The practices of anesthesiology, surgery, and critical care are continuously improving. Through advances in each field, a number of patients with increasingly severe comorbidities are undergoing riskier and more complex operations and experiencing better outcomes. In recent years, intraoperative mortality has decreased by a factor of 10. Nonetheless, perioperative morbidity and mortality remain high. If perioperative mortality were classified as a disease, it would be the third leading cause of death in the United States. Thirty-day postoperative mortality after noncardiac surgery could be as high as 1% and 2% for inpatients in the United States. Despite an enhanced ability to effectively care for this growing high-risk group, these patients remain at substantial risk for the development of perioperative organ dysfunction—myocardial, pulmonary, neurologic, and renal. The degree of dysfunction ranges from mild (sometimes silent) and even undetected injury to profound organ injury, coma, or death. The implications of the more immediate and severe injury occurring in the perioperative period have long been identified, but only recently has it been noted that injury thought to be transient may have long-term consequences. This realization is at the core of this book. In this chapter, we focus on identifying perioperative morbidity and touch on strategies to prevent or to treat these complications, many of which will be described further in subsequent chapters.
Cardiac Injury
Myocardial injury has long been a dreaded complication during and after surgery. Each year, over 1 million people having noncardiac surgery will experience a cardiovascular complication. Although the number of patients with documented acute myocardial infarctions within 30 days of surgery is significant, the number of patients who likely experience silent and undetected myocardial injury during and after surgery is sobering. This injury now has a name: “MINS” (myocardial injury after noncardiac surgery). MINS is defined as a peak troponin T of 0.03 ng/mL or greater judged to be due to myocardial ischemia (i.e., no evidence of a nonischemic etiology causing the troponin T elevation); the definition does not require the presence of an ischemic feature such as electrocardiogram changes or anginal symptoms. Due to the common absence of ischemic features, it is estimated that more than 80% of MINS events will be missed without routine monitoring of troponin levels after surgery. However, the 30-day mortality increase for patients with MINS suggests that this is an important perioperative event with implications for changes in clinical management. The VISION (Vascular Events in Noncardiac Surgery Patients Cohort Evaluation) trial was a prospective international study of more than 15,000 patients who received routine troponin monitoring for 72 hours postoperatively. It demonstrated that patients with peak troponin T concentrations less than 0.01 ng/mL had a 1.0% mortality, whereas patients with concentrations of 0.02 ng/mL, 0.03–0.29 ng/mL, or 0.30 ng/mL or greater had 30-day mortality rates of 4.0%, 9.3%, or 16.9%, respectively. A composite of nonfatal cardiac arrest, congestive heart failure, stroke, and death occurred in 18.8% of the MINS cohort and only 2.4% of patients without MINS in the VISION study, an eightfold increase. A similar study in a colorectal surgery population echoed these results. In this study, mortality of patients with troponin levels greater than 0.01 ng/mL within the first 48 hours after surgery was 20%.
Importantly, the mortality attributed to MINS is not exclusively cardiac in nature. Nevertheless, recognition of MINS by the perioperative physician is an opportunity to improve outcomes. In the colorectal study, 17 of 40 patients with elevated troponin levels went on to receive an ischemic evaluation and were started on medical therapy, which may have prevented worse outcomes. Research on whether instituting medical therapy in MINS will reduce mortality is ongoing.
While recognizing and responding to myocardial injury postoperatively is an important area to target to improve patient outcomes, preventing myocardial injury in the first place has been an area of intense study over the past two decades. Large prospective clinical trials investigating the ability of pharmaceutical interventions to reduce myocardial injury, morbidity, and mortality were some of the first trials with adequate power and long-term outcome assessment to lead to an understanding of the implications of perioperative injury. The use of perioperative β-blockade to reduce myocardial injury became popular two decades ago. The POISE (Perioperative Ischemic Evaluation) trial in 2008 challenged widespread use of β-blockers, showing a reduction in ischemic events but an increase in bradycardia, hypotension, strokes, and all-cause mortality for high-risk patients not receiving a β-blocker prior to surgery. A 2014 systemic review of randomized controlled trials (RCTs) investigating new institution of perioperative β-blockage supported the POISE trial’s results, suggesting that the reduction in nonfatal myocardial infarction was offset by increases in nonfatal strokes, bradycardia, and hypotension. However, for patients with coronary or peripheral arterial disease or more than two risk factors for coronary disease who were maintained on a β-blocker prior to surgery, perioperative withdrawal of β-blockage was associated with increased 30-day and 1-year mortality. A Department of Veterans Affairs (VA) study of angiotensin-converting enzyme (ACE) inhibitors showed that resumption of the drug within 2 weeks of surgery was associated with decreased 1-month mortality. The INPRESS (Intraoperative Norepinephrine to Control Arterial Pressure) study was a multicenter RCT which concluded that an individualized approach to blood pressure management which kept systolic blood pressures within 10% of baseline blood pressure, as compared with a standard management strategy, reduced the incidence of postoperative organ dysfunction by 40% in high-risk surgical patients.
The good news is that from 2004 to 2013, major adverse cardiac events decreased from 3.1% to 2.6%, mortality decreased from 2.0% to 1.3%, and the rate of acute myocardial infarction decreased from 1.0% to 0.8%. Regardless of whether we can distinguish association from causality, we can, from a prediction standpoint, identify the individuals who are at greater risk and identify strategies to reduce the probability of later morbidity and mortality and improve outcomes for high-risk patients.
Pulmonary Injury
Perhaps the most common perioperative complications, and possibly the most preventable, are perioperative pulmonary complications (PPCs), which include respiratory infection, respiratory failure, pleural effusion, atelectasis, pneumothorax, bronchospasm, aspiration pneumonitis, pulmonary embolus, and acute respiratory distress syndrome. Though the incidence of these complications varies widely by study, the 30-day mortality rate for patients who have one of these complications is between 14% and 30%. A 2017 study published in JAMA Surgery of 1200 American Society of Anesthesiologists physical status 3 patients suggested a PPC rate of 33.4%, with a significant proportion of these complications being prolonged need for oxygen therapy and atelectasis. However, any PPC increased early postoperative mortality, intensive care unit (ICU) admission, and ICU and hospital length of stay. The Closed Claims Project identified 92 claims in which respiratory depression caused significant harm and concluded that 97% of these events were preventable.
Focusing on strategies to optimize nonmodifiable risk factors and minimize modifiable risk factors can reduce these numbers. Smoking cessation before major surgery can reduce postoperative morbidity. One study published in JAMA Surgery suggested that limiting colloid administration, reducing blood loss and anesthetic duration, and low-tidal volume ventilation might all reduce the incidence of PPC. A study of 69,235 patients comparing protective ventilation of less than 10 mL/kg with plateau pressures less than 30 cm H 2 O and positive end-expiratory pressure (PEEP) greater than 5 cm H 2 O versus nonprotective ventilation showed a decrease in a composite of major respiratory complications (adjusted odds ratio, 0.90; 95% confidence interval [CI], 0.82 to 0.98; P = 0.013). A meta-analysis of 15 RCTs of over 2000 general surgery patients showed that a low-tidal volume ventilation strategy (< 8 mL/kg ideal body weight) reduced PPCs by an adjusted relative risk of 0.64. Although less compelling, there is some evidence suggesting that moderate PEEP of 6–8 cm H 2 O combined with low tidal volumes also reduced PPC. Inadequate reversal of neuromuscular blockade is the most common reason for respiratory failure in the postoperative care unit. A retrospective observational study of 11,355 patients at VA hospitals showed an odds ratio of 1.75 for PPC in patients who did not receive neostigmine reversal as compared with patients who did. The POPULAR (Post-Operative Pulmonary Complications After Use of Muscle Relaxants in Europe) study was a prospective observational study of 22,803 patients which suggested that patients receiving a neuromuscular blocking agent had a 4.4% relative risk of PPCs as compared with patients receiving general anesthesia without neuromuscular blockade. This increased risk was independent of the use of reversal agents.
Patients at high risk for postoperative extubation failure may be managed successfully with noninvasive ventilation strategies. A 2009 RCT in 500 cardiac surgery patients showed that continuous positive airway pressure (CPAP) of 10 cm H 2 O for 6 hours after extubation reduced PPCs. Multiple studies have found that use of a high-flow nasal cannula was noninferior to CPAP or bilevel positive airway pressure for preventing reintubation and respiratory failure.
Up to 24%–41% of patients presenting for surgery have obstructive sleep apnea (OSA), and many of them are undiagnosed. These patients are at increased risk of desaturation events and respiratory arrest around the time of surgery. A retrospective case–controlled study of OSA patients undergoing orthopedic procedures demonstrated longer hospital stays and 2.5 times the number of postoperative complications for OSA patients (24% vs 9% for control subjects). Those patients who experienced respiratory and cardiovascular complications had 28-day mortality rates of 26.1% and 17.7%, respectively. Improving outcomes for these patients could be as simple as providing end-tidal carbon dioxide monitoring or continuous pulse oximetry for 24 hours after surgery. A meta-analysis of capnography monitoring reported that it recognized six times more respiratory events than pulse oximetry. Perioperative monitoring is recommended until these patients are no longer at risk for respiratory depression, though monitoring has never been proven to improve outcomes in a clinical trial. The use of multimodal analgesia and postoperative CPAP also reduced the risk of apneic events and overall complication rates.
Neurologic Injury
Perioperative neurologic injury ranges from acute stroke to a spectrum of neurocognitive disorders ( Table 1.1 ). Stroke is a leading cause of severe disability today, and perioperative strokes (estimated to be 15%–20% of all strokes) remain significant contributors. The rate of perioperative stroke in noncardiac surgery has increased from 0.5% to 0.8%, possibly related to the use of β-blockers. However, perioperative neurocognitive dysfunction is increasingly recognized as a serious, common clinical entity, and its characterization and prevention are the focus of major research and clinical initiatives.
| Time period preoperative | Term and Definition | Comments as in community | ||
|---|---|---|---|---|
| Mild NCD | Major NCD | |||
| Perioperative cognitive disorders | ||||
| Emergence form anaesthesia | Emergence excitation or delirium | |||
| From : Immediately postoperative Until : Expected recovery (to 30 days) ⁎ | Delirium (postoperative † ) OR Delayed neurocognitive recovery | Delayed neurocognitive recovery ‡ | Delayed neurocognitive recovery ‡ | The time for expected resolution is based on perioperative conditions, e.g., complications/infection/prolonged hospitalization |
| From: Expected recovery (30 days) Until : 12months | Mild NCD postoperative (POCD) | Major NCD postoperative (POCD) | POCD is an indicator of the temporal associated with the anaesthesia/surgery/event | |
| Beyond 12 months | Mild NCD | Major NCD | As in community if a new diagnosis after this time | |
⁎ i.e., the time point from at least 30 days when the effects of anaesthesia and surgery should have resolved.
† The postoperative specifier will be applied where delirium is persistent beyond hospital discharge.
‡ Also applies to assessments prior to medical readiness for discharge although it should be noted any decline prior to readiness for dischrage is unlikely to represent a delay in recovery.
Few would question the significance of perioperative stroke, but perioperative neurocognitive dysfunction has been thought of historically as a transient process without substantial long-term implications. However, it is now known that the incidence of postoperative delirium is significant and that delirium and delayed neurocognitive recovery (if mental status remains altered up to 30 days) can lead to long-lasting cognitive deficits and increased mortality. While statistics vary widely, the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, states that delirium occurs in 15%–53% of elderly patients and 70%–87% of ICU patients. A 2018 study noted that those patients with the most severe delirium had three times the rate of cognitive decline in the 3 years following their surgery. Patients with preexisting cognitive impairment have a synergistically elevated risk of developing new and worsening cognitive deficits. Regardless of whether the defined association indicates that more severe perioperative injury causes greater long-term morbidity and mortality or whether these individuals were at higher risk for later decline because the numbers of elderly patients and patients with cognitive impairment who present for surgery are increasing, this problem should be a major focus in improving perioperative care.
In 2015, the American Society of Anesthesiologists proposed the Brain Health Initiative as a multidisciplinary group to “create a low barrier access program to minimize the impact of pre-existing cognitive deficits, and optimize the cognitive recovery and perioperative experience for adults 65 years and older undergoing surgery.” The initiative highlights many ways in which care can be improved. The American Geriatrics Society’s (AGS) Beers criteria list medications that have an unfavorable risk profile for elderly patients, many of which can be used during surgery. This list includes avoidance of benzodiazepines, diphenhydramine, and anticholinergics; sparing use of antipsychotics; and preferential use of dexmedetomidine. Although the evidence is inconclusive, best practice guidelines from the American College of Surgeons, AGS, and National Surgical Quality Improvement Program suggest preferential use of regional and neuraxial anesthesia over general anesthesia. The use of multimodal analgesia and opioid-sparing techniques, early mobilization and avoidance of pressure ulcers, aggressive pulmonary hygiene, judicious use of fluids, and meticulous management of indicated cardiac medications are other best practice suggestions ( Table 1.2 ). There has been intense interest in the use of processed electroencephalogram (EEG) monitoring as a way to reduce anesthetic dosages and delirium. Evidence is mixed. The 2012 CODA (Cognitive Dysfunction after Anesthesia) trial was an RCT of bispectral index (BIS)-guided anesthetic delivery versus usual care for elderly patients, which concluded that there was a significant reduction in delirium and postoperative cognitive dysfunction in the intervention arm. Low intraoperative BIS values, deep anesthesia for long periods, and large doses of anesthetic were predictors of postoperative cognitive dysfunction. However, the results of the ENGAGES (Electroencephalography Guidance of Anesthesia) trial refute this. This trial was an RCT of 1232 elderly patients that also looked at processed EEG-guided management versus usual care and did not find a reduction in delirium, despite a reduced dose of anesthetic in the EEG group.
