Improper triage seems to contribute to the poor outcomes experienced by some elderly trauma patients. The effectiveness of triage can be evaluated by looking at the interaction between injury severity and complication rate, mortality, or requirement for intervention.26 Although there are data suggesting that undertriage of elderly patients increases mortality rates, a recent population based study disputes this notion.27 These investigators evaluated data from 6015 patients in three counties in California and four from Utah. There was no difference in mortality in patients taken to trauma centers when compared to those taken to nontrauma centers. It should be noted that only 244 patients had an ISS greater than 15. Caution must be exercised in interpreting the conclusions of this study, but it certainly suggests that more work is needed to evaluate the impact triage has on elderly trauma patients. Elderly patients with severe injuries who are not treated with full trauma-team activation are considered undertriaged. Multiple studies have shown a large incidence of undertriage as compared to younger patients, so undertriage can be viewed as a modifiable risk factor for poor outcome in the older patient.26,27,28,29,30,50 Lehmann et al demonstrated that the classic physiologic criteria for trauma team activation, that is, blood pressure and heart rate, both failed to independently predict hospital mortality or the need for urgent interventions.28 These authors attributed the older patient’s “pseudostability” to declining functional reserve and the interaction of premorbid medications. Use of initial vital signs in the elderly population can be misleading. In a study from Los Angeles, patients 70 years of age and older who were admitted to the trauma center were reviewed. The hypotension or tachycardia criteria for trauma team activation (TTA) were not met in 63% of patients with an ISS greater than 15% and 25% of patients with an ISS greater than 30. In this study the overall mortality in “stable” patients not meeting any of the standard TTA criteria was 16%.26 The National Trauma Triage Protocol now suggests that a systolic blood pressure (SBP) of 110 mm Hg be used as a criterion for transport to a trauma center for patients above the age of 65 years. Brown et al from Pittsburgh recently evaluated data from the NTDB and were able to show that the new criterion of a SBP of 110 mm Hg in patients older than 65 increased the sensitivity in predicting mortality in geriatric trauma patients.31 In a similar approach, the group at the University of Arizona has suggested that the Shock Index (SI = HR/SBP) can be used as a field triage tool for elderly patients. Using data from the NTDB, these investigators were able to demonstrate that an SI greater than 1 reliably predicted mortality with statistically better fidelity than the individual components of the index.32
Interestingly, an apparently “minor” injury can have major impact on the older patient. Rib fractures and pulmonary contusions can lead to an abrupt decompensation, and injuries such as intracranial hemorrhage are commonly underappreciated. It has been suggested that a patient age of 70 or older be used as a criterion for trauma team activation.26 The age at which triage and management issues become problematic is controversial, also. The current recommendation of the ATLS program is 55 years of age.33 This is based on data from the Major Trauma Outcome Study (MTOS) which noted a significant increase in mortality between the ages of 45 and 54 years.34 TRISS uses a similar age cutoff, although a recent work examining TRISS methodology seems to indicate an older age is more accurate. Using the Ohio trauma registry, Caterino et al examined mortality trends. Regression analysis identified 70 years of age to be the most promising cutoff for predicting increased odds of mortality.29
The airway of the elderly patient poses specific challenges for providers as these individuals have a significant loss of protective airway reflexes. Patients may have dentures or be edentulous with the former making bag-mask ventilation easier, while arthritic changes may make mouth opening difficult. Finally, when performing rapid sequence intubation, the doses of barbiturates, benzodiazepines, and etomidate should be reduced between 20 and 40% to minimize the risk of cardiovascular depression.35
The anatomical and physiologic changes in the respiratory system associated with aging are reviewed above. Changes in the compliance of both the lungs and the chest wall result in an increased work of breathing with aging. These changes associated with the possibility of nutritional deficits and the supine position place the elderly trauma patient at high risk for respiratory failure. Given a suppressed heart rate response to aging, respiratory failure may present in a more insidious fashion. Diagnosis can sometimes be difficult in interpreting clinical and laboratory information in the face of preexisting respiratory disease or nonpathologic changes in ventilation associated with age. Frequently, decisions to secure a patient’s airway and provide mechanical ventilation may be made prior to fully appreciating underlying premorbid respiratory conditions. As noted above, the risk of ventilator associated pneumonia and the possibility of prolonged ventilation are significant.35 The role for noninvasive mechanical ventilation in the acute resuscitative phases of trauma care seems to be very limited and is likely associated with significant risk to the patient.
Age-related changes in the cardiovascular system place the elderly trauma patient at significant risk for being mislabeled as being “hemodynamically normal.” Since the elderly patient may have a fixed heart rate and cardiac output, the response to hypovolemia will occur by increasing systemic vascular resistance. To further demonstrate the lack of classic symptoms as they relate to cardiovascular pathology, Chong et al evaluated troponin I levels following emergency orthopaedic surgery in 102 patients over the age of 60, and 52.9% had elevated levels. The majority of patients with elevated troponin levels had no cardiac symptoms, but had an increased mortality within 1 year of the event. Furthermore, since many elderly patients have preexisting hypertension, the seemingly “acceptable” blood pressure may truly reflect a relative hypotensive state. As such, identifying the patient who has significant tissue hypoperfusion is mandatory. Several measurements continue to be used to make this diagnosis. These include base deficit, serum lactate, age-adjusted Shock Index, and tissue-specific end-points.8,36,37,38 Resuscitation of the geriatric hypoperfused patient is the same as all other patients and based upon appropriate fluid and blood administration. The elderly trauma patient with evidence of circulatory failure should be assumed to be bleeding. Given the incidence of elderly people with preexisting disease states, however, one should keep in mind that some physiologic event may have triggered the incident leading to injury. Ultimately an aggressive approach to resuscitation of the elderly patient with overt shock or tissue hypoperfusion should result in acceptable outcomes. Less aggressive measures based upon the patient’s age are not acceptable.
Traumatic brain injury (TBI) is a problem of epidemic proportion in the elderly population, and older age is a known variable for a poor outcome following brain injury.39 Aging will cause the dura to become more adherent to the skull. Additionally, older patients are more commonly prescribed anticoagulant and antiplatelet medications for preexisting medical conditions. These two factors place the elderly individual at high risk for intracranial hemorrhage. Atherosclerotic disease is common with aging and may contribute to a primary or secondary brain injury. Moderate cerebral atrophy will permit intracranial pathology to initially present with a normal neurologic examination. Early identification and timely appropriate support including correction of therapeutic anticoagulation can improve outcomes.40
Musculoskeletal changes associated with the aging process pose special concerns during the initial assessment of the elderly trauma patient. Loss of subcutaneous fat, nutritional deficiencies, chronic medical conditions, and associated medical therapies will place the elderly patient at risk for hypothermia and the risks associated with immobility (pressure ulcers, delirium). Rapid evaluation and early mobilization will prove to minimize morbidity.
In the elderly age group, traumatic brain injury (TBI) accounts for more than 80,000 emergency department visits each year of which a majority result in hospitalization. Falls are the leading cause of TBI for people over the age of 65 years (51%), followed by motor vehicle collisions (9%).39 Health care costs associated with the treatment of TBI in the elderly exceeded $2.2 billion in 2003.41
The current evidence-based approach to treating severe TBI in the adult is a “one size fits all” approach that neglects specific issues of the older adult. An investigation using the New York State Trauma Registry compared mortality and functional outcome in elderly versus younger patients.42 In this study, Susman et al demonstrated increased mortality in patients older than 65 years, but also showed that mortality increased as patients aged. These investigators also showed that a majority of elderly patients sustain TBI from falls and appear to have only a mild traumatic brain injury at admission, but still have a much higher mortality as compared to younger patients. This same group then looked at the total effect of age on mortality after TBI.43 They concluded that mortality from TBI increases after 30 years of age, but has a sharp rise after the age of 70. Given the anatomic changes associated with aging on the brain and the fact that a majority of elderly patients present with a Glasgow Coma Scale (GCS) consistent with a mild brain injury, a high index of suspicion must be maintained with the elderly patient presenting with any mechanism of head trauma. To address this, Mack et al investigated the use of computed tomography (CT) of the brain in elderly patients.44 This study specifically looked at mild brain injury (GCS 13–15). In their study of 133 elderly patients, 14.3% had radiographic evidence of acute intracranial pathology. The authors noted that there were no useful clinical predictors of intracranial injury and recommended liberal use of CT of the brain, also. In addition to a higher mortality in general, patients surviving hospitalization will require aggressive rehabilitation. In a multi-institutional trial, a group of elderly patients surviving their initial moderate to severe brain injury (Head AIS = 3) were evaluated following discharge from acute care. In this cohort there were few patients with a low GCS who survived, and this left patients with a GCS of 13–15 to be evaluated. Functional outcome for these patients, as measured by the Glasgow Outcome Scale (GOS) and modified Functional Independence Measure (FIM), was good to excellent. Older patients, however, required more inpatient rehabilitation and took longer to recover when compared to younger patients.45 Aggressive initial management and long-term rehabilitation in the elderly patient with TBI is required for acceptable outcomes.
Rib fractures in the elderly pose a significant risk for morbidity and mortality when compared to younger patients who, in general, suffer little morbidity. The morbidities in the elderly include inadequate pain management, need for intubation, prolonged ventilatory support, and the development of pneumonia. Bulger et al investigated the impact of rib fractures after blunt chest trauma in the elderly.46 This study showed a linear relationship between age, number of rib fractures, complications and mortality. In a similar study, Holcomb et al retrospectively evaluated 171 patients.47 These authors demonstrated an increase in negative outcomes based upon increasing age and number of rib fractures (Fig. 44-2). By grouping ages and number of rib fractures their data revealed that patients with more than four rib fractures who are older than 45 years had increased morbidity (ICU length of stay [LOS], total LOS, ventilator days, and pulmonary complications). Given the impact of type of analgesia at reducing ventilator days and pulmonary complications, the authors attempted to determine the impact of epidural analgesia. Their data did not demonstrate a decreased incidence of morbidity and mortality. Given that this was only a portion of their entire population, it is possible that their results suffer from a type II statistical error. The Eastern Association for the Surgery of Trauma has a Practice Management Guideline on analgesia for management of chest trauma.48 Readers are encouraged to refer to this document on evidence based guidelines for analgesia. In an evaluation of data from the NTDB of the American College of Surgeons Committee on Trauma, Flagel et al reviewed a large patient population.49 These investigators showed that the overall mortality rate for patients with rib fractures was 10%. This rate increased for each additional rib fracture independent of age. A similar trend noted increasing pulmonary complications with additional rib fractures. The incidence of pneumonia in patients with up to five rib fractures was from 3 to 5.2%. This increased to 6.8–8.4% for patients with 6 or more rib fractures. These authors were unable to demonstrate that age is a risk factor for mortality in patients with rib fractures. Most recently, data from a multicenter study of 1621 patients was published by the Research Consortium of the New England Centers for Trauma.50 These investigators evaluated patients over the age of 50 years with nearly isolated rib fractures. Of interest, 35% of the patients were admitted to the ICU with an average ICU LOS of 16.5 days and a total hospital length of stay of 27.5 days. Intubation was required in 12% of patients, and 4.3% went on to require tracheostomy. Univariate analysis of the data revealed that risk factors for mortality were preexisting coronary artery disease or congestive heart failure (CHF), increasing age, Injury Severity Score (ISS), number of ribs fractured, and increasing Abbreviated Injury Scale (AIS) for associated body regions. On multivariate analysis the strongest predictors of mortality were admission to a high-volume trauma center, preexisting CHF, intubation and increasing age. Patient controlled analgesia showed a trend toward improved survival, but was not statistically significant. The only therapeutic maneuver that proved to be predictive of survival was tracheostomy. Identification of the elderly patient with rib fractures and early recognition of respiratory failure with aggressive supportive maneuvers will potentially reduce morbidity and mortality.
Nonoperative management of injuries to solid organs has generally become the preferred method of treatment for the hemodynamically normal patient; however, this approach must be used with caution in the elderly trauma patient. The classical physiologic response to hemorrhage which may be used as a criterion to attempt nonoperative care or as a marker for “failure” of nonoperative management may not be present. In a 6-year retrospective analysis of patients sustaining blunt splenic injuries, Albrecht et al evaluated the utility of nonoperative management of these injuries in patients older than 55 years.51 In this small study of 37 patients meeting inclusion criteria, 13 patients went directly to the operating room. Of the remaining 23 patients, nonoperative management was successful in 15 (62.5%) and failed in 8 (33.3%). Characteristics of the group that failed nonoperative management included higher AAST splenic organ injury scales and large hemoperitoneums. In a larger retrospective study of 1482 patients, 15% (n = 224) 55 years or older, the mortality of the older population was significantly higher (43% vs 23%) than the younger group. In this study 80% of patients over the age of 55 years were successfully managed nonoperatively, with 24 patients of the original 132 patients subsequently requiring exploration. Although not statistically significant, there was a trend toward an increased failure rate with increasing grade of injury. When evaluated by grade of injury, grade I injuries had a success rate for nonoperative management which was similar for younger and older patients. Success rates for older patients were lower for grade II (73% vs 54%) and grade III (52% vs 28%) injuries. All elderly patients with grade IV–V injuries required operation, either immediately or for failed nonoperative management.52 The data suggest that nonoperative management of splenic injury in the elderly patient should be undertaken with caution. An appreciation of the grade of injury and assuring hemodynamic stability are absolute requirements. Elderly patients who fail nonoperative management may have a higher mortality as compared to younger patients.
Fractured bones are extremely common after trauma in the elderly population, and women are at particular risk because of osteoporosis. Pelvic fractures pose a significant special risk for elderly patients, with challenges involving the acute phase of fracture management, timing of operation and functional outcome. The group from the R Adams Cowley Shock Trauma Center compared outcomes of patients (<55>) sustaining pelvic fractures over a 2-year period.53 A blood transfusion was required in 62% of older patients compared to 36% in the younger population (p < 0.0035). Further analysis revealed that patients who received transfusion were 2.8 times more likely to be over the age of 55. Mortality for the younger group was 6.2% compared to 20.5% for the older group (p = 0.0034). Even after adjusting for ISS, death was four times more likely in the elderly patients. The authors suggested that every elderly patient with a pelvic fracture should be considered hemodynamically unstable until proven otherwise.
Another consideration in the management of the elderly patient with orthopaedic trauma is the timing of operation. The decision to proceed with an orthopaedic operation is made based on extra-orthopaedic injuries, physiologic status, and the magnitude of the operation planned. In a series of 367 elderly patients with hip fractures, a delay in operation for more than 2 days was associated with more than double the risk of death within the first postoperative year.54 Most believe that an orthopaedic operation to allow for mobilization from bed should occur as soon as possible when physiologic conditions have been optimized. Functional outcome is an additional consideration in caring for fractures in the elderly. Therefore, careful consideration by the orthopaedic traumatologist is required when deciding between operative versus nonoperative management as it relates to function. Osteopenic bones need to be taken into consideration when dealing with periarticular fractures, also. Aggressive physical and occupational rehabilitation have been shown to improve outcomes as determined by patient satisfaction.
The elderly burn patient has a higher mortality for a given size burn than younger patients. In fact, predictors of survival from burn injury continue to rely on age as a significantly weighted variable. Despite significant advances in the science and management of burn injury, the LD50 for burns in the 65 years or older population remains approximately 35% total body surface area (TBSA). Older individuals are at significant risk for burn injury due to impaired mobility, diminished senses, and slower reaction time, all of which make it difficult to escape harm, ultimately leading to deeper and more extensive burns.
Predicting survival in older burn patients is important. It provides a framework for discussing realistic clinical expectations with patients and their families and may allow for appropriate utilization of resources associated with the intensive care of burn patients. Wibbenmeyer et al attempted to address this predictive model in relation to modern burn care.55 They reviewed 308 burn patients over the age of 60 years. A majority of these patients sustained flame burns including 41.4% in household incidents. At least one preexisting medical problem was present in 64% of the patients, and the median TBSA size was 13%. The mortality for the cohort was 30.2%, with an LD50 of 30%. When this was further evaluated, the LD50 was age dependent. Specifically, it was 43.1% for patients aged 60–69.9, 25.9% for ages 70–79.9, and 13.1% for those 80 years and older. As expected the presence of an inhalation injury had a significant negative impact on survival. These authors concluded that death was significantly related to age, TBSA burn, and presence of inhalation injury, while comorbidities did not impact mortality. The Baux score, calculated as the sum of the age of the patient and the TBSA burn, is an estimate of the percent mortality and is supported by data from this study. The Abbreviated Burn Severity Injury (ABSI) score, calculated as the weighted sum of age, gender, TBSA, percentage of full thickness injury, and presence of inhalation injury, was also predictive of survival, but less so than the Baux score. The authors were unable to demonstrate an improvement in survival over the 20-year period of data collection. Another study from the Burnett Burn Center at Kansas University Medical Center reviewed retrospectively three decades of care of elderly burn patients.56 A total of 201 patients over the age of 75 years was evaluated. This group had a 77% mortality rate in the 1970s, compared to a combined overall mortality of 41% during the 1980s and 1990s; however, the LD50 remained at 20% TBSA.
Ho et al reported outcome data on 94 patients over the age of 60 years from their regional burn center.57 When compared to a younger population of burn patients, the older patients had a longer length of stay, a mortality of 7.4% and no differences in outcome based upon early versus late excision. Deitch et al had shown earlier that early wound closure decreased LOS, number of septic complications and improved mortality.58 The studies from Ho and Wibbenmeyer did not show the benefits of early excision.55,57 Finally, gender seems to impact outcomes in elderly burn patients. Chang et al from the University of Utah demonstrated a higher mortality, longer LOS, and less likelihood of being discharged home for older female patients when compared to older men.59
Maintaining an aggressive approach to the older burn patient is warranted in the context of survivable injuries. Advanced age, larger burns, and the presence of inhalation injury are all negative predictors of survival in this population. Realistic therapeutic expectations are important for patients, their families, and the burn team when taking care of these very challenging patients.
Falls are the most common mechanism of injury reported in the NTDB (42%).60 Of these patients, more than half are over the age of 65. For older trauma patients, many of these falls are from a standing height, but still lead to significant injuries requiring hospitalization.61 Preexisting medical conditions, medications, and other variables play a significant role in ground level falls. Unfavorable discharge dispositions and mortality are significant in this group of patients. The group from Harborview Medical Center examined 1352 elderly patients admitted to their trauma center after a ground level fall. Deaths occurred in 12% of the patients during the index admission. Of the survivors, 50% were discharged to a skilled nursing facility (SNF) and only 6% were discharged to home. Nearly 45% of patients were readmitted within 1 year following discharge, with patients requiring an admission to the ICU during the index admission to be at highest risk for readmission. The overall 1-year mortality for the entire group, including those who died during the index admission, was 33%. Patients discharged to a SNF were three times more likely to die than a patient discharged home.62
As the societal and financial impact for older adults sustaining falls is significant, developing effective prevention measures is mandatory. Optimizing medications to prevent hypotension and hypoglycemia, encouraging physical exercise to prevent osteoporosis and prevent skeletal muscle loss, and environmental modifications to reduce clutter and provide safety features all play a significant role in preventing falls.
With the growth in the proportion of elderly people there will be an obligatory rise in the number of elderly trauma patients. By far, blunt trauma is the predominant mechanism for injury; however, there are still a significant number of people over the age of 65 years who are victims of penetrating injury. It should be remembered that many of these patients may be victims of an attempt at suicide, a major cause of death in the geriatric population. When examining patients over the age of 55 between the years 1982 and 1987 in the original MTOS, elderly patients had a higher mortality when compared to a younger cohort.63 Also, Finelli et al reported a significantly higher mortality in the elderly (52%) compared to younger patients (20%) sustaining gunshot wounds.64 In contrast, Roth et al from Los Angeles County Hospital and the University of Southern California retrospectively reviewed 79 patients over the age of 55 years who suffered penetrating injuries. Their data showed no difference in mortality between elderly and younger patients (23% vs 18%). Interestingly, 50% of the elderly patients who died presented with “normal” vital signs.65
In a large retrospective statewide trauma registry review of 22,571 patients with blunt trauma of whom 7117 were elderly, ICU utilization was evaluated.66 The entire population had an ICU admission rate of 42.7% with a mean ICU LOS of 5.77 ± 8.86 days. In contrast, the elderly patients had a lower ICU admission rate of 36.7% when compared to the younger population (45.5%). Interestingly, those elderly patients admitted to the ICU had a significantly longer ICU LOS. The lower utilization of ICU resources in the elderly may be explained by a higher early mortality and/or end-of-life decisions that may have precluded admission to an ICU setting.
Preventable complications in the elderly trauma patient significantly impact outcome. DeMaria et al demonstrated an association of complications with death (32%) and an association of deaths from multisystem organ failure (62%).67 The relationship between senescence and infection deserves mention. Bochicchio et al showed an increase risk of infection in elderly patients (39%) when compared to a younger population (17%) (p < 0.05).68 Once infected, the elderly patient is at much higher risk for death (28%) than his or her younger counterparts (5%) (p < 0.005). An aggressive approach to the older patient with infection is essential for improved outcomes.
Scalea et al suggested that early invasive monitoring of the geriatric trauma patient can improve outcome.69 Noninvasive monitoring of elderly patients has been investigated in the critical care setting, but little has been concluded regarding the critically ill trauma patient.71 Recently, Murthi et al have looked at the utility of cardiac ultrasound to evaluate hemodynamics and resuscitation.70 In this study these investigators compared focused rapid echocardiographic evaluation (FREE) to assessment using a vascular catheter. They were able to show that the echocardiographic assessment correlated well to data from a pulmonary artery catheter, with the benefit of noninvasiveness and “real time” evaluation after therapeutic interventions. Although not specifically studied in elderly patients, this noninvasive method will likely become a mandatory diagnostic technique for intensivists.
Because the elderly will fail to demonstrate the classic physiologic responses to shock it is important to maintain an aggressive approach toward monitoring and resuscitation. The utility of invasive monitoring and aggressive critical care is demonstrated in a study on outcomes of elderly patients with the acute respiratory distress syndrome (ARDS). In this study, Eachempati et al evaluated a protocol approach to ARDS utilizing lung protective ventilator strategy and invasive hemodynamic monitoring. In a group of 210 elderly patients with ARDS, the investigators were able to demonstrate a lower mortality when compared to historical controls despite a higher severity of illness.72
Delirium in the elderly trauma population is now recognized as a significant risk factor for morbidity and mortality in the ICU. Delirium is characterized as a disturbance of consciousness associated with a fluctuating mental status and disorganized thought and occurs in at least 20% of hospitalized patients over the age of 65 years. Also, it increases hospitals costs by $2500 per patient resulting in 6.9 billion dollars of Medicare hospital expenditures.73 There is also a threefold higher mortality over 6 months following a single episode of delirium in the ICU.74 Understanding the impact of delirium in the trauma patient has only recently been investigated. Pandharipande et al undertook a prevalence study of delirium in their trauma and surgical ICUs. Using the CAM-ICU, a validated screening tool, these investigators found an overall prevalence of delirium of 70% (73% in surgical patients and 67% in trauma patients).75 The group from the trauma center at Denver Health Medical Center performed a 4-month study of 69 patients admitted to their trauma ICU following injury. They found a 59% incidence of delirium overall and higher if the patient was mechanically ventilated. Upon univariate analysis age proved to be a predictor for the development of delirium; however, on multivariate analysis the strongest predictors for transitioning into delirium were lower arrival GCS, higher packed red blood cell transfusion and higher multiple organ failure (MOF) score.76 Most recently, a two center study examining the impact of delirium on outcomes was performed. A total of 134 patients were evaluated, and 63% progressed to delirium during their ICU stay. The patients with delirium had more ventilator days and longer ICU and hospital LOS. These investigators were unable to show a relationship of increasing age and the development of delirium.77 The association of advanced age and delirium is strong in most studies, and delirium and dementia are highly interrelated; however, the nature of this interrelationship remains poorly examined.73 The association of delirium with opioid narcotics and benzodiazepines is well-established and care must be taken in prescribing these medications to all patients, but especially the elderly. In addition, the issue of appropriate and goal-directed analgesia and sedation is commonly faced by the ICU staff. Finally, delirium represents one of the most common preventable adverse events among older persons during hospitalization.78 As such, the development of delirium in a patient may reflect processes of care in the hospital and, therefore, a reflection of the quality of care of that individual. Those caring for elderly injured patients must be aware of the development, diagnosis, prevention and, if needed, appropriate treatment of delirium in order to provide the highest quality of care to this population.79
Injured elderly patients admitted to the hospital consume significant health care resources. Average hospital lengths of stay are generally around 10 days, and usually the LOS in the ICU is longer than for younger patients (except for early deaths). And, hospital acquired complications are independent predictors of mortality in this age group. In 2002, Richmond et al reported their 10-year retrospective review of geriatric trauma from a statewide trauma registry.80 They evaluated all patients 65 years or older who were included in the registry. Nearly 62% of the patients were injured as a result of a fall, while 22.6% were injured in a vehicle collision. Operation was required in 28% of patients, and 37% of the cohort had a preexisting medical condition. The average length of stay was 11.5 days, 1/3 of the group required admission to the ICU and 10% of all patients died. These investigators found that as age increased, the patients had higher mortality, more complications, and more required discharge to a facility other than home. The patients who died had a greater number of injuries with more body regions involved with a resultant higher Injury Severity Score (ISS).
Admission physiologic markers may predict those at risk for mortality. In a statewide trauma registry review from Pennsylvania, elderly patients who were either hypotensive, had a GCS of 3, or had a respiratory rate of less than 10 all had a significantly increased risk of death.81 A recent meta-analysis attempted to understand predictors of mortality in severely injured elderly patients. The authors were able to show that increasing age and injury severity both contributed to mortality, but the impact of age plateaued at 84 years. The study also corroborated increased mortality in patients presenting with a systolic blood pressure of less than 110 mm Hg.82
The interaction between injury and patients factors has been studied extensively, but data are conflicting regarding the impact of these factors on mortality. Despite this, there is a large body of evidence documenting that preexisting conditions (PECs) will impact morbidity and, likely, mortality. Morris et al identified five PECs that appeared to influence outcomes.83 In that study of over 3000 patients, one-fourth of patients over the age of 65 years had one of the five PECs. These investigators identified cirrhosis, congenital coagulopathy, chronic obstructive pulmonary disease (COPD), ischemic heart disease and diabetes mellitus as PECs which influenced outcomes in trauma patients. Patients with one or more of these PECs were nearly two times more likely to die than those without PECs. The same authors then reported the interaction between injury and host factors, which included age, gender, and PECs.84 Although injury severity was the primary determinant of mortality, host factors also played a significant role. These studies were later corroborated by Grossman et al in which an ISS greater than 30 was considered the LD50. Each additional year of age greater than 65 years carried a mortality increase of 6.8%, and PECs with the greatest impact on mortality were hepatic disease, renal disease, cancer, and congestive heart failure.85 The impact of PECs was nicely demonstrated in a study from the group at Vanderbilt University in which they studied a predictive mortality model incorporating PECs (University Health System Expected Mortality [UHC-EM]) and compared it to TRISS methodology. The UHC-EM is derived from data on diagnosis related groups (DRG) and includes both anatomic injury and PECs, whereas TRISS uses only unadjusted vital signs and pattern of injury. The study was a 7-year retrospective review of all patients admitted to the Vanderbilt Trauma Center, and elderly patients were compared to a younger cohort. The UHC-EM was able to predict mortality better than TRISS in elderly patients demonstrating the impact of PECs.86
There seems to be a relationship between PECs and the development of post-injury complications which increases mortality. In a study reviewing a large statewide trauma registry by Taylor et al, 6.2% of elderly patients developed pneumonia with preexisting pulmonary disease and increased ISS as risk factors. Pneumonia predicted both an increased ICU and hospital LOS. In this study, 5.9% of patients developed an acute kidney injury which resulted in a greater than 10-fold increase in mortality. Finally, the development of sepsis, which occurred in only 1.2% of patients, significantly increased mortality, ICU LOS, and hospital LOS. The only risk factor identified with the development of sepsis was an increased ISS.81 The impact of failure to rescue in the elderly has been investigated in a large database of emergency surgical patients. The failure to rescue rate was twofold higher in the elderly as compared to younger patients, correlating with a similar disproportionate increase in mortality.87
The association between age as an independent predictor and outcome as documented by increased mortality and hospital LOS is well supported in the literature. Although hospital LOS is increased, ICU utilization is less than in a younger population.81,88 As previously noted, this is due to an increased early mortality seen in the elderly as well as decreased ICU admission resulting from advanced directives. Data from the Healthcare Financing Administration demonstrated that ICU use decreased with age and was least likely for patients over 85 years.41 Young et al looked at outcome in a cohort of elderly trauma patients.89 Although the older group had a higher mortality rate, there were no differences in hospital or ICU LOS or in ICU admission rate. In the study by Taylor et al, the elderly group had a lower ICU admission rate (36.7%) compared to younger trauma patients (45.5%); however, elderly patients, once admitted to the ICU, had a longer ICU LOS.81 The impact of withholding or withdrawal of care may influence ICU LOS. Plaisier et al examined the practices of end-of-life care in a group of trauma patients. They divided their patients into elderly and younger groups which they cared for in a single institution. These investigators found there was no age bias in the elderly trauma patients admitted to the ICU.90
Strategies for Improving Outcomes
A recent population based study of elderly trauma patients suggested that outcomes are equivalent in trauma centers and nontrauma centers. Given the small number of seriously injured patients included in the study one must interpret these data with caution.27 There are numerous studies suggesting that elderly trauma patients are undertriaged to trauma centers despite the proven mortality benefit of trauma center care. Statewide data from Maryland demonstrated that fewer elderly trauma patients were transported to trauma centers despite meeting physiologic criteria for trauma triage.91 In a similar study evaluating the Washington state trauma registry, Lehman et al demonstrated that patients over the age of 65 years, despite having an ISS greater than 15, were less likely to have the prehospital trauma system activated, the trauma team activated, and a full trauma team response.28 The group of undertriaged elderly patients had a significantly higher mortality, discharge disability rate, and complication rate when compared to a group of younger patients. Meldon et al demonstrated a significant survival benefit for severely injured elderly patients when treated in a trauma center (56%) versus a nontrauma center (8%).92 It is not clear why older patients are undertriaged; however, this group is less likely to demonstrate hypotension or tachycardia than are younger patients as previously noted.
The increased awareness in medication reconciliation and its impact on patient safety is prominent in current health care literature. The acuteness of the hospital admission and, at times, the initial underappreciation of significant anatomical injury results in a potentially modifiable situation highlighting the importance of knowledge of a patient’s premorbid medications and the potential interaction of any newly prescribed medications. The Beers criteria are used by consensus for safe medication use in elderly patients.93 Using this system along with expert input from geriatricians and pharmacists may reduce adverse drug reactions in older patients.
There are a number of medications specifically related to the injured patient that deserve mention. β-Blockers are used in approximately 20% of elderly patients with coronary artery disease and 10% of patients with hypertension. The inherent physiologic blockade of the expectant response to hypovolemia may provide obstacles to triage and treatment. Additionally, the impact of prolonged β-blockade on the progression and outcome of trauma and burn injury is not fully understood.
Anticoagulation with warfarin and/or antiplatelet therapy poses significant problems for the bleeding patient. Warfarin therapy carries a 1% per year risk of spontaneous intracranial hemorrhage. Assessment of the therapeutic level is important as many patients will be sub- or supratherapeutic. Many studies assessing the impact of warfarin on trauma and, specifically, trauma to the brain, are limited in that they do not address the clinical level of anticoagulation. To address this limitation, a study of 3242 trauma patients over the age of 50 years was performed. In this retrospective analysis data using the international normalized ratio (INR) was a surrogate for warfarin anticoagulation. These investigators demonstrated a mortality rate of 22.6% in the group with an INR greater than 1.5 compared to 8.2% for those with an INR less than 1.5. After adjusting for age and ISS, the odds of death for each one-unit increase in INR was 30%.94 Those patients with supratherapeutic anticoagulation fare even worse. A study of 49 patients with severe traumatic brain injuries and an average INR of 6.5 demonstrated an attributable mortality of 87.8%.95 A more recent study comparing patients with brain versus nonbrain injuries while on anticoagulation to a matched control group showed no difference in mortality.95 A limitation in this study was that INR data were not included.
Antiplatelet therapy is less well-studied. This class of drugs includes aspirin and clopidogrel and is increasingly used for a wide variety of cardiovascular problems. In a registry review of patients over the age of 50 years who sustained a traumatic brain injury, patients taking aspirin or clopidogrel were compared to those not receiving antiplatelet therapy. The group receiving antiplatelet therapy had a mortality of 23% compared to 8.9% (p = 0.016) in the control group. Risk factors for death in this study included age more than 76 years and GCS less than 12.97 Peck et al from San Diego retrospectively looked at a group of elderly patients with a blunt TBI. They compared those taking anticoagulants and/or antiplatelet medications at the time of injury with patients not taking such medications. The anticoagulation/antiplatelet group had a higher mortality, and the authors attributed this to the use of antiplatelet agents, the majority of which were clopidogrel.98 Patients sustaining a TBI while taking clopidogrel have been shown to have progression of intracranial hematomas (ICH). Joseph et al retrospectively reviewed 71 patients with an ICH who were taking clopidogrel at the time of injury. There was a fivefold increase in progression of ICH and a twofold increase in progression of ICH based upon clinical deterioration. These authors suggested routine repeat head CTs in patients with an ICH who had taken clopidogrel.99
Rapid reversal of anticoagulation with transfusion of fresh frozen plasma is beneficial. Reversal of an elevated INR to normal within 2 hours reduced mortality from 48% to 10% in one study.100 Recent Guidelines from the American College of Chest Physicians recommend the use of four-factor prothrombin concentrate (PCC) for the rapid reversal of the effects of Coumadin. The role for factor VIIa is not completely clear, though it may have a role in reversing anticoagulation in patients receiving warfarin.101
The role for the geriatrician in the management of the elderly trauma patient is logical. There are age specific nuances in the care of the elderly that make the geriatric specialist an invaluable member of the extended trauma team. Landefeld et al have shown that geriatric medical patients cared for in a special geriatric unit were less likely to be transferred to a long-term care facility than those cared for on a standard medical ward.102 Fallon et al developed the idea of a “geriatric trauma team.”103 This involved mandatory consultation from a group of interested and dedicated geriatricians. The consultants were able to address a variety of medical conditions in elderly trauma patients at study sites. These included pain management, rehabilitation, delirium, and advanced care planning to name a few. Most notably, the group of patients seen in consultation had a lower mortality than a group of elderly trauma patients not seen by a geriatrician. To study this concept prospectively, Lenartowicz et al from Toronto established a Geriatric Trauma Consultative Service (GTCS) that was comprised, at a minimum, of an advanced practice nurse in geriatrics and a geriatrician who performed a comprehensive geriatric assessment. All patients 60 years and older were evaluated by the GTCS. The authors looked at geriatric specific complications (falls, delirium, and restraint use) and trauma specific complications (pressure ulcers, venous thromboembolism, myocardial infarct, pneumonia, cardiac arrest, and missed injury). There were less consult requests to internal medicine and psychiatry and, ultimately, there was less delirium and a decreased incidence of patients being discharged to long-term care facilities.104
There are age specific interventions that may improve outcome in the elderly injured population. As previously noted, one of the first attempts to address this was work from the trauma service at Kings County Hospital Center in Brooklyn, NY. In this study the authors were able to demonstrate an outcome benefit from early invasive monitoring and intervention.69 A study from Nathens et al investigated the impact of an intensivist-model ICU on trauma related mortality.105 In this large multicenter study the authors concluded that an intensivist-model ICU when compared to an open-model ICU significantly reduced mortality. Furthermore, the greatest risk reduction was seen in the elderly population (RR, 0.55 [0.39–0.77]).
The ICU is an environment in which end of life (EOL) decisions are made on a regular basis. In fact, 38% of all hospitalized patients who die in the United States spent at least 10 days in the ICU in one review.106 Older patients, having an understanding of their own wishes regarding health care, frequently utilize advanced directives and health care proxies. Many times patients know these legal entities through interaction with an astute family physician or by way of a previous elective hospital admission. Understanding the logistics of these documents is mandatory when caring for trauma patients. Given the acute onset of trauma, as well as a patient’s inability to predict future injury, requires trauma practitioners to rely on a written document which may have been written years prior or by interactions with a health care proxy. The health care proxy is an individual who is appointed by a competent adult. In the event that the individual is no longer able to make health care decisions, the proxy would “speak” on behalf of the patient. There is evidence that physicians frequently have differing opinions regarding a clinical outcome compared to the health care proxy, as well as personal beliefs which can influence EOL decisions.106 Withholding and withdrawing life support is commonly practiced in ICUs. Common reasons for this are futility, patient suffering, and the anticipation of a poor quality of life. In order to investigate the potential bias of age in withholding or withdrawing support, Plaiser et al studied this practice in their ICU. Despite a small patient population, this single institution study was unable to show a difference between young and old regarding these EOL decisions.90
Futility is a concept with a distinct ethical definition. Unfortunately, the term is applied frequently in situations that are strongly influenced by the local environment and culture. What may be viewed as futile by one physician may not be seen the same way by another. As such, utilization of an ethics consultation will be important. The perception of suffering of elderly trauma patients is a cause of emotional distress for the ICU staff.107 The ubiquitous use of analgesics and sedatives in the ICU is necessary, but the science is imprecise. Most surgical ICU patients who survive indicate that they would repeat the experience again if necessary.108 EOL decisions based upon this argument are not indicated. Using the argument of a poor quality of life is commonly based upon the perceptions, beliefs, and personal interpretation of quality of life of the health care providers. Care must be taken to not interject one’s own beliefs when counseling families. Most ICU survivors indicate that they have an acceptable quality of life and would undergo treatment again.109 Data on the quality of life following ICU treatment after injury is not widely available. Limiting care in the elderly trauma patient in whom survival is unlikely is a necessary and ethical aspect of care provided by trauma practitioners. Utilizing standard ethical concepts and precise diagnoses will help facilitate discussions with families. Palliative care consultations have been used successfully with trauma patients, which can improve family and staff satisfaction with end-of-life care.110