3 General Assessment and Optimization for Surgery
A patient with a hip fracture presents a challenging, high-risk situation for surgical, anesthetic, and medical consultants. Fracture repair is neither elective nor truly an emergent situation with immediate risk to limb or life of the patient. Bushnell et al classified this type of surgery as “necessary” or urgent.1 In patients with a hip fracture, the operation is performed as soon as it is safe, and the perioperative medical consultation must be both thorough and rapid.1 This chapter describes the medical assessment and optimization of a patient hospitalized with a hip fracture. Evidence is available to help guide perioperative planning and risk estimation. Controversy and debate are ongoing about whether these surgical procedures should be delayed for medical optimization, how much delay is acceptable, which medical problems should be optimized, and whether this optimization improves perioperative outcomes.
The medical assessment of the surgical patient includes the following steps (Table 3-1): (1) assessing the cause of the fall that resulted in hip fracture, (2) establishing the patient’s prior comorbidities, (3) assessing any acute physiologic derangements, (4) searching for active cardiac conditions that are known to increase risk, (5) estimating the patient’s risk, (6) refining this risk based on specific investigations, and (7) performing targeted optimization of the patient with an elevated risk in preparation for the surgical procedure. A patient at significant risk for developing a particular outcome (e.g., ischemia, arrhythmia, heart failure, delirium) should also be monitored closely for this event postoperatively (e.g., electrocardiograms and cardiac markers, telemetry, screening for delirium) in an appropriate setting (e.g., intensive care unit [ICU], step-down unit, or ward).
|Determine the cause of fall resulting in hip fracture||Cardiac syncope|
|Establish underlying comorbidities||Stroke and transient ischemic attack|
Coronary artery disease
|Assess for acute physiologic derangements||Hyponatremia|
Acute asthma flare
|Risk Stratification||Conditions and Actions|
|Search for cardiac conditions associated with increased risk 22||Unstable or severe angina|
Acute congestive heart failure
Severe stenotic valvular disease
|Estimate patient’s perioperative risk||Calculate Revised Cardiac Risk Index|
Calculate Respiratory Failure Risk Index
|Perform investigations to refine patient risk||Perform echocardiography|
Obtain dipyridamole technetium cardiac scan
|Optimize patient to minimize risk||Treat pneumonia|
Optimize fluid status in congestive heart failure
Continue specific medications during fasting by the intravenous route, if necessary
Ensure deep vein thrombosis prophylaxis
Hip fractures occur most frequently in elderly persons, and the risk of hip fracture increases exponentially with increasing age.2 In 2004 in the United States, more than 320,000 patients were admitted to the hospital with a hip fracture.3 This number represents a rate of approximately 850 per 100,000 population.3 Comparably, in Canada in 2008, there were 456 fractures per 100,000 seniors4 and 78,554 admissions in 2002 in England.5 This population has a very high risk of in-hospital mortality, reported to be up to 7%.4 The 30-day mortality has been reported to be between 5% and 12%, and the reported 1-year mortality is 12% to 37%.2,6 The expected mortality in this age group is approximately 10%.7 Generally, this increased risk of death from hip fracture occurs soon after the event. Leibson et al found that the risk of death at 3 months was 12% in patients who had hip fractures, compared with 3% in aged-matched patients who did not have hip fracture and who were seeking general medical advice (risk ratio, 4). At 1 year, the risk ratio for death fell to 1.8 (20% versus 11%) and to 1.2 after 5 years (52% versus 44%),2 a decline that, in part, may also represent survivor advantage.
This striking mortality may result from the increasing numbers of comorbidities with age and consequently the higher risk of surgical complications.8 Most (approximately 88%) of these patients are 65 years old or older, and their mean age is 83 years (range, 52 to 105 years).4,8 The average age at presentation is also rising (from 67 years in 1944 to 79 years in 1993).9 Some of the common comorbidities include dementia (23%), diabetes mellitus (18%), congestive heart failure (CHF; 16%), and chronic obstructive pulmonary disease (COPD; 14%). The presence of these comorbidities has been shown to increase the risk of postoperative complications two- to threefold.8
Some investigators have implicated age itself as an independent risk factor for perioperative complications. In 2009, Kheterpal et al analyzed more than 7500 patients and identified that age 68 years or older was an independent predictor of perioperative cardiovascular adverse events.10 Age was also identified as a risk factor for perioperative complications by Detsky et al,11 as well as in the 2002 version of the American Heart Association/American College of Cardiology guidelines.12 However, the most widely used preoperative risk stratification index (the Revised Cardiac Risk Index [RCRI]) does not include age as an independent predictor of adverse events.13
The development of postoperative complications increases mortality rate. Roche et al studied 2448 consecutive patients with acute hip fracture.5 This population had mortality rates of 9.6% at 30 days and 33% at 1 year, findings similar to those of many other studies. Patients who developed respiratory infections had increased mortality rates of 43% at 30 days and 71% at 1 year. Postoperative CHF was associated with even worse outcomes, mortality rates of 65% at 30 days and 92% at 1 year. Roche et al also found that thromboembolic disease that developed despite prophylactic medications increased mortality rates 4.5-fold.5 Whether perioperative involvement of medical specialists changes these outcomes is not known.
The high risk of mortality and morbidity requires the entire medical team, including the surgeon, the anesthesiologist, and the internist, to collaborate on the timing of surgery, the preoperative optimization of the patient, the minimizing of postoperative complications, and the appropriate management of any complications that may occur.14
Deleterious cardiac outcomes include myocardial infarction, pulmonary edema, ventricular fibrillation, cardiac arrest, and complete heart block. Less serious are supraventricular arrhythmias including atrial fibrillation, demand-driven angina, and mild fluid overload. Pneumonia, exacerbation of underlying lung disease (asthma and COPD), and hypercapnic respiratory failure are commonly encountered postoperative respiratory events. Most physicians concentrate on the cardiac risk and respiratory risk assessment because these parameters are most directly linked to mortality and morbidity and are potentially modifiable. The thromboembolic risk has been well described, and standard protocols to reduce the rates of thromboembolism are recommended in the American College of Chest Physicians (ACCP) guidelines.15 Some adverse outcomes may have subtle manifestations. For example, in patients with delirium, electrolyte abnormalities, hyperglycemia, and transient renal dysfunction, the diagnosis is often missed or even dismissed.
The timing of surgery in patients with hip fracture depends on several factors. The availability of operating room time, the availability and preference of orthopaedic surgeons, and medical investigation and optimization potentially contribute to the delay in surgical repair.6,16,17 Orosz et al classified the potential reasons for delay of surgery (Table 3-2).16 A delay of 24 to 48 hours was the result of delayed routine medical clearance in 52% of patients; in 41% of the patients, the delay was related to unavailability of the operating room or surgeon; and 8% of the patients in this study required medical stabilization. A delay of more than 48 hours resulted from medical clearance issues in 63% of these patients; in 44% of patients, it was caused by unavailability of the surgeon or operating room; and the delay reflected a need for medical stabilization in 35% of these patients. Many of the patients in this study had more than one reason for the postponement of their surgical procedures.16
|Awaiting medical consultation or clearance|
|Not having available operating room or surgeon|
|Awaiting family discussion|
|Awaiting laboratory results/other studies|
|Awaiting stabilization of a medical problem|
|Admitting patient too late in the day|
Adapted from Orosz GM, Hannan EL, Magaziner J, et al. Hip fracture in the older patient: reasons for delay in hospitalization and timing of surgical repair. J Am Geriatr Soc 2002;50:1336-40.
The time of day and the day of the week of admission may further delay the surgery. In a Canadian review of hip operations, patients admitted to hospital between midnight and noon or on weekends were more likely to have their operations within 48 hours. Although this finding seems counterintuitive from the perspective of staffing during these times, this discrepancy may reflect the impact of elective surgery on operating room availability.17
The timing of surgery may be complicated further by delayed presentation of the patient to the hospital after the incident hip fracture. In a prospective review of 571 patients with hip fracture who presented to 4 major metropolitan hospitals in the United States, 17% of the patients sought help more than 24 hours after the injury. Approximately half of those patients presented more than 72 hours after the initial fall. Most of these patients (77%) did not realize that they had suffered a hip fracture, and some did not present to the hospital earlier because they were unable to communicate their injury to others.16 Whether this delay contributes to an increased rate of complications and mortality has been addressed by several studies.
The timing (delay) of surgery may affect the rate of postoperative complications, functional recovery, functional independence, hospital length of stay, and possibly mortality. Increasing rates of deep vein thrombosis (DVT; ≤62%), decubitus ulcers, infections (pneumonia, urinary tract infections), and loss of bone density and muscle mass have been reported even after a relatively short period of immobility.8,16–19 These and other variables that remain unaccounted for have the potential to increase mortality rates in patients whose time from injury to surgical treatment is prolonged.17
Multiple prospective and retrospective studies have tried to address the relationship between timing of surgery and outcomes. If an association between delay and poor outcomes (especially mortality) exists, one must consider whether the delay itself is the culprit or whether it is a marker for patient-related complexity that reflects the need for preoperative testing and optimization (e.g., reversal of anticoagulation, management of decompensated heart failure).17 Ideally, this association would be better understood from randomized controlled trials. For ethical and feasibility reasons, such studies will likely never be performed.6 Most data on the effect of surgical delay on patient outcomes are observational. These studies are often prone to bias and may establish association rather than causation.
A meta-analysis by Shiga et al of 16 English-language studies that involved more than 250,000 patients identified an increased risk of 30-day mortality when the operation was delayed by 48 hours after admission (odds ratio, 1.4; number needed to harm, 20).6 Similarly, there was an increased odds ratio of 1-year mortality of 1.3 (number needed to harm, 40). Unfortunately, the quality of these studies was poor, with an average quality score of 14 out of a maximum of 32.6 The metaregression determined that delaying operation in patients with low baseline risk and in young patients increased the risk of all-cause mortality.6 This finding suggests that early surgical intervention may be more beneficial in these patients. Alternatively, the delay resulting from optimization of higher-risk patients may have lowered this group’s mortality rates.
Orosz et al demonstrated the relationships among delay in surgery, comorbidities, and outcome.20 In a prospective study of 1178 consecutive patients admitted to 4 different hospitals, the hazard ratio for mortality in the early-surgery group (<24 hours) was 0.68 (95% confidence interval [CI], 0.48 to 0.97). After propensity score matching of the two groups, no difference in mortality was noted (odds ratio, 0.98; 95% CI, 0.63 to 1.50).20 Earlier surgical treatment resulted in fewer days of severe and very severe pain (difference of 0.22 day) and a shorter hospital stay (difference of 1.94 days). Postoperatively, the variables of pain and length of hospital stay were not affected by earlier operation. Early surgical treatment in patients who were medically stable preoperatively resulted in a significant reduction in the incidence of major complications (odds ratio, 0.26; 95% CI, 0.07 to 0.95).
Thus, early surgical treatment is the goal in patients who are medically stable. Delay in seeking medical consultation for the purpose of clearing a patient for surgery (in patients who do not require optimization) increases the incidence of major postoperative complications. In general, delaying patients’ operations for more than 48 hours may increase mortality. Preoperative medical consultation and optimization should be targeted to correcting physiologic abnormalities associated with worse outcomes (Table 3-3). Delaying surgery for more than 48 hours to continue optimization may not be beneficial because of the increased rate of complications (e.g., decubitus ulcers, urinary infections, venous thrombosis) associated with postponement of operation. The role of specific interventions beyond stabilization of physiologic derangements (e.g., beta-blockade, imaging, revascularization) is addressed later in this chapter.
|Cardiovascular||Hypotension (systolic blood pressure <90 mm Hg)|
Hypertension (systolic >180 and/or diastolic >110 mm Hg)
|Respiratory||Lower respiratory tract infection (bronchitis, pneumonia)|
Acute asthma/COPD exacerbation
Respiratory failure (Pco2 >55 mm Hg, Po2 <60 mm Hg, oxygen saturation <90%)
|Renal||Acute renal failure (creatinine >185 μmol/L, urea >14 mmol/L)|
Electrolyte disturbance (Na <125 or >155 mEq/L, K < 2.5 or >6 mEq/L)
Acid-base disturbance (HCO3– <18 or >36 mEq/L)
|Endocrine||Uncontrolled diabetes (glucose >25 mmol/L)|
Thyroid disorders (thyrotoxicosis, hypothyroidism)
|Hematologic||Anemia (hemoglobin <80 g/L)|
Coagulopathy (INR >1.5)
COPD, chronic obstructive pulmonary disease; HCO3–, bicarbonate; INR, international normalized ratio; K, potassium; Na, sodium; Pco2, partial pressure of carbon dioxide; Po2, partial pressure of oxygen.
Adapted from Orosz GM, Magaziner J, Hannan EL, et al. Association of timing of surgery for hip fracture and patient outcomes. JAMA 2004;291(14);1738-43.
Perioperative risk is the combination of surgery-specific risk and intrinsic patient-specific risk based on preexisting comorbidities and acute physiologic derangements. Determination of this risk can help to guide the team in deciding whether preoperative optimization is necessary or worthwhile and whether patients should be subjected to aggressive postoperative surveillance, monitoring, and management.10 A patient with a low predicted perioperative risk not only may derive no benefit from optimization, but also may suffer both as a result of the delay and from complications related to unwarranted investigations unrelated to surgery. Conversely, a high-risk patient undergoing a high-risk procedure should be critically evaluated and optimized by the anesthesiologist and medical specialist before the proposed operation.10,14
The magnitude of the risk to the patient will determine the degree of invasive monitoring required (arterial lines, central lines, and pulmonary artery catheter) and the appropriate postoperative setting (ICU, step-up unit, cardiac-monitored bed, or general ward bed) after discharge from the postanesthesia care unit. The surgical team may decide on an alternate plan of management (e.g., a nonoperative or less ideal, less invasive procedure) if the patient is deemed to be at too high a risk of complications and mortality.
Factors that contribute to increased surgical risk include the duration of the operation, blood loss, fluid shifts, and the extent of surgical insult resulting from manipulation of bone and soft tissue. Extracapsular fractures often have significant blood loss (≤1 L) associated with the fracture itself, and the surgical procedure may result in further blood loss.7 Mortality is also increased in patients with these types of fractures (38% 1-year mortality versus 29% in patients with intracapsular fractures).9 Operative duration of 3.8 hours or longer and administration of 1 or more units of packed red blood cells are independent predictors of adverse cardiac outcomes.10 Bone marrow instrumentation and application of cement put the patient at a risk of systemic embolization.14 The surgeon’s expertise will guide the decision about the type of operative management. Total hip replacement performed for subtrochanteric hip fractures is much more invasive and is associated with higher rates of blood loss, fluid shifts, cardiac arrest, and death compared with hip replacement in displaced femoral neck fractures. Pins, intramedullary nails, and other fixation devices are less invasive and consequently are associated with a lower perioperative risk.14
Intrinsic patient risk results from the interaction of preexisting comorbidities, functional and cardiac status, and acute physiologic derangements. Patients presenting with a hip fracture often have one or more acute physiologic derangements. These abnormalities may be associated with increased perioperative complications and thus may warrant preoperative correction.
Some common preoperative abnormalities are listed in Table 3-3. In a prospective cohort study by McLaughlin et al, 34% of patients with hip fracture had minor abnormalities (e.g., systolic blood pressure >180 mm Hg, chest pain with normal electrocardiogram, glucose >25 mmol/L).8 Major abnormalities (e.g., fever from pneumonia, acute pulmonary edema, severe hyponatremia with serum sodium <125 mEq/L) were present in 23% of patients. The presence of two or more major abnormalities on admission to the hospital was associated with a greater than fourfold increase in postoperative complications. When these major abnormalities were still present preoperatively, the risk of postoperative complications further increased 12-fold. These investigators observed that patients with only minor or no abnormalities preoperatively had a postoperative probability of complications of 7%. That risk increased to 21% in the presence of major abnormalities.8
To date, no prospective, randomized clinical trials have determined which of these abnormalities and what degree of derangements should be corrected. Most of the literature is expert-based opinion or is derived from nonperioperative settings. Overzealous correction and normalization of abnormalities are often unnecessary and may delay surgical treatment. Rapid correction of abnormalities that are thought to be acute but in fact are chronic may harm the patient. For example, rapid correction of hyponatremia when the sodium concentration is less than 125 mmol/L may result in central pontine myelinolysis with devastating neurologic consequences.21 Rapid correction of blood pressure in a chronically hypertensive patient may result in cerebral hypoperfusion and stroke, myocardial infarction, or renal failure.
The American Heart Association/American College of Cardiology (ACC/AHA) 2007 perioperative guidelines outline several conditions that are major predictors of poor outcome (Table 3-4). In patients undergoing elective surgical procedures, the presence of one or more of these conditions requires further investigation, possible optimization, and delay or cancellation of the operation.22 In the hip fracture patient, the degree of investigation and optimization is constrained by the short period of time available. It is possible to delay repair of acute hip fracture for up to 48 hours without undue increased risk to the patient (see “Optimal Timing of Surgery,” earlier). Recent myocardial infarction and severe valvular disease optimization require investigations and procedures that may extend past the short time frame available. For example, therapy for severe aortic stenosis requires echocardiographic assessment, surgical consultation, valvular replacement, and postoperative convalescence. Unstable coronary artery disease usually involves assessment with stress testing, angiography, angioplasty, or surgery, even though no evidence supports the use of these interventions perioperatively. Only patients with acute myocardial infarction or patients with hemodynamic instability related to their underlying coronary disease would potentially benefit from such interventions. Conversely, the consultant can potentially intervene and stabilize patients with decompensated heart failure or unstable or uncontrolled arrhythmia and thereby negate the effect of a major predictor of poor cardiac outcome.
|Unstable coronary syndrome||Myocardial infarction within 30 days|
Canadian cardiovascular class III or IV angina
|Decompensated congestive heart failure||New-onset heart failure|
Worsening heart failure
Severe symptomatic chronic heart failure (New York Heart Association class IV)
|Uncontrolled arrhythmia||Rapid atrial fibrillation (heart rate >100 beats/min)|
Third-degree heart block
Symptomatic bradycardia of other origin
|Severe valvular disease||Aortic stenosis with valve area <1.0 cm2 or a mean gradient >40 mm Hg|
Severe mitral stenosis (symptomatic of syncope, presyncope, dyspnea, or congestive heart failure)
Adapted from Fleisher LA, Beckman JA, Brown KA, et al. ACC/AHA 2007 Guidelines on perioperative cardiovascular evaluation and care for noncardiac surgery: executive summary. J Am Coll Cardiol 2007;50(17):1707-32.
Estimation of a patient’s perioperative cardiac risk is commonly used in planning for surgery and in directing cardiac investigations with the view of optimizing the patient’s preoperative care before elective operations. The role of risk assessment in a patient with an acute hip fracture is less clear because the evidence of benefit of cardiac optimization is not well defined. Nonetheless, knowledge of the risk of complications should be the basis for educating the patient and his or her family about the expected postoperative course and the perioperative care plan. Cardiac optimization immediately before the operation is discussed later in the chapter.
One of the first systematic attempts of quantifying patients’ perioperative cardiac risk was reported in 1977 by Goldman et al.23 Subsequently, Detsky et al published a Bayesian approach to establishing patients’ risk.11 A more recent, and currently most commonly used approach, is the use of the RCRI.13 Lee at al identified six risk factors as predictors of adverse cardiac outcomes (outcome defined as myocardial infarction, pulmonary edema, ventricular fibrillation, cardiac arrest, and complete heart block) These factors are listed in Table 3-5.13 In this system, increasing numbers of factors are associated with an increasing rate of perioperative cardiac events (Table 3-6). This simple risk index has predictive characteristics similar to those of the Goldman and Detsky indices, with reduced complexity and more current patient data reflecting recent advances in surgical and anesthetic techniques. Unfortunately, the data used to derive the RCRI were derived from a population of patients who were undergoing elective noncardiac surgery.13 This factor may make the index less valid in urgent/emergent surgical patients. Indeed, Detsky et al identified emergency surgery as a high-risk situation with a risk similar to that of recent myocardial infarction, CHF in the week preceding the operation, or moderate to severe coronary artery disease (Canadian Cardiovascular Society class III to IV angina).11
|High-risk type of surgery|
|Ischemic heart disease|
|History of congestive heart failure|
|History of cerebrovascular disease|
|Insulin therapy for diabetes|
|Preoperative serum creatinine >2.0 mg/dL|
From Lee TH, Marcantonio ER, Mangione CM, et al. Derivation and prospective validation of a simple index for prediction of cardiac risk of major noncardiac surgery. Circulation 1999;100:1043-9.
|Class (No. of Risk Factors)||Cardiac Event Rate (%)|
|IV (3 or more)||11.0|
From Lee TH, Marcantonio ER, Mangione CM, et al. Derivation and prospective validation of a simple index for prediction of cardiac risk of major noncardiac surgery. Circulation 1999;100:1043-9.
A validation study of the Lee cardiac risk index identified orthopaedic surgery in general as low to intermediate risk, with approximately one third of the risk of other intermediate-risk procedures (e.g., abdominal, thoracic, head and neck) and an odds ratio for perioperative cardiac death at 30 days of 2.8 versus 10.3.24 In the same study, emergency surgery carried a greater than 10-fold increase in risk (6.1% versus 0.5%).24 The interaction of preexisting patient comorbidity with the emergent nature of hip fracture surgery likely explains the high perioperative mortality rates observed in this patient population. It is likely that the RCRI underestimates risk in the hip fracture patient.