More recent studies have yielded varying results (150,151), certainly not confirmatory of a pronounced “predestined” decline in oxygenation with age, and questioning the hypothesis that progressive disruption of the matching of ventilation and perfusion in the elderly is actually the cause of whatever decline actually occurs (152). The related questions of rise in (A–a) gradient with age, and “normal” age-related decline in PaO2 are, similarly, quantified by different investigators (148,149,153,154).
Malnutrition, also known as undernutrition, is a common companion of elderly individuals and frequently a complicating factor in the efficient and successful management of an elderly ICU patient. The natural decline in energy expenditure with age begins at about age 30 and accompanies the age-related increase in body fat-to-protein ratio (155–157). The evolution of nutritional intake with age is one of decline that exceeds the decrease in energy expenditure (158,159) for various reasons. Thus, even the healthy individual will eat less and lose weight with time, and will be at risk for malnutrition if illness occurs or social support wanes. For example, in elderly nursing home patients, a population in whom initially minor medical problems can quickly blossom into life-threatening conditions, the incidence of undernutrition can approach 85% (160,161). Malnutrition at the onset of critical illness portends poor outcome, as does insufficient nutritional support during the course of the illness (162,163). Mortality is considerably higher in the malnourished elderly patient, compared to those who are nutritionally replete (162). Undernutrition has several common causes: (i) functional decline and social isolation from family and other support systems, (ii) anorexia associated with older age—the so-called anorexia of aging—or chronic illness, (iii) anatomic or gustatory impediments to mastication or swallowing, (iv) abuse or neglect, and (v) insufficient financial resources (164–167). Therefore, the prevalence of undernutrition in hospitalized, geriatric patients is relatively high (168,169) and is often unrecognized unless sought specifically (170). Identification of malnutrition in the elderly patient (170) may be facilitated by the routine employment of easily used physical examination and laboratory screening tools as part of an organized, proactive nutrition screening program (171). There is little literature addressing nutrition in the geriatric ICU patient per se, and the principles set forth below are generally applicable to any ill elderly patient.
Undernutrition imposes a considerable burden on the marginally compensated geriatric patient. The conditions known as protein–energy malnutrition (PEM) and micronutrient deficiency complicate the treatment of several conditions seen in the ICU. These include the contribution of gastrointestinal (GI) tract nonintegrity to multiorgan system failure (172,173), and other common CV (174,175), pulmonary (176–178), and infectious issues (179). Wound healing is impeded by a poor nutritional state (180–182); in particular, development of decubitus ulcers is more common in malnourished elderly individuals, and successful management is decidedly more difficult (181). Patients with PEM are at increased risk for serious complications while in the hospital (182), with slower recovery (183), poorer functional status at discharge, and higher rates of mortality after discharge (184–187).
Malnutrition is a disorder of body composition in which macronutrient and/or micronutrient deficiencies occur when nutrient intake is insufficient, resulting in reduced organ function, abnormal blood chemistry studies, and suboptimal clinical outcomes (188). Nutritional deficiency is found in 35% to 65% of elderly hospitalized patients (189). Of the available screening techniques reflecting nutritional status, one of the most revealing is the dietary and weight loss history, as found in such structured nutritional questionnaires as the Mini Nutritional Assessment (MNA) and other tools (190,191) and nutrition evaluation steps taken on admission (192). Although probably more applicable to the long-term outpatient setting, certain pieces of information gathered from the patient or family via the MNA are helpful in providing a “snapshot” of the patient’s nutritional status as the initial steps in the continuing attention that must be paid to nutritional integrity (193). Obtaining the patient’s weight immediately on admission is an obvious step in assessing nutritional condition. Because of the various types of body habitus found in ICU patients, a calculation of the Quetelet body mass index (BMI) is helpful to standardize weight to height, providing a relatively standardized estimate of body fat (194):
BMI = weight (in kg) divided by height (in m2)
The Department of Health and Human Services defines normal BMI as being within the range of 18 to 24.9, with those with BMIs less than 18 being underweight, the overweight range being 25 to 29.5, and those displaying a BMI above 30 being obese (195). These data, however, cover—in the United States—the adult population as a whole. The picture in a unique subset such as the elderly is more complex. In the geriatric population, BMI less than 20 is predictive of nearly 50% 1-year mortality (196), a stronger predictor of mortality than is diagnosis; similar results were found among critically ill adults with a BMI less than or equal to the 15th percentile (197). Such data lead researchers to suggest that the optimal BMI lies higher in the elderly than in the general population (198); this supposition has been supported by a large study demonstrating that the detrimental effect on mortality of excess body weight declines with age (199). Furthermore, the BMI calculation does not differentiate between differences in body morphology; obese, malnourished individuals whose BMIs fall within the normal range may go unidentified using this formula (200). Because an age-associated loss of height can be significant in the geriatric population, especially in kyphotic individuals, substitution of arm span as the denominator of the BMI calculation has been suggested to give a more accurate comparison of an individual patient’s BMI to the standards that were originally established in younger persons (201,202). Arm span is identical to height in younger years; although height may decline with age, arm span remains unchanged, providing more accuracy within the previously determined younger age frame of BMI reference. Knee height, as measured from plantar surface to top of patella with the ankle at 90 degrees, is another measurement (203) that can be substituted in corrected BMI calculations in those with diminished stature who are unsuitable for arm span measurement. Triceps skin fold thickness and mid-arm circumference can also provide an idea of body fat content (204).
In general, however, use of the BMI in the elderly is suspect, regardless of the height measurement used, as there are few normal BMI data that specifically describe those older than 65 years. The ages of geriatric patients included in nutrition studies vary, anthropometric characteristics vary in different advanced decades, and incidence of weight-changing diseases and conditions—such as cancer or the anorexia of aging—increases with age (205). These factors make the formulation of accurate statements and recommendations addressing ideal weight and BMI in the elderly difficult to formulate (156). Although the percentage of older Americans falling into the definition of obese continues to climb (206), one should not make the assumption of nutritional integrity. Age-related redistribution of caloric stores may disguise the overweight elderly patient with severe PEM (200) as one who is obese in the mind of the unwary clinician who is not familiar with the metabolism of geriatric patients and the pathophysiologic implications of these changes (207). Misguided hypocaloric feeding, directed at mobilizing excess fat stores in the obese, but malnourished, elderly patient may worsen the situation by leaving the ongoing catabolic protein breakdown associated with critical illness uncorrected (208). Several easily measured laboratory parameters are reflective of nutritional status on admission, and some can be followed periodically to assess the success of nutritional support. Albumin is a product of hepatic metabolism, synthesized ultimately from ingested or infused nitrogenous precursors in the presence of adequate caloric support. Although it is held that the serum albumin level is reflective of the nutritional state, various factors influencing serum albumin levels make it only vaguely reflective of overall nutritional status (209), with an ROC (receiver operating characteristic) curve rating of 0.58 compared to the clinical subjective global assessment tool. Serum albumin level does decline somewhat with age—0.8 g/L per decade for individuals older than 60 years of age—but generally remains within the numerical normal range. Significantly reduced albumin concentration, therefore, should be attributed to disease processes (210,211) and be aggressively investigated. A substantial decline in serum albumin concentration is accurately predictive of mortality and worse outcome among the elderly, both in the setting of apparent health and illness (212–216), possibly reflective of the presence of chronic disease- or inflammation-induced mediators that simultaneously suppress albumin gene expression (217). The half-life of albumin, 18 to 19 days (218–220), makes its use less than optimal in monitoring metabolic and synthetic functions, in which rapid change is significant. The reliability of previously favorite nutritional indicators such as prealbumin and retinol-binding protein in demonstrating the adequacy of nutritional support in critically ill patients has recently been called into question (221).
As critical illness induces substantial catabolism (220,222,223), resting energy expenditure (REE) rises during the first 2 weeks of this state, with mobilization of nitrogen stores as a component of the associated inflammatory response to physiologic insult. Total energy expenditure (TEE) may rise dramatically in critically ill, septic, or trauma patients, repletion of which is most difficult without correcting the underlying inciting process (224,225). In the elderly individual with marginal nutritional reserve at the onset of critical illness, early provision of caloric and protein support is warranted. Catabolic processes characteristic of critical illness are not reversible by nutrient supplementation alone; they are incited by inflammatory mediators rather than by pre-existing deficiency or inadequate repletion and are thus not forestalled by aggressive nutritional support. Traditional guidance recommends 25 kcal/kg/day of nutritional support, with an additional protein supply of 1.2 to 1.5 g/kg/day (188,226) based on actual body weight. Obese individuals, defined as above (227,228), warrant feeding based on ideal, rather than actual, body weight (IBW):
Men: IBW (kg) = 50 + 2.3 kg per inch over 5 ft
Women: IBW (kg) = 45.5 + 2.3 kg per inch over 5 ft
Greater accuracy can be achieved using one of several formulas to calculate REE (229) the Harris–Benedict equation is commonly used (230):
This may be insufficient in the critically ill geriatric patient in the throes of the inflammatory response, unless the higher stress and activity factor is used (231). Resting metabolic rate may be nearly double in the critically ill or injured individual (224) compared to the healthy uninjured person. Protein supplementation for the most critically ill ranges from 1.2–1.5 g/kg to 2.0–2.5 g/kg (232–234) although, in the initial stages of such a condition, the rate of catabolism may just not be ameliorable despite aggressive support in appropriate quantities (235). Initial empiric dosages should subsequently be adjusted based on indirect calorimetry and nitrogen balance studies if there is suspicion of inadequate nutritional support (236–240). Enthusiastic overprovision of macronutrients in a misguided and vain attempt to thwart and correct inflammatory catabolism, on the other hand, leads to a host of complications and considerable morbidity (241) for which the geriatric patient may be unable to compensate. Most recently, it appears that the optimal outcome attributable to nutritional support is achieved only when both the energy (as determined by indirect calorimetry) and protein requirements are reached within 1 to 2 days (242); the wisdom of “permissive underfeeding” of predicted energy requirements has recently also been documented (243). The confounding factor of obesity sometimes seen in the nutritionally deficient geriatric patient makes the recipe that provides optimal nutritional support frustratingly difficult to determine. In such situations, measurements of energy expenditure performed at frequent intervals are even more strongly advisable, because energy requirements fluctuate with time and medical condition, and vary significantly from those of younger patients on whose metabolism nutritional recommendations are often based. In general, most, although not all, studies show that enteral nutrition is preferred because of the purported preservative effects on intestinal mucosal integrity, cost issues, and a lesser degree of risk exposure to the patient, both infectious and mechanical, associated with placement of flexible nasointestinal feeding tube versus central line for parenteral nutrition (188,244–249). This statement, however, is the source of endless controversy and the basis of considerable investigation (249–251). The optimal site of delivery of the enteral solution, gastric versus postpyloric, surprisingly, remains controversial (252,253) as does the importance of gastric residual volume (GRV) measurement and its impact on outcome (254,255). The risks and benefits of the common routes of nutritional support have been reviewed in considerable detail (256,257).
One additional point that is critical to remember is that of the possibility of development of potentially fatal refeeding syndrome in a critically ill patient who is already nutritionally deficient, as many elderly people are. Meticulous attention to the introduction of nutrients in those at risk, with frequent monitoring and generous replenishment of electrolytes, is very important to avoid this morbid complication (258).
Deterioration of renal function in a critically ill patient has a dramatic impact on survival. Despite this, it is often not recognized in a timely manner, is managed poorly or incompletely, and is often preventable or able to be ameliorated by suitable intervention (259). Acute kidney injury (AKI, the name given to what was previously known as acute renal failure [ARF]) carries a mortality of nearly 30% in a general ICU population; a decline of renal function of even lesser severity also impacts mortality significantly (260), more so in the geriatric population. An elderly patient with compromised renal function will often succumb to the added insult of renal failure after a complex surgical intervention or traumatic injury. The chance for at least partial renal functional recovery after critical illness-related AKI is greater than 90% among those alive a year after their illness (261). Presently there is little available for treatment of renal insufficiency or failure other than identification of the etiology with certainly so as to administer the correct theraputic medications when warranted (e.g. steroids, imuran, etc.), optimization of hemodynamics, prevention of further renal damage by removal of any nephrotoxic medications or processes, aggressive management of complications such as hyperkalemia, and initiation of renal replacement therapy if indicated and deemed appropriate within the patient’s wishes. The intensivist holds a pivotal role in the understanding of renal physiology and the principles of renal protection to minimize the impact of critical illness on renal function and its influence on outcome.
Just as in other physiologic systems, there is a gradual deterioration of renal function with age, beginning at age 30 years (262,263). It is well described that renal blood flow declines after the fourth decade (264,265). When the sixth decade is reached, this deterioration generally continues, although with a very wide bell curve of distribution (266). There is loss of renal—primarily cortical—mass (267) and the onset of glomerular sclerosis and involution, causing a decrease in the number of functional glomeruli (268), in turn causing a decrease in glomerular filtration rate (GFR) of 30% to 40% by the age of 80 years (262,269). Deterioration of tubular function parallels that of the glomerulus (270). In the elderly patient, factors other than age-related deterioration may complicate renal function, including pre-existing renovascular disease, hypertensive nephrosclerosis, or hypotension associated with trauma or neglect. Laboratory measurement of serum blood urea nitrogen (BUN) and creatinine (Scr), used individually or in a ratio, act as surrogates of renal function; they are, however, less accurately reflective of renal function in the elderly than in a younger person. BUN rises slightly with age over 60 years, paralleling the gradual decline in renal function; Scr reflects muscle mass and, while completely filtered (and only minimally secreted) into the tubule and therefore generally reflective of GFR, may not climb as expected despite age-related falling renal filtration (271). The age-related muscle mass diminution, frequently paralleling deterioration of renal function, generates less creatine (and thus, creatinine), leading to what may erroneously be looked upon as a normal baseline Scr. Assessment of GFR should be individualized by using the Cockcroft–Gault formula (272) to generate a more accurate estimate of function based on weight, age, and serum creatinine:
Creatinine clearance = [(140 – age) (weight in kg)]/(72 × Scr)
(arithmetic result × 0.85 = clearance for female patients)
This formula provides a “snapshot” of function at a given time and is most useful if calculated on ICU arrival and daily thereafter. Other laboratory surrogates of GFR have been devised, such as the measurement of cystatin C (273–275) and MDRD (modification of diet in renal disease) equations (276), but the ease with which the Cockcroft–Gault calculation is performed, especially when performed daily to “trend” the result rather than depend on one individual number, makes its routine replacement unlikely. The CKD-EPI formula (277) has been found to be more accurate in some circumstances. One must be mindful that any assessment of renal function utilizing Scr to infer creatinine clearance as reflective of GFR is limited by the nonlinear rise in Scr as renal function declines (278,279) in that a small change in a close-to-normal value Scr represents an insignificant change in GFR, while a similar numerical change in an already-elevated Scr likely represents further compromise of already impaired renal function. If GFR remains uncertain, urine collection for measurement of creatinine clearance can be done with fair accuracy using at least an 8-hour urine collection period (280,281); 24-hour collection is preferred in critically ill patients and is easily done in patients with indwelling urinary catheters. A variety of methods exist to assess renal function, both by measurement and by estimation (282,283).
Fluid and electrolyte handling is altered in the aging kidney, related to tubular dysfunction which is proportional to the GFR decline. Although baseline electrolyte values and fluid status are likely within normal range in the previously healthy geriatric patient, age-related tubular dysfunction narrows the limits of correction of water and sodium aberrations that the elderly kidney can readily accomplish. Sodium excretion and reabsorption declines in efficiency, with those older than 60 years requiring considerably more time to achieve homeostasis in the face of sodium overload or deprivation (284). Similarly, the range of specific gravity and osmolarity achievable in the face of water excess or deficit is narrowed in comparison to that of a younger individual (285); rectification of acid–base perturbations is similarly deficient (286). The stresses of critical illness or injury typical of the elderly ICU patient intensify the effects of these functional deficiencies, and must be foreseen and addressed aggressively to forestall the profound effects of deterioration of renal function on morbidity and mortality. These stresses include volume depletion from GI bleeding, severe dehydration, diarrhea, aggressive diuresis, insensible losses in burn patients or those with drainage from wounds or fistulas, and disruption of renal blood flow from sepsis, shock, or surgical causes such as complex renovascular surgery. Management of deteriorating renal function requires accurate diagnosis of the inciting cause, while addressing complicating or resultant metabolic derangements and preventing further insult. The details of the diagnosis of renal pathology are not specific to the geriatric patient and are addressed elsewhere in this text (see Chapter 132).
It is important to recognize that AKI occurs in as many as 67% of ICU admissions (260), as identified by RIFLE (risk, injury, failure, loss, end-stage) criteria (287), and that the effect of renal deterioration is quite detrimental to the elderly individual. Initial evaluation must include performance of a physical examination that may reveal an occluded urinary catheter causing an enlarged bladder; bladder scanning can be performed on those in whom an enlarged bladder might not be palpable. Hypovolemia, both absolute, as in severe dehydration, and relative, as in sepsis, must be aggressively corrected with appropriate fluid and blood products; invasive monitoring is warranted in this population of patients with limited reserve. Dosage adjustment of potentially nephrotoxic medications is mandatory, using assessment of GFR as a guide. Antimicrobials such as cephalosporins and aminoglycosides, nonsteroidal anti-inflammatory medications, certain chemotherapeutic medications, and angiotensin-converting enzyme inhibitors are common offenders (288). The use of “protective” medications such as N-acetylcysteine, dopamine, mannitol, or loop diuretics to minimize the detrimental impact of contrast material on renal function has generally been demonstrated to be ineffective (289,290). On the other hand, pre-procedure isotonic fluid loading, mindful of the possibility of occult HF in elderly patients, is the strategy most likely to benefit postcontrast renal function (291). The use of isotonic bicarbonate solution, while possibly decreasing the incidence of CIN compared to isotonic saline, does not decrease the subsequent incidence of dialysis or in-hospital mortality (292,293).
Beyond awareness of medications that impact renal function, there is the effect of age-related diminished renal function on drug metabolism and excretion (294). Recall again that common indicators of renal function, BUN and Scr, although appearing normal in the elderly, may mask a compromised GFR, risking medication-induced complications if this fact is overlooked, and mandating more specific assessment of GFR (see above) if question arises. Early nephrology consultation is encouraged when RIFLE criteria suggest compromised renal function; similarly, a critical care pharmacologist can assist in clarifying renal-active medication issues in these complex patients.
ASSESSMENT AND MANAGEMENT OF TRAUMATIC INJURIES
Elderly individuals suffer a significant number of severe and often lethal traumatic injuries, the analysis and management of which can be frustratingly complex (295). In the 55- to 64-year-old age group, unintentional injury was the THIRD leading cause of death in 2013; in those older than 65 years of age, 45,942 deaths were attributed to trauma (296) compared with 35,000 in 2003 (297). Most serious injuries are caused by falls, the occurrence rising dramatically as age advances into the 60s and beyond (298). This predominance continued to be seen through 2015 (299). Falls from a standing or even sitting position, imparting an apparently trivial amount of kinetic energy to frail tissue, may result in fatal injury, accounting for half the trauma-related deaths as compared to those of younger people (300). Most remaining significant traumatic injuries to the elderly involve motor vehicles, either as vehicle occupants or as pedestrians (301), while there is a small but persistent incidence of injury and death from penetrating trauma in the geriatric population, declining to less than 1% in those older than 75 years (299,302,303).
Evaluation and management of the injured elderly requires familiarity with characteristic injury patterns and knowledge of comorbid diseases and particulars of geriatric physiology that impact treatment (304). Practice management guidelines for geriatric trauma (305) are helpful in this situation. Triage of injured geriatric patients to more experienced trauma centers improves outcomes, to such an extent that some practitioners in the field of trauma management advocate the hyperspecialization of some centers to be the location of management of seriously injured elders (306).
Immediate assessment of the resuscitation status of any patient arriving in the ICU, whether from the operating room, emergency department, or elsewhere in the hospital, is imperative. The paucity of overt physical findings of intravascular fluid deficiency seen in the elderly patient adds additional urgency to its accurate analysis, while the lusitropic compromise that typifies the geriatric patient mandates avoidance of overgenerous fluid repletion. Although standard protocols may serve as a guide to ensure that all systems are evaluated, one must remain mindful that standard and acceptable initial hemodynamic measurements may actually conceal unsuspected injury or bleeding in the confused elderly trauma patient who may be taking medications that affect vital signs. Airway management in the elderly carries its own set of difficulties. Age-associated arthritic spinal, mandibular, and arytenoid deformities, and an increased incidence of occult cervical spinal injuries (307–309) may be seen; marginal respiratory drive and compromised airway reflexes may warrant securing the airway preemptively, avoiding a later “crash” difficult airway emergency. A thorough and detailed physical examination is fundamental. Timely sequential measurement of routine hematology tests, even in the stable geriatric patient, may reveal unsuspected hemorrhage. Arterial blood gas analysis is a convenient tool because it is quickly performed and allows frequent measurement of hemoglobin, base deficit, and lactate. The latter two values are powerful indicators of resuscitation status and, when elevated, predict increased mortality in the elderly population (310,311). In one study of elderly trauma patients, mortality was decreased from 54% to 34% (p < 0.003) by institution of a protocol of trauma team activation and early noninvasive and subsequently invasive monitoring for resuscitation of all patients older than age 70 years with an injury severity score (ISS) greater than 15, even for those with nonworrisome initial vital signs and fairly minor injuries (311). This supports the precept that achieving adequate tissue perfusion early, while often difficult to accomplish, is fundamental to successful trauma management. One must be mindful, furthermore, that while invasive monitoring carries its own risks, judicious use of these tools can improve outcome and survival in the elderly trauma patient (312–314).
Certain patterns of injury are found in geriatric patients. Traumatic brain injury (TBI) afflicts the elderly with extraordinary severity. High mortality leaves fewer survivors, most of whom suffer debilitating sequelae (315). A large meta-analysis revealed an overall mortality of 38.3% in patients 60 years and older with moderate and severe TBI (316). In 2003, there were 90,000 emergency department visits involving TBI in those older than 65 years, of whom 38.4% died (317); some series document mortality rates for severe TBI in those older than 55 years of age as high as 80% (318). Initial neurologic examination of an elder with significant intracranial injury may be deceptively normal (319); the reliability of the Glasgow Coma Scale in identifying elders with severe TBI is not as great as in younger individuals (320). A high index of suspicion for the presence of an occult central nervous system (CNS) injury must be maintained if such individuals arrive in the ICU without radiologic evaluation having been performed, warranting frequent, sequential neurologic evaluations by the same examiner and a conservative approach to ordering a cranial CT scan. Those elderly whose cause of TBI is a fall—nearly 50%, from 1988 to 1998—are likely to have three or more significant comorbid conditions complicating ICU management (321,322). Outcome after TBI is optimized by using meticulous clinical assessment, timely radiologic re-evaluation, and aggressive invasive monitoring to facilitate immediate recognition of worsening status, such as that due to recurrent intracranial hemorrhage, while minimizing secondary injury. Elderly patients taking anticoagulant medications and antiplatelet (ACAP) agents experience higher TBI-related in-hospitality mortality (323). Secondary injury may occur when even transient episodes of hypoxia or hypotension affect cerebral perfusion pressure (CPP) (324) and, in the setting of elevated intracranial pressure (325), with hyperglycemia (326), hyperthermia (327), or aggressive hyperventilation (328,329). Infusion of hypertonic saline (330) may supplement the management of elevated intracranial pressure that resists control by the usual initial measures. The cornerstone of TBI treatment is the maintenance of cerebral oxygenation by ensuring adequate oxygen content and CPP, guided by data derived via invasive intracranial monitoring devices that are inserted based on specific indications (331). Little, however, has been written specifically addressing geriatric CPP requirements. Although ∼60 mmHg is considered the threshold below which the CPP should not be allowed to drop (332), this has not been rigorously studied specifically in the geriatric population. Cerebral autoregulation in the elderly is subject to the same influences as those that affect the younger individual. In this population, the abundance of comorbidities, such as untreated hypertension, may have acclimated the cerebral vasculature to a new baseline, making invasive cerebral monitoring even more critical in ensuring adequate perfusion for the aging brain. The profound influence of even mild TBI (316,332–335) on short- and long-term outcome in the elderly patient mandates aggressive monitoring, optimization of cerebral perfusion, and meticulous attention to hemodynamic parameters. The impact on long-term survival of severe head-injured elderly trauma patients, compared to other multi-trauma patients without TBI is considerable (336).
Cervical spine injury is common in the geriatric trauma patient (337); plain radiographs (307,338) may be unrevealing of fracture or difficult to interpret because of age-related boney changes obscuring acute pathology (308). Fracture of the upper cervical spine is more common in the elderly than in younger individuals, especially in those who have fallen (309), and is more likely to be unstable (339). There is a very high incidence of odontoid fracture in the very elderly, with significant associated mortality (340,341). Helical CT is superior to plain radiographs to identify cervical spine injury in this population (338,339,342,343). Cervical spine pathology may exist in totally asymptomatic individuals with unremarkable examination findings, only to be discovered by a diligent clinician who takes extra steps to search for such an injury (344,345). In the geriatric patient with a cervical injury, the likelihood of coincident painful injury—a distracting injury in which the pain from another injury distracts the patient’s attention from the perception of neck pain or a condition such as altered mental status—that would affect the examiner’s decision to forgo radiographic evaluation is so high as to make such an evaluation imperative in nearly all cases. Again, one must be mindful of the greater likelihood that cervical injuries in the elderly often occur in the arthritis-prone superior vertebrae, which are notoriously difficult to depict on plain films (346), and consider CT evaluation of virtually all geriatric trauma patients in whom even subtle symptoms, history, or mechanism of injury suggest cervical injury, regardless of the initial examination findings or any comforting results of a protocol-based decision-assisting algorithm that suggest the safety of less aggressive investigation.
Traumatic rib fractures impose substantial morbidity; those older than 45 years with more than four fractures are particularly affected (347). In a study of patients traditionally defined as elderly—those older than 64 years—rib fractures profoundly affected morbidity and mortality, with longer length of stay (LOS) in the ICU, more frequent pneumonia, and overall mortality rate of 22% compared to 10% (p < 0.001) in those less seriously injured (348). Of note in this study was that rates of mortality and pneumonia both increased with each additional rib fracture. Epidural analgesia would appear to be the ideal technique to alleviate the pain associated with rib fractures to optimize pulmonary status and, indeed, has been found to be successful in nongeriatric adults (349,350). In one recent study, however, the opposite has been demonstrated in an elderly population when compared to parenteral analgesia (351).
The management of abdominal trauma follows pathways similar to those for younger patients, with certain caveats: findings on physical examination indicating serious abdominal pathology can be subtle, especially when complicated by distracting orthopedic or mild head injury. Liberal use of CT scanning is strongly recommended if mechanism of injury, external abdominal findings such as a seat belt mark, or laboratory evidence of hypoperfusion (elevated base deficit or serum lactate) suggest visceral injury. Nonoperative management of certain radiologically well-characterized injuries of solid organs—namely the spleen, the liver, and the pancreas—in the hemodynamically stable elderly patient is becoming increasingly accepted as evidence of the success of this approach accumulates (352–354).
Serious orthopedic injuries frequently befall older victims of polytrauma, and portend a substantial risk of mortality (355). Decrease in bone mineral density (BMD) in patients older than about 30 years heightens the risk of fracture in general; this phenomenon is observed in varying degrees in both genders and all races, but is particularly severe in postmenopausal Caucasian women (356,357). Pelvic fracture in the aged is associated with a greater likelihood of significant blood volume transfusion and mortality (p < 0.005) (358). Open pelvic fracture often has substantial associated bleeding, which is seldom treatable, with the exception of arterial bleeding, in any way other than with early stabilization, aggressive transfusion, and correction of coagulopathy in hopes of eventual tamponade of the retroperitoneal bleeding source. Arterial bleeding from lacerated pelvic vessels warrants embolization (359). The more typical scenario, however, is that of diffuse venous oozing, which, nonetheless, may render the elderly patient hemodynamically unstable, requiring large-volume transfusion of blood products as a temporizing measure until anatomic stabilization can be achieved (359,360). The presence of an open pelvic fracture, with frequent associated visceral injuries (361,362), further worsens outcome (358). Hemodynamic consequences of large-volume transfusion and frequent septic complications can drive the mortality in both younger and older adults to nearly 80% (363). Long-bone fractures, in general, warrant early immobilization and stabilization to minimize ongoing hemorrhage and generation of fat emboli; such fixation improves mortality significantly (364). Optimal timing of surgical stabilization of these quite morbid fractures, however, is a complex issue to resolve when they occur in the larger setting of the patient with severe head, chest, or abdominal injuries (365). Although postponing the operative stabilization of a femur or complex pelvic fracture to allow time to achieve hemodynamic stability in a traumatized patient has benefit, it is also not without risks (366,367). Prolonged immobilization of the elderly patient with such a fracture prior to stabilization results in compromised respiratory status, likely exposing the patient to extended intubation, pulmonary thromboembolism, and infection.
Studies of elderly trauma patients have consistently documented the increase in mortality in this population (302,313,367,368). The mortality rate begins to climb for those in their sixth decade, even for less severe injuries (369) when compared to younger individuals. For those with moderate injuries, the mortality curves steepen beginning in the fifth decade, with another even steeper turn in the seventh. With advancing age, trauma-related mortality rates for those in the seventh decade and above range as high as 47% for those with an ISS more than 30, compared to those 45 years old or younger (20.1%) (370). Identification of parameters which might predict mortality in the individual patient is an important area of study (371). Within the context to which allusion was earlier made—that of future payoff in return for resource use—the complexity and enormity of the issue grows as health care costs rise, and as the percentage of the population represented by the elderly increases. As these rising numbers of individuals cease working, and, thereby, are no longer able to generate an income that can be taxed to finance public health care funding programs such as Medicare, or be used to pay for personal private health insurance to cover costs of traumatic injuries, the costs of providing that trauma—and, indeed, all—care will have to be borne by a source other than the patients themselves. It is clear that focusing the resources of the modern ICU on the management of elderly trauma patients improves outcome (372), so the skills associated with successful management of elderly trauma patients are improving with experience. Based on the size of these costs and the likelihood of marginally or poorly acceptable outcomes among a substantial minority of geriatric trauma patients (3,370,373–375), investigations have tried to answer two important trauma outcome-related questions. These are as follows: (1) is it possible to identify an elderly trauma patient who will certainly die later, even if the patient survives the initial period of resuscitation, surgery, and further stabilization, and (2) to what level of functioning will the elderly survivor of trauma-related intensive care return on discharge? For many elderly individuals, the prospect of lingering in the netherworld of prolonged posttrauma multi-organ system failure with the certainty of death pushed back “only as long as the machines keep me running,” or existing debilitated in a non-home environment where even bowel function and bathing are at the behest of another, is worse than death itself, not really living at all, and is the basis of much concern among the elderly.
There are, however, grounds for hope. In one study of victims of penetrating trauma more than 60 years old, 91% were discharged home, most without assistance (303). The postdischarge level of functioning in elderly patients surviving blunt trauma varies widely, as would be expected in a population whose baseline physiologic attributes are so diverse. Clearly, even the healthiest octogenarian is not the physiologic equal of a two-sets-of-tennis 65-year-old and will have a significantly decreased likelihood of returning to premorbid functionality, although both individuals may be described as elderly. Nonetheless, even after significant traumatic injuries, a substantial percentage of recovering geriatric patients, even the very old, will be able to live relatively independently, albeit for some patients, in a protected environment with assistance. Many will be able to return home with or without periodic professional assistance (373,376–378). In one retrospective study of 38,707 elderly trauma patients with a mean ISS of 11.7 ± 0.05 (standard error of mean), in which 10.3% died in hospital, 52.2% of the survivors went home. The percentage of patients returning home after serious traumatic injuries, many requiring prolonged intensive care, varied considerably with age, from 66.7% of those 65 to 74 years to 30.5% of those 85 years of age or older (378). With aggressive rehabilitation, improvement in function and independence can continue for substantial periods of time after discharge, including in those who have suffered TBI (379,380). In another study, recovery of elderly trauma patients was improved by early involvement of physicians from a geriatric trauma consult service, who assisted in recognition and treatment of medical issues, and in advanced care and disposition planning (381). Additionally, determining the optimal destination for posthospitalization rehabilitation can be facilitated by employment of assessment tools that include such parameters as the 15-item Trauma Specific Frailty Index, the Barthel Index, and others (382). The likelihood of leaving the hospital after a trauma-related ICU admission can be improved from the outset, as noted earlier, by aggressive attention to adequate resuscitation to rectify suboptimal perfusion, by attention to maintenance of acceptable CPP in TBI, and by recognition and treatment of the early subtle signs of cardiac and respiratory decompensation. Finally, trauma care outcome must be scrutinized within the context of profound personal and social issues, beyond those solely of medical success, that are integral to ICU care in patients in this age group.
OUTCOME AFTER A CRITICAL ILLNESS IN THE GERIATRIC POPULATION
Life expectancy in the United States is presently 78.8 years (383), and while it is greater at any given age now than it was even 15 years ago, objective evaluation must be made of the appropriateness—and likelihood of successful outcome—of aggressive critical care medical services provided to geriatric patients. Presently, geriatric patients represent between 25% and 50% of all ICU admissions (9,11). In 2000, ICU costs represented 13.3% of hospital costs, 4.2% of health care expenditures, and 0.56% of the US gross domestic product (384). By 2005, the latter value had increased to 0.66%, with Critical Care expenditures representing 13.4% (∼ $82 billion) of hospital expenditures (385). As of this writing in 2016, expenditure for CCM services represents approximately 1% of the US GDP. The enormous expense associated with ICU care has prompted some analysts to raise the subject of limits on expenditures for the elderly (386,387), because, for example, an 80-year-old who is supported through a 3-week bout of sepsis is not likely to return to the revenue-generating work force. Indeed, the literature dealing with geriatric medical issues is liberally populated with articles addressing ageism in the context of delivery of services to the elderly (388–390), raising the concept of providing less aggressive or intensive levels of acute care to an elderly person on the basis of age, the inference being that such care provides a less robust postillness benefit to the patient and to society as a whole. Meaningful discussions addressing the more philosophical issues of critical care such as the correct level of aggressiveness of care and appropriateness of withdrawal of care, to say nothing of the financial issues, simply cannot be addressed in any rational way without an accurate picture of what critical care accomplishes in these elderly patients.
A successful ICU admission is certainly defined within cultural and social, as well as personal, contexts. Although the family member’s “do everything for Granddad” dictum is familiar, it often represents an unrealistic appraisal of the possible benefit from certain modalities of care that can be, but possibly should not be, carried out. Although the ICU is designed as a temporary environment that allows support of body functions during recovery, the complicated technology and meticulous attention to detail that characterize that environment are not the basis for such “magical” accomplishments as saving the life of a patient who has a lethal condition, despite the expectations and exhortations of some. Indeed, death can often skillfully be forestalled with polished and professional ICU care to such a degree that it may occur immediately after a de-escalation of such care, or later while the patient is on the general ward, in a step-down unit or rehabilitation facility, or after returning home (either early or late) (391). Meaningful discussions with elderly patients and their families, whether prior to complex morbid surgical procedures or as an ICU stay extends past the first few days, must include accurate outcome data, so as to facilitate informed decisions regarding the specifics and suitability of continued care. Studies addressing outcome in the critically ill geriatric population have produced various results that vary with the metric employed, the duration over which the outcome is monitored and broad intrapopulation patient variability. The latter category highlights differences in age, premorbid physical status, statements of preference regarding aggressiveness of long-term medical care, and patient and family declarations addressing such subjective concepts as posthospitalization quality of life (QOL). As the numbers of geriatric patients admitted to ICUs increased with the growing geriatric population, some more meaningful data identifying which elderly patients are likely to survive and return to meaningful posthospitalization lives is becoming available (392). In one recent study from Canada, 25% of elderly ICU patients 80 years of age and older survived and had returned to their premorbid levels of functioning by 1 year later (393).
The term geriatric population encompasses a quite heterogeneous group of individuals from the standpoint of age, premorbid general medical health as a reflection of functional status, the severity of the event justifying ICU admission, and cultural mores as they impact interaction with the modern health care structure of the country in question. Studies assessing the results of care delivered to the elderly may or may not reflect this diversity (394), making interpretation of individual study conclusions and their application to individual clinical situations suspect. Furthermore, the term outcome must be specifically defined as to the depth of support required by the post-ICU elder and its correspondence with that autonomous person’s preferences which, again, may vary widely based on cultural, religious, national, and other parameters. Although many elders prefer a less aggressive care regimen designed around EOL comfort at the expense of duration of remaining life, some may desire life extension in the face of critical illness by use of complex technology despite a vanishingly small or nonexistent expectation of recovery (9,395,396). Furthermore, the clinician’s perceptions of the patient’s desires may not be accurate, and thus may lead to withdrawal of care or withholding of a modality of treatment in a manner that would not be considered in the care of a younger patient (395). It is important to remember that while age may be associated with worse outcome from critical illness, numerous investigations have demonstrated that age, in and of itself, is less a factor than is the severity of the specific condition that warrants intensive care or the general medical condition (i.e., frailty) (397) of the patient prior to the institution of intensive care (398–403). Despite being subjected to procedures that are potentially morbid, the otherwise healthy elderly patient may fare quite as well as a younger individual (400,404). In one study of outcome after intensive care in octogenarians, postdischarge survival was more accurately forecast by care dependency at the time of discharge, as a reflection of premorbid condition and severity of illness, rather than solely by LOS (405). The subjective term Quality of Life in the post-ICU elder does not necessarily imply inferiority to that of younger individuals (406); indeed, overall QOL has been demonstrated to be similarly good across age groups ranging from middle aged to very old (above 80 years) (407,408). It must be remembered, however, when evaluating outcome data in elderly ICU patients, that while ICU survival is less a function of age than of premorbid condition or severity of illness (36,405,406,409,410), when the aggressive ICU support is de-escalated with recovery, physiologic reserve may no longer suffice to forestall death in the few months after discharge, and thus may not be reflected in ICU outcome statistics. With the wide variability of desires for aggressiveness of care displayed by the elderly and the inaccuracy with which they are analyzed by many physicians (411), the most important function of the geriatric intensivist may be that of conducting a thorough discussion at the outset of care with the patient and involved family members so as to tailor intensiveness of care to the patient’s educated and informed preferences. Fluency in initiating and conducting EOL discussions with patient and family is important for intensivists to possess, and can be learned with clear guidance and polished with experience (412). Depending on the intensivist’s point of view and experience, some modalities of available ICU treatment may be viewed as “futile”—or possibly more objectively spoken as “medically inappropriate.” Although this may lead to strong differences of opinion regarding the depth of and extent to which intensive care plans are formulated and should be executed (413,414), it is incumbent upon the ICU practitioner to expend all possible effort to resolve the differences equitably and objectively; the assistance of Ethics Consultative services may be beneficial (415).
DRUG DOSING IN THE ELDERLY
As more patients live longer and are placed on a larger number of medications, it is necessary for health care providers to understand the risks, benefits, and consequences of drug therapy in older patients. Several important pharmacologic and nonpharmacologic issues influence the safety and effectiveness of drug therapy in this population. Pharmacokinetics, the study of the action of a drug in the body over a period of time, changes with age. The physiologic changes accompanying aging affect the pharmacologic processes of absorption, distribution, metabolism, and excretion (Table 70.1). The effects of these age-related changes are variable and difficult to predict; some changes are related solely to aging, whereas others are most likely due to the combined effects of age, disease, and the environment. Although increasing age is often accompanied by decreased physiologic reserve in many organ systems independent of the effects of disease, this change is not uniform. The alterations in pharmacokinetics and pharmacodynamics that occur with increasing age suggest a pharmacologic basis for concern about the vulnerability of the elderly to the effects of medications. Unfortunately, the results of epidemiologic studies that explore these relationships are unclear, in part due, in this area of medical investigation as in many others (416), to the small number of older people included in premarketing studies relative to the patient population most likely to be exposed to the drug. The oldest—those aged 80 years or older—have not generally been included in clinical trials of investigational drugs, and those older subjects who do participate in such trials tend to be healthy “young-old” people. Thus, the results of these trials and the side effects reported often have limited application to the older patient with multiple illnesses, taking several medications. In general, consideration of the individual patient, his or her physiologic status—hydration, nutrition, and cardiac output, and how this status affects the pharmacology of a particular drug—are more important in prescribing that drug than any specific age-related changes.
|TABLE 70.1 Age-Related Changes Relevant to Drug Pharmacology|