Geriatrics (Anesthesia Text)


Approximately 1/3 of surgical patients are 65 and older, although this number is rising. The incidence of perioperative morbidity and mortality is higher in geriatric patients, however this is most likely due to age-related diseases themselves and not the age itself (ie a healthy 70 year old may be at lower risk than a 55 year old with multiple comorbidities). It is helpful to think of elderly (and all) patients in terms of organ reserve function – as individuals age past their 30s or 40s, the “functional reserve” of each organ system begins to decline

Changes Associated with Aging


Brain mass declines with age, although cognitive function may remain intact. Postoperative delirium and cognitive dysfunction are higher in elderly patients. Note that it may be difficult to distinguish between changes associated with surgery and those associated with aging – in the past it was thought that CABG led to “pump head” ie accelerated cognitive decline, but a prospective, observational study comparing to CABG to medical management in 244 patients suggested that the cognitive decline following CABG is age-related and not due to the surgery itself [Selnes OA et al. Ann Neurol 63: 581, 2008]

General anesthetic requirements fall as patients age (as much as 30% according to some studies [Gold et al. Anesthesiology 79: 710, 1994; Nakajima et al. Br J Anaesth 70: 273, 1993]). The reasons are unknown, but may be secondary to the loss of neurons which accompanies aging (interestingly, cognitive decline does not seem to be related to neuronal loss). Another possible contributing factor is the increased threshold for stimulation in almost every sensory modality (vision, hearing, touch, proprioception, etc.)

Required doses of local anesthesia are reduced in spinals [Cameron et al. Anaesthesia 36: 318, 1981] as well as in epidurals [Sharrock NE. Anesthesiology 49: 425, 1978], although these changes, while real, may be clinically insignificant [Curatolo et al. Acta Anaesth Scand 38: 646, 1994]


As cardiac output is generally coupled with metabolic needs, aging patients have a reduced (but normal) cardiac output. A study of 106 patients 62-98 years of age (mean age 82) showed no change in mean resting LVEF even into the 10th decade of life (91-98 years of age, n = 9) [Aronow WS et al. Am J Cardiol 63: 368, 1989], however this study did not examine cardiac function under duress. A study of elderly vs. young volunteers subjected to postural stress suggested that they may not [Shannon et al. Am J Cardiol 67: 1110, 1991]

Systolic function can be remarkably preserved [Stoelting RK. Basics of Anesthesia, 5th ed. Elsevier (China) p. 519, 2007]. Cardiac responsiveness to adrenergic stimulation is reduced, however, and therefore maximal heart rate (and cardiac output) falls with age. Also, the heart becomes thicker with age, which impedes diastolic filling, further affecting cardiac output and increasing reliance on normal sinus rhythm. Geriatric patients may be highly pre-load dependent, with small decreases in preload (hemorrhage, decreased PO intake) having profound effects on cardiac output. Mid-systolic ejection murmurs are common and are secondary to aortic cusp thickening

Normal physiologic changes of the vascular system include atherosclerosis (leading to hardened arteries, reduced vascular compliance, and a widened pulse pressure), increased arterial wall thickness and decreased β2-mediated vasodilation. Vascular impedance increases, which ultimately leads to increased myocardial wall stress and oxygen consumption. Interestingly, afterload due to vascular impedance can increase even if SVR is unchanged – Nichols et al. studied 22 patients undergoing catheterization and found that there was no difference in aortic input resistance between hypertensive and non-hypertensive patients, but impedance was higher in the hypertensive group (142 vs 72 dyne-sec/cm5, p < 0.005) [Nichols et al. Circulation 74: 455, 1986]. Arterial pressure waves, which normally reflect back towards the heart (arriving during diastole in young, healthy patients), travel more quickly and reach the heart during systole, forcing it to pump against itself and significantly reducing ventricular efficiency

Aging also produces a state of beta-adrenergic insensitivity, and elderly patients are markedly less responsive to beta-agonists (ex. isoproterenol). Alpha sensitivity remains. The adrenal tissues atrophy some with age, but circulating norepinephrine levels increase ~ 2-4 fold in order to compensate for beta insensitivity [Ziegler et al. Nature 261: 33, 1976]. The end result is that autonomic responses are impaired. Furthermore, elderly patients seem to respond differently to hypotension – a study of 9 patients aged 74 years and 7 patients aged 27 years suggested that both groups maintained MAP values when subjected to a 60 degree tilt, however the young adults did so by increasing cardiac contractility and heart rate, whereas elderly patients relied on increased vascular resistance [Shannon et al. Am J Cardiol 67: 1110, 1991]


Lung tissue elasticity decreases with age, but so does chest wall compliance (kyphoscoliosis, decreased disc height), thus net pulmonary compliance is unchanged [Dauchot PJ et al. Prob Anesth 9: 498, 1997]. Loss of elasticity results in a 15% reduction in functional alveolar area by 70 years of age [Stoelting RK. Basics of Anesthesia, 5th ed. Elsevier (China) p. 519, 2007], resulting in an emphysema-like state (despite the fact that net pulmonary compliance is the same). Loss of alveoli in random regions of the lung leads to increased dead space an increase in V/Q mismatch

Large airways get bigger and small airways get smaller, which has a neutral effect on total resistance. Total lung capacity decreases slightly, FRC increases, and thus vital capacity must necessarily decrease. Closing capacity approaches FRC, thus it becomes possible for elderly patients to collapse their airways during expiration. Data from Rose’s prospectively collected PACU patients (24,157) suggested that the risk of critical respiratory events postoperatively is almost twofold higher (RR 1.92, adjusted relative odds ratio 1.54) in patients > 60 years of age [Rose DK et al. Anesthesiology 81: 410, 1994]

Both cardiovascular and pulmonary responses to hypoxemia and/or hypercarbia diminish with age [Peterson DD et al. Am Rev Respir Dis 124: 387, 1981]. Aspiration is more likely as the vocal cords become less sensitive


By 80 years the kidneys have lost 50% of their functional glomeruli, but these changes may not be clinically significant. Elderly patients do not require a specific fluid regimen, however they are less able to manage hypovolemia or hypervolemia (ie renal compensation is delayed)


Liver mass declines ~ 40% by 80 years of age, thus the effects of opiates, vecuronium, and other hepatically cleared drugs are prolonged in the elderly.


The most obvious pharmacologic issue with elderly patients is polypharmacy. Furthermore, protein binding sites and lean mass are reduced, potentially changing pharmacokinetics, although Barash claims that the clinical utility of this is negligible [Barash, PG. Clinical Anesthesia, 5th ed. (Philadelphia), p. 1222, 2006]. It is not known whether the increased potency of pharmacologic agents has do to with pharmacodynamic effects or changes in redistribution [Barash, PG. Clinical Anesthesia, 5th ed. (Philadelphia), p. 1224, 2006]


As in other patient populations, routine preoperative testing of elderly patients is not recommended unless coexisting medical illnesses are known or suspected. Indeed, a study of 544 patients > 70 years of age undergoing NCS showed that only ASA status and surgical risk were predictive of adverse outcomes (laboratory values were not) [Dzankic S et al. Anesth Analg 93: 301, 2001; FREE Full-text at Anesthesia & Analgesia]

There is debate about the utility of preoperative EKGs – Liu’s study of 513 patients > 70 years of age undergoing NCS showed that 75.2% had at least one abnormality on their preoperative ECGs and that the presence of abnormalities on preoperative ECGs was not associated with an increased risk of postoperative cardiac complications (OR 0.63, P = 0.26). The authors did not stratify patients based on type of EKG changes [Liu et. al. J Am Geriatr Soc 50: 1186, 2002]. That said, Noordzij et. al. retrospectively studied 23,036 patients (not restricted by age or NCS vs. cardiac surgery) who underwent 28,457 surgical procedures over 9 years. Patients with abnormal ECG findings had a greater incidence of cardiovascular death than those with normal ECG results (1.8% vs 0.3%; adjusted OR 4.5, CI 3.3 to 6.0) [Noordzij et. al. Am J Cardiol 97:1103, 2006]. Certainly a normal EKG does not rule out cardiac disease, that said the predictive value of an abnormal EKG is debatable. Chest X-rays are most likely to be beneficial if lung resection is planned

Blood Pressure

Howell et al reviewed 30 observational studies and found that the OR for perioperative cardiac complications was 1.35 in the presence of hypertension, but claimed this was clinically insignificant. Additionally, Howell found little evidence to support improved outcomes with admission BP < 180/110. Importantly, almost all studies of admission blood pressure are underpowered, and all but a two treat hypertension as a discontinuous variable [Howell SJ et al. Br J Anaesth 92: 570, 2004]

The two studies which do treat blood pressure as a continuous variable are both retrospective studies by Howell’s group [Howell SJ et al. Br J Anaesth 80: 14, 1998; Howell SJ et al. Br J Anaesth 82: 679, 1999], and neither show a significant relationship between admission blood pressure and outcome. Further complicating matters is “white coat hypertension,” which can occur in up to 20% of patients [Pickering TG et al. Blood Press Minot 4: 333, 1999]

Cardiovascular medications should generally be continued up to and including the morning of surgery, with the exception of diuretics [Stoelting RK. Basics of Anesthesia, 5th ed. Elsevier (China) p. 523, 2007]

Perioperative Management

There is no such thing as being “too old” for surgery. It is generally thought that comorbidities, which increase with age, are more important than age itself. That said, Forrest’s study of 17,201 patients showed that, following multivariate analysis (multiple stepwise logistic regressions), risk of severe outcomes is declines from 3% to 2% from the 20s to the 40s, but rises linearly afterwards (from 2% in the 40s to 6% in the 80s) [Forrest et al. Anesthesiology 76: 3, 1992]

Little information is available about the incidence of severe adverse outcomes, and even less information is available about the identification and quantification of independent predictors of severe perioperative adverse outcomes. The purpose of this study was to identify and quantitate independent predictors of severe perioperative adverse outcomes in a prospective randomized clinical trial of general anesthesia in 17,201 patients. Twenty-nine prognostic variables for 15 severe outcomes in 847 patients were tested by multiple stepwise logistic regressions from which 20 significant (P less than 0.05) predictors were identified. A history of cardiac failure or myocardial infarction less than or equal to 1 yr; ASA physical status 3 or 4; age greater than 50 yr; cardiovascular, thoracic, abdominal or neurologic surgery; and the study anesthetics were significant predictors of any severe outcome, including death. There were 17 significant predictors for 10 severe cardiovascular outcomes in 608 patients, including a history of ventricular arrhythmia, hypertension, cardiac failure, myocardial ischemia, myocardial infarction less than or equal to 1 yr or myocardial infarction greater than 1 yr, and smoking; ASA physical status; age; cardiovascular, thoracic, abdominal, eyes-ears-nose-throat/endocrine, neurologic, musculoskeletal, or gynecologic surgery; and the study anesthetics. There were 9 significant predictors for 4 severe respiratory outcomes in 163 patients, including a history of cardiac failure, myocardial ischemia, or chronic obstructive pulmonary disease; obesity; smoking; male gender; ASA physical status; abdominal surgery; and the study anesthetics. Colinearity between related prognostic variables (such as disease and ASA physical status) was assessed using progressively segregated groups of variables in eight stepwise logistic regressions. We conclude that the comprehensive stepwise logistic regression of 29 prognostic variables reported here provides a valid estimate of the risks of severe perioperative outcomes associated with general anesthesia

Preoperative functional status is another confounding factor that is known to affect outcomes [Eisler et al. Clin Orthopaed Rel Res 399: 52, 2002] and which is generally worse in the geriatric population

There is no “ideal” anesthetic regimen for elderly patients [Stoelting RK. Basics of Anesthesia, 5th ed. Elsevier (China) p. 524, 2007]. The majority of anesthetic agents are more potent in elderly patients, with the exception of atropine (the dose of which must be increased to generate a HR response). Volatile and intravenous agents generally last longer due to an increased volume of distribution. Volatile anesthetics are more potent in elderly, thus MAC requirements decrease (although onset of action may increase with diminished cardiac output). Propofol may produce an exaggerated drop in blood pressure. Etomidate doses can be decreased by 50% in individuals > 80 years of age. Midazolam is more potent and lasts longer. NMBDs are relatively unchanged

Regional vs. GA

Data showing a difference in outcome between general and regional anesthetic techniques are lacking [Go AS et al. Anesthesiology 84: 1, 1996]. That said, regional anesthetic techniques offer compelling theoretical advantages, and the lack of outcome differences likely reflects the safety of general anesthesia, and not the lack of any inherent difference between the two. The largest randomized, controlled trial comparing general to regional techniques includes only 423 patients undergoing femoral bypass procedures [Bode at al. Anesthesiology 84: 3, 1996] – this study was stopped after it was determined that, based on the complication rates at interim analysis, it would be impossible to find a statistically significant difference even if one did in fact exist [Go AS et al. Anesthesiology 84: 1, 1996]

Importantly, regional techniques are not without risks, and the risk of nerve palsies, paresthesias, and other complications are increased in the elderly population [Warner et al. Anesthesiology 81: 6, 1994; Hampl et al. Anesthesiology 88: 629, 1998; Martinez-Bourio R et al. Anesthesiology 88: 624, 1998]

The effect of anesthetic depth on outcome is an area of controversy – an observational study of 1064 patients suggested that time under BIS45 was associated with increased mortality rates at one year [Monk et al. Anesthesia and Analgesia 100: 4, 2005 ; FREE Full-text at Anesthesia & Analgesia], however the methodology and conflicts of interest inherent in this study (ex. use of an Aspect employee for statistical calculations) has come into question [Sclarlett J et al. Anesthesia and Analgesia 101: 1880, 2005]. A second study of 5056 consecutive patients who underwent BIS monitoring suggested that BIS less than 45 is related to mortality only in the subset of patients who had a preexisting malignancy with an unfavorable prognosis [Lindhold M et al. Anesth and Analg 108: 508, 2009; FREE Full-text at Anesthesia & Analgesia]

Cardiovascular responses to neuraxial blocks can be profound. Rooke studied 15 men from 59-80 years of age, with prior MI, CHF, or UA, given spinal anesthesia with 50 mg lidocaine. MAP decreased 33%, SVR dropped -26%, and cardiac dropped -10%. EF increased from 53% to 58% while LVEDV) decreased -19% [Rooke GA et al. Anesth Analg 85: 99, 1997]

Specific Strategies in the Elderly


Congestive heart failure will develop in 10% of patients aged 80 or older [Tresch D et al. J Am Geriatr Soc 45: 1128, 1997]. Many elderly patients develop diastolic heart failure, and in the general population it is thought that diastolic failure makes up ~ 1/3 of CHF cases [Vasan R et al. JACC 26: 1565, 1995]. Diastolic heart failure manifests as pulmonary congestion following ventricular overload, as well as exercise intolerance. Left ventricular relaxation can be predicted by examining the ratio between the E (peak early filling) wave and A (atrial filling) wave on Doppler – normal E/A ratio is 1-2. In patients with diastolic dysfunction, adequate filling is key – attempt to maintain a slow heart rate and NSR, optimize blood volume, and be wary of blood pressure changes or ischemia.

Because of the inherent risks of performing surgery on patients with significant diastolic dysfunction, some authors have proposed regional or central neuraxial blockade in these patients. A metaanalysis of 141 randomized trials including over 9500 patients showed reductions in mortality, DVT, pulmonary embolisms, transfusion requirements, pneumonia, and respiratory depression [Rodgers A et al. BMJ 321: 1493, 2001]

These benefits were not found in all studies, however Bode et al. randomized 423 patients with peripheral vascular disease to GA vs. epidural vs. spinal anesthesia, and found MACE incidences of 16.7%, 21.3%, and 15.4%, respectively, with an absolute risk difference between general and regional anesthesia of only -1.6% (95% confidence interval -9.2%, 6.1%) [Bode RH et al. Anesthesiology 84: 3, 1996]



Previous authors have claimed that geriatric patients a) do not experience as much pain as their younger counterparts and b) cannot tolerate opiates as well, but this has not been proven [Stoelting RK. Basics of Anesthesia, 5th ed. Elsevier (China) p. 525, 2007]. Epidural anesthesia should be strongly considered in elderly patients, as they offer improved function (ie decreased atelectasis, shorter time to extubation, and less ICU time according to a randomized trial of 915 abdominal surgery patients [Rigg JR et al. Lancet 359: 1276, 2002])

Mental Status

Delirium is associated with a significant mortality rate, but the relationship is not likely one of causation. In the past, it was believed that general anesthesia contributed to delirium, mainly based on Berggren’s study of 57 femoral neck fractures (half of whom received epidural analgesia, half of whom received halothane analgesia) [Berggren D et al. Anesth Analg 66: 497, 1987]

A much larger, and more recent (although retrospective) study of 9,400 hip fracture patients showed decreased unadjusted mortality in the general anesthesia group (compared to the epidural/spinal group, OR 0.80, CI 0.66-0.97), although the adjusted OR was higher (OR 1.08, CI 0.84-1.38). There was no difference in postoperative mental status changes [O’Hara DA et al. Anesthesiology 92: 947, 2000]

Recovery from Anesthesia

Full recovery of psychomotor function can be delayed in the geriatric population [Shinozaki M et al. Can J Anaesth 49: 927, 2002]. Full functional recovery may take up to two weeks [Muravchick S. Clin Anaesthesiol 4: 1035, 1986], and full neurocognitive function (as measured by neurocognitive testing) may take several months to return to baseline, with the highest incidence of 3 month dysfunction in patients > 70 years of age (14%), those who had a second operation (14%), and those who had a respiratory complication (14%) [Moller KT et al. Lancet 351: 857, 1998]