:236-243.
34. Charlson ME, Pompei P, Ales KL, MacKenzie CR: A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40:373-383.
35. van der Sluis CK, Klasen HJ, Eisma WH, ten Duis HJ: Major trauma in young and old: what is the difference? J Trauma 1996;40:78-82.
36. Shapiro MB, Dechert RE, Colwell C, Bartlett RH, Rodriguez JL: Geriatric trauma: aggressive intensive care unit management is justified. Am Surg 1994;60:695-698.
37. Smith DP, Enderson BL, Maull KI: Trauma in the elderly: determinants of outcome. South.Med J 1990;83:171-177.
38. Howard MA, Gross AS, Dacey RGJ, Winn HR: Acute subdural hematomas: an age-dependent clinical entity [see comments]. J Neurosurg 1989;71:858-863.
39. Rakier A, Guilburd JN, Soustiel JF, Zaaroor M, Feinsod M: Head injuries in the elderly.  Brain Inj. 1995;9:187-193.
40. Jamjoom A, Nelson R, Stranjalis G, et al: Outcome following surgical evacuation of traumatic intracranial haematomas in the elderly. Br J Neurosurg. 1992;6:27-32.
41. Kotwica Z, Jakubowski JK: Acute head injuries in the elderly. An analysis of 136 consecutive patients. Acta Neurochir.(Wien.) 1992;118:98-102.
42. Pennings JL, Bachulis BL, Simons CT, Slazinski T: Survival after severe brain injury in the aged. Arch Surg 1993;128:787-793.
43. Rozzelle CJ, Wofford JL, Branch CL: Predictors of hospital mortality in older patients with subdural hematoma. J Am Geriatr.Soc 1995;43:240-244.
44. Kilaru S, Garb J, Emhoff T, et al: Long-term functional status and mortality of elderly patients with severe closed head injuries.  J Trauma 1996;41:957-963.
45. Ross AM, Pitts LH, Kobayashi S: Prognosticators of outcome after major head injury in the elderly. J Neurosci.Nurs. 1992;24:88-93.
46. Reuter F: Traumatic intracranial hemorrhages in elderly people. Advances in Neurosurgery 1989;17:43-48.
47. Cagetti B, Cossu M, Pau A, Rivano C, Viale G: The outcome from acute subdural and epidural intracranial haematomas in very elderly patients. Br J Neurosurg 1992;6:227-231.
48. Amacher AL, Bybee DE: Toleration of head injury by the elderly.  Neurosurgery 1987;20:954-958.

Table

Evidentiary Table: Triage Issues in Geriatric Trauma