Introduction
Patients undergoing thoracic surgeries are different from those undergoing most noncardiac surgeries. Although the risk for perioperative major cardiac complications is significant, postoperative pulmonary complications (PPCs) contribute equally, if not more, to perioperative morbidity and mortality in this group of patients. Almost by definition, lung resection leaves patients with impaired pulmonary function (with some important exceptions, see later). The purpose of preoperative assessment of thoracic surgery patients is to identify high-risk patients (including those in whom the risk may preclude surgery), to plan the intraoperative and postoperative management, and to treat concomitant medical issues, with the goal of optimizing patient outcomes.
In this chapter we will provide an approach to the preoperative evaluation of patients undergoing thoracic surgery in light of the most recent evidence about the physiologic demands that these surgeries impose on patients. We will also briefly touch upon important issues during the intraoperative and immediate postoperative period that may influence outcome and may be of interest to pulmonologists who are involved in managing this interesting and challenging group of patients.
Definition and Impact of Postoperative Pulmonary Complications in Thoracic Surgery
Unsurprisingly, postthoracic surgery patients have a higher risk for PPCs ( Table 27-1 ) than patients having upper or lower abdominal surgeries (19% to 59% compared to 16% to 17% and 0% to 5%, respectively ).
|
One reason for the wide variability in the reported incidence of PPCs is the variability in defining what constitutes a PPC. For the purposes of this chapter we will define PPCs as shown in Table 27-1 . PPCs after lung resection are a major cause or contributing factor to postoperative deaths, accounting for up to 84% of all deaths. PPCs have a disproportionate impact on hospital costs as well, with one study finding that PPCs (defined by the authors as pneumonia, unplanned intubation, and failure to wean from mechanical ventilation) added more to hospital costs than either cardiovascular, infectious, or thromboembolic complications. An important study using data from the prospectively collected National Surgical Quality Improvement Program showed that a postoperative complication within the first 30 days after eight common surgeries (including lung resection) was independently associated with increased short-term (30-day) and long-term (1- and 5-year) mortality. Compared to wound complications, the other most common postoperative complication, PPCs had a disproportionate adverse impact on survival, as shown in Figure 27-1 .
Mortality and Morbidity in Thoracic Surgery: Can We Identify Patients at the Highest Risk for Adverse Outcomes?
Lung cancer is the most common indication for lung resection in the Western world, and, for patients with localized cancer, lung resection provides the best chance of a cure. The poor outcomes associated with nonoperative management of lung cancer together with improvements in surgical technique, anesthetic management, and postoperative management have resulted in larger numbers of sicker patients being offered surgery. It is now well established that the operative volume of the hospital where lung resection is performed has an important impact on outcomes. Mortality has also been reported to be lower when surgeries are performed by board-certified thoracic surgeons compared to nonspecialists, even though the board-certified thoracic surgeons often operate on patients with a higher burden of comorbid disease.
Consequently, there is significant interest in the ability of clinicians to be able to predict the risk for mortality and major morbidity following thoracic surgery, both to ensure quality and to provide patients with a reasonable estimate of the risk involved before undertaking surgery. Currently available tools are limited in their utility by the quality of the databases that are used to generate them. The Thoracoscore was developed in France using data obtained from more than 15,000 patients who were enrolled in a nationally representative thoracic surgery database (Epithor, developed by the French Society of Thoracic and Cardiovascular Surgery). The authors identified nine factors that predicted increased mortality: age, sex, dyspnea score, American Society of Anesthesiologists status, performance status, priority of surgery, diagnosis, procedure class, and comorbid disease. The model was subsequently validated in the United States and incorporated into the British Thoracic Society guidelines for risk assessment of patients with lung cancer ; it is available in a Web-based calculator ( http://sfctcv.fr/pages/epithor/thoracoscore_engl.php ). However, more recent studies have found the Thoracoscore to have a lower predictive power than reported earlier. Recently Kozower and colleagues reported on a model of perioperative risk for mortality and major morbidity from a database of more than 18,000 patients—the Society of Thoracic Surgeons (STS) General Thoracic Database. They found 12 risk factors to be associated with mortality, including American Society of Anesthesiologists status, the Zubrod functional status scale, renal dysfunction, induction chemoradiation, forced expiratory volume in the first second (FEV 1 ), body mass index (an increase was protective), male sex, and importantly, the type of surgery (pneumonectomy and bilobectomy had significantly higher mortality risks). An important limitation of both the Thoracoscore and the STS models is the lack of incorporation of diffusing capacity for carbon monoxide (D l CO ) data into their models because the majority of the patients in this database did not have measurement of diffusion capacity. In a subset of the STS database patients that did have D l CO values (almost 7900 patients), D l CO was found to be a strong independent predictor of mortality, in addition to the factors mentioned previously. Although these and other predictive models are still far from ideal, they do provide the clinician with objective data that may help supplement individual judgment in planning the best course of action for complex patients.
Assessment of Cardiac Risk in Patients with Lung Disease
Patients with lung cancer often have cardiovascular disease, because cigarette smoking is a shared risk factor for both diseases. The revised guidelines from the American College of Cardiology and the American Heart Association (ACC/AHA) are particularly helpful in evaluating cardiac risk ( Fig. 27-2 ).
The pathway for both emergent and elective surgeries in patients with active cardiac conditions (a myocardial infarction [MI] < 30 days ago, unstable angina, decompensated heart failure, significant arrhythmias, or severe valvular disease) is relatively straightforward. Most thoracic procedures fall into the intermediate risk category (with a combined risk for cardiac death and nonfatal MI between 1% and 5%), and will need a cardiac workup if they have both a poor functional capacity—defined as an exercise capacity of less than 4 metabolic equivalents (METs; 1 MET is approximately the level of effort required to climb up a flight of stairs) and more than one risk factor on the Revised Cardiac Risk Index (comprising a history of ischemic heart disease, cerebrovascular disease, heart failure, diabetes mellitus, and renal insufficiency). Most thoracic surgery patients in this group are likely to benefit from perioperative heart rate control. The ACC/AHA guidelines recommend noninvasive stress testing in this group of patients only if it is likely to change clinical management. Some patients may fall into relatively ill-defined areas in the guidelines, and they benefit from the opinion of an experienced cardiologist. A few specific situations are examined in the following section.
Indications for Coronary Revascularization
There has been a consensus over the last decade that invasive testing and coronary revascularization (either with bypass surgery or percutaneous coronary intervention) are unlikely to improve outcomes in noncardiac surgery unless the intervention is independently indicated for an acute coronary syndrome. In the Coronary Artery Revascularization Prophylaxis trial for patients undergoing vascular procedures, McFalls and associates randomly assigned 510 patients with significant coronary artery stenosis from among 5859 patients to either coronary artery revascularization or no revascularization before surgery. The short-term risk for death or MI or long-term outcomes was similar in both groups. It is important to remember that the study excluded patients with more than 50% left main vessel disease, a left ventricular ejection fraction of less than 20%, and severe aortic stenosis. Subsequent studies have supported these conclusions. (These studies were performed in the vascular surgery population, because this group of patients is at the highest risk for perioperative cardiac death/nonfatal MI. ) Although prospective data in thoracic surgery patients are lacking, it is not unreasonable to extrapolate these data to the thoracic surgery population. A number of retrospective studies show a higher incidence of major cardiac adverse events in patients who undergo lung surgery after the placement of either a bare metal stent or a drug-eluting stent. The difficult problem of a patient who needs both coronary revascularization and lung resection for cancer is best resolved on a case-by-case basis. The ACC/AHA recommended approach to nonemergent surgery in patients who have had a percutaneous coronary intervention is laid out in Figure 27-3 .
β-Blockade
The publication of the POISE (PeriOperative ISchemic Evaluation) trial in 2008 resulted in significant changes to the ACC/AHA recommendations on perioperative β-blockade that were laid out in the 2009 focused update on perioperative β-blockade. The POISE investigators randomized 8351 patients with or at risk for atherosclerotic disease to a fixed, relatively high dose of extended-release metoprolol (200 mg/day) compared to placebo, beginning with 100 mg 2 to 4 hours before surgery and continuing for 30 days. Fewer patients in the metoprolol group had an MI. However, more patients died in the metoprolol group than in the placebo group. These deaths were attributed to a higher incidence of strokes in the metoprolol group, perhaps secondary to a higher incidence of hypotension and bradycardia in these patients. The POISE study therefore raised the possibility that initiating β-blockers perioperatively in β-blocker-naive patients without titrating the dose to heart rate and blood pressure parameters may be associated with worse outcomes. Consequently, although continuing β-blockers for patients who are already on these medications remains a class I recommendation, the ACC/AHA has downgraded its recommendation for initiating β-blockers in high-risk vascular surgery patients from class I to class IIa and added the proviso that β-blockers should be titrated to heart rate and blood pressure. A similar recommendation applies to high-risk thoracic surgical patients. A new recommendation has been added advising against the initiation of high-dose β-blockers in the absence of titration (class III). Another issue with perioperative β-blockade is that clinicians are sometimes reluctant to use these drugs on elderly patients and in patients with chronic obstructive pulmonary disease (COPD) for fear of exacerbating airway obstruction. However, the data suggest that, not only is it safe to administer β-blockers to such patients, but that failing to use these drugs may be associated with worse outcomes.
Atrial Fibrillation Prophylaxis
Atrial fibrillation (AF), the most common arrhythmia following general thoracic surgical procedures, develops in 12% to 44% of patients after lung resection and esophageal surgeries. Patients with postoperative AF have increased lengths of hospital stay, increased medical costs, and increased risks for stroke, cognitive dysfunction, and death. Consequently, effective prophylaxis to prevent postoperative AF is an important goal. The recommendations in this section are derived from clinical practice guidelines on AF prophylaxis by the STS. As in the ACC/AHA guidelines regarding perioperative β-blockade, it is important to continue β-blockers in patients already taking them to reduce the risk for β-blocker withdrawal. The data here come chiefly from the cardiac surgery literature, where propranolol withdrawal has been associated with new-onset AF. The dose of β-blockers in these patients should be adjusted, and “hold-parameters” should be in place because many of them are likely to have epidural catheters in place for analgesia, increasing the risk for hypotension and bradycardia. In patients not taking β-blockers preoperatively and who do not need to be started on them based on the ACC/AHA recommendations described earlier, calcium channel blockers such as diltiazem have proven to be effective at AF prophylaxis. Amiodarone is effective at AF prophylaxis ; however, in one study, amiodarone was associated with an increase in the incidence of acute respiratory distress syndrome—particularly if given after a pneumonectomy. The use of amiodarone may also be associated with a higher incidence of postoperative acute respiratory distress syndrome and, if there is preexisting pulmonary disease, a higher risk for amiodarone-induced pulmonary toxicity. On the other hand, two more recent studies by Tisdale and coworkers in patients after pulmonary resection and after esophagectomy did not find increased pulmonary toxicity with amiodarone, although the doses used were lower. For the present the STS recommends the use of amiodarone only in very limited circumstances and in carefully controlled doses. It recommends against the use of amiodarone for AF prophylaxis in patients undergoing pneumonectomy. Digitalis and flecainide should not be used for the prophylaxis of AF. Therefore the current recommendations for AF prophylaxis are to continue β-blockade throughout the perioperative period if the patient is already taking β-blockers. If the patient is not taking them, then calcium channel blockers would be recommended for prophylaxis.
Special Patient Populations
Patients Who Smoke
Smoking is one of the most important causes of preventable death in the world. Studies suggest that smoking is responsible for almost half a million deaths and more than $200 billion in health care costs and lost productivity in the United States alone. At the time of diagnosis of lung cancer, it is estimated that up to 18% of patients have never smoked, 58% are former smokers, and 24% to 40% are current smokers. About 20% of patients smoke at the time of cancer surgery, and half of these continue to smoke afterward. There are extensive data documenting the adverse effects of smoking on both pulmonary function and clinical outcomes. Smoking decreases macrophage function, impairs vascular endothelial function, decreases coronary reserve, and places patients at increased risk for tachycardia, hypertension, and ischemia. In addition, smoking increases the risk for arterial desaturation and laryngospasm during anesthesia. Smokers also have increased levels of carboxyhemoglobin (between 3% and 15%) that both compromise the total oxygen content of blood and impairs oxygen release at the tissue level by shifting the oxygen dissociation curve to the left. Recent multicenter outcome studies have confirmed the association between smoking and worse perioperative outcomes. Using data from the American College of Surgeons National Surgical Quality Improvement Program database, Turan and colleagues showed that smokers had an approximately 30% higher adjusted odds of postoperative mortality and major complications compared to nonsmokers. Another large database analysis using the Veterans Affairs Surgical Quality Improvement Program (with more than 390,000 patients) showed that past smokers have an approximately 20% higher odds of postoperative mortality or major complication. In an STS database with almost 8000 patients undergoing lung resection surgery for cancer, Mason and associates showed that smokers have a significantly increased risk for hospital mortality and pulmonary complications (hospital mortality 1.5% in smokers compared to 0.39% in nonsmokers), and that smoking cessation gradually mitigated these risks. However, they could not identify an optimal interval for smoking cessation.
There is some controversy over the timing of smoking cessation because some studies suggested an increase in PPCs in patients who quit smoking less than 8 weeks before surgery. The risk seemed to be higher in patients who quit closest to the day of surgery. However, more recent studies have not been able to show this increased risk in recent quitters. It does seem that the benefits increase with the duration of smoking cessation. A meta-analysis of 6 randomized trials and 15 observational studies of smoking cessation demonstrated a relative risk reduction of 41% in PPCs. In addition, each week of cessation increased the magnitude of the effect by 19%. Another recent meta-analysis examined postoperative outcomes in patients who quit smoking less than 8 weeks before surgery compared to those who continued to smoke. They reported that there was no evidence of any positive or negative effect of late smoking cessation on PPCs. Given these data, it seems prudent to encourage smoking cessation irrespective of surgical timing. The U.S. Public Health Service recommends that physicians strongly advise smokers to quit smoking because physicians’ advice to quit has been associated with increased abstinence rates. Effective interventions include medical advice and pharmacotherapy such as nicotine replacement (which is generally safe in the perioperative period) and non-nicotine options such as bupropion and varenicline. Varenicline seems to be the most effective pharmacologic intervention to promote abstinence from smoking. It is important to note that the Food and Drug Administration has issued boxed warnings for both drugs because of reports of increased psychiatric symptoms or suicidal ideation.
Asthma
Well-controlled asthma is unlikely to be a risk factor for either intraoperative or postoperative complications (for clinical aspects of asthma, see Chapter 42 ). However, poorly controlled asthma, evidenced by active wheezing, can increase PPCs. The combination of inhaled β-agonists such as albuterol and inhaled steroids with long-acting β-agonists is often very useful to achieve control of symptoms before surgery. If bronchospasm persists, a short course of low-dose systemic steroids may be considered and does not seem to have an impact on postoperative wound healing. It is important to keep in mind that symptoms suggestive of asthma could in fact be due to other pathologic conditions such as pulmonary carcinoid, tracheal stenosis, and other endobronchial tumors, and difficult-to-treat asthma should prompt a workup for these rarer conditions.
Chronic Obstructive Pulmonary Disease
COPD is increasingly being recognized as a disorder with both pulmonary and extrapulmonary manifestations, including an increased incidence of lung cancer (for clinical aspects of COPD, see Chapter 44 ). Between 50% and 80% of patients with lung cancer have COPD, and the association is independent of the intensity of smoking. Unlike well-controlled asthma, COPD is associated with an increased risk for PPCs. As a general rule, the management of COPD before surgery is the same as for patients who will not have surgery. The optimal management of COPD, including surgical approaches such as lung volume-reduction surgery, is discussed elsewhere in this textbook. (For clinical discussion of lung volume-reduction surgery, see Chapter 44 .) Preoperative pulmonary rehabilitation may improve perioperative outcomes in these patients (see later).
Obesity and Obstructive Sleep Apnea
More than 65% of Americans are overweight or obese. Obesity is associated with multiple comorbidities, including diabetes and heart disease. There is evidence that obesity is an independent risk factor for worse perioperative outcomes in many types of surgery —however, in thoracic surgeries, the effect of obesity on outcome is less clear. As mentioned earlier, a high body mass index appeared to be protective in the STS database risk model study by Kozower and coworkers. A recent propensity-matched comparison of survival after lung resection in high versus low body mass index groups also suggested a protective effect of obesity in the setting of lung cancer. In another retrospective study, Dhakal and associates did not find any adverse effect of obesity on postoperative morbidity or mortality following lung resection. It is unclear why obesity has no harmful or even a protective effect for perioperative outcomes—it is tempting to speculate that, in the setting of lung cancer, obesity is a marker of better nutritional and immune status, leading to better outcomes.
In contrast, patients with obstructive sleep apnea (OSA) have a higher incidence of (mainly respiratory) PPCs than those without OSA (for discussion of OSA, see Chapter 88 ). In those undergoing hip or knee replacements, 39% of patients with OSA developed PPCs or cardiac complications compared to 18% of patients without OSA; 24% of the patients with OSA required intensive care unit admissions compared to 9% of patients without OSA. Few investigations have identified the perioperative risks for OSA surgical patients, but left heart failure or right heart dysfunction due to pulmonary hypertension could be responsible and should be sought. In particular, pulmonary hypertension is associated with significant perioperative risk. Because cardiovascular dysfunction in OSA patients can be modified by treatment, it is important to identify patients with OSA and begin treatment. The use of standardized questionnaires such as the STOP ( Snoring, Tiredness during daytime, Observed apnea, high blood Pressure ) and STOP-Bang ( Body mass index, Age, Neck circumference, Gender ) forms have been validated in the perioperative setting and provide a simple and useful screening tool for OSA. Current studies suggest that diagnosis and treatment of OSA can improve postoperative outcomes.
Lung Resection
The preoperative evaluation of patients for lung surgery differs from that for other surgeries. Not only are these patients more likely to have pulmonary disease, but the postoperative function may be impaired permanently because of the effects of the operation. In most other operations, the evaluation is primarily to optimize the anesthetic and surgical plan and to prepare the patient, but in pulmonary resection, the evaluation determines whether or not the surgery can proceed.
Predicting Postoperative Function
The challenge in assessing patients for pulmonary resection is predicting the postoperative course, both for the acute changes and postoperative morbidity, and for the final postoperative status. Depending on the type of operation, the eventual status may be better or worse than the current status ( Table 27-2 ).
| Worse Function | Better Function |
|---|---|
|
|
The typical method used to predict postoperative function is to use a regional ventilation-perfusion scan (V/Q scan using a radioactive gas) to estimate the proportion of lung function expected to be lost. This assessment was based on older studies that showed an excellent correlation. More recent work, however, shows the nuances of postoperative functional prediction. For example, in one prospective study, the immediate postoperative function was significantly worse than the predicted postoperative FEV 1% , and, although there was recovery in the first week, the FEV 1% did not reach the level predicted ( Fig. 27-4 ). In a study that followed patients over 3 months after discharge for lobectomy, pulmonary function was found to reach predicted levels at 1 month and in fact to surpass them at 3 months by approximately 10% ( Fig. 27-5 ).
In an STS report on their large registry of pulmonary resections (18,000 operations, 111 centers), pulmonary resection was associated with a 2% overall mortality. Significant predictors of morbidity and mortality included low preoperative FEV 1 , renal failure, high American Society of Anesthesiologists class, steroid use, induction chemoradiation, and especially the type of resection, with pneumonectomy carrying more risk than lobectomy. Thoracoscopic surgery was somewhat protective.
Guidelines for Lung Resection
There are several consensus published guidelines for assessing a patient’s suitability for pulmonary resection. Typically some measure of pulmonary function is used. Especially for more extensive resection (e.g., pneumonectomy), the possibility that the diseased lung may not be contributing a proportional share of function and thus may be safely resected is addressed by predicted postoperative (PPO) values. The estimate is usually calculated using ventilation-perfusion scans.
The American Thoracic Society/American College of Chest Physicians guidelines review cardiopulmonary exercise testing in detail but touch on preoperative evaluation only briefly. They recommend FEV 1 and D l CO criteria as the primary measures and, when these values are borderline, cardiopulmonary exercise testing has utility. The British Thoracic Society recommends an initial FEV 1 evaluation and more detailed analysis in marginal cases.
Principles of Cardiopulmonary Exercise Testing—Strengths and Weaknesses (see Chapter 26 )
Cardiopulmonary testing increases the demands of both the cardiovascular and respiratory system, but, given that the respiratory capacity is utilized only at peak exercise in healthy patients, exercise is usually limited by cardiac factors before respiratory function limits exercise capacity in normal individuals. Exercise testing in patients is associated with severe complications in 1 : 10,000, and testing is contraindicated in patients with severe cardiac disease (e.g., cardiomyopathy, unstable angina, arrhythmias), uncontrolled asthma, significant infections, metabolic derangements, and uncooperative patients.
Not all investigations have found an association between exercise performance and a risk for perioperative complications. Some of the issues with oxygen consumption measurements are that they are affected by muscle mass, body size, and level of fitness. Therefore using maximum oxygen consumption ( 
) may bias the preoperative assessment against older and obese individuals.
Some other issues that need to be considered are that 
predictions based on preoperative measurements have significantly overestimated the degree of exercise capacity loss after an operation. Although the PPO 
did not accurately predict postoperative exercise capacity, it was nonetheless the best predictor for postoperative morbidity and mortality.
It is also unclear whether preoperative exercise performance has any influence on long-term clinical outcomes. In one report, 68 of 86 patients categorized as high risk, with preoperative measured 
less than 15 as well as having PPO FEV 1 of less than 33%, underwent lung resections. Although they had higher morbidities, the high-risk group had a low mortality of only 4% (3/68). Furthermore, the 68 high-risk patients who underwent resection had better 5-year outcomes than similar high-risk patients who were denied surgery. Patient outcomes after surgery are influenced by improvements in surgical techniques and postoperative care, as well as nonsurgical therapies for early-stage lung cancer, so that guidelines and other recommendations need to be revisited on a regular basis.
Preoperative Predictors
Although a variety of testing schemes have been used for predicting postoperative complications after lung resection, 
is currently considered the best predictor. In a 2007 meta-analysis of 
as a predictor that pooled 14 studies and nearly 1000 patients, a higher 
(20 versus 16 mL/kg/min) was found in the group without PPCs. D l CO and FEV 1 were also somewhat higher in the group without complications; however, they were found to be less useful clinically than the 
. 
was also evaluated in a Cancer and Leukaemia Group B study that included 400 patients and a two-tier resectability evaluation. The first group based on FEV 1 and FEV 1 -PPO criteria were considered “low risk.” Patients failing FEV 1 criteria were further stratified by a value of 
greater than 15 mL/kg/min into “high risk” and “only at physician discretion.” The mortality and morbidity rate was higher for the high-risk group compared to the low-risk group, but overall survival was much better in all operative patients. There was also a tendency for the higher-risk patients to get a less extensive resection.
Recent Guidelines for Preoperative Testing in Lung Cancer Patients
A recent evidence-based clinical practice guideline supported by the American College of Chest Physicians was published in the third edition of their supplement concerning the diagnosis and management of lung cancer. Twelve recommendations were made and given grades for the level of existing evidence. These recommendations for patients with lung cancer being considered for surgery are listed with our comments:
- 1.
The preoperative assessment should be conducted by a multidisciplinary team. [This is reasonable, given that the patient will be cared for by a pulmonologist before and after surgery but will be cared for intraoperatively and perioperatively by an anesthesiologist and a surgeon.]
- 2.
Evaluations should be done for patients regardless of age. [Many perioperative outcome studies have documented that the physiologic age is more important than the duration of life.]
- 3.
Cardiovascular risk should be managed according to existing guidelines for noncardiac surgery. [See earlier discussion of indications for coronary revascularization.]
- 4.
FEV 1 and D l CO should be measured, and both PPO FEV 1 and PPO D l CO should be calculated. If both of these PPO measurements are greater than 60% predicted, no further tests are recommended.
- 5.
However, if they are less than 60% predicted but above 30% predicted, a low-technology exercise test should be performed.
- 6.
If either of the PPO FEV 1 or the PPO D l CO are less than 30% predicted, a formal cardiopulmonary exercise test should be done with measurement of the
.
- 7.
Similarly, if a patient is considered for surgery who walks less than 25 shuttles [<400 m] on the shuttle walk test or climbs less than 22 m on a stair-climbing test, a formal cardiopulmonary exercise test with measurement of
is also recommended.
- 8.
If patients with lung cancer being considered for surgery have a
less than 10 mL/kg/min or less than 35% predicted, they should be counseled about minimally invasive surgery, sublobar resections, or nonsurgical treatment options. [See earlier comments regarding patients who were deemed high risk due to their exercise test results but had low perioperative mortality. ]
- 9.
In patients with lung cancer who are being considered for surgery who undergo neoadjuvant therapy, it is suggested that repeat pulmonary function testing with diffusing capacity be performed after completion of the neoadjuvant therapy. [These patients may be more prone to problems with drugs such as amiodarone—see earlier discussion.]
- 10.
In patients with lung cancer in an area of upper lobe emphysema who are candidates for lung volume-reduction surgery, combined lung volume-reduction surgery and lung cancer resection is suggested.
- 11.
In all patients with lung cancer being considered for surgery who are actively smoking, tobacco dependence treatment is recommended.
- 12.
In patients with lung cancer being considered for surgery and deemed at high risk [PPO FEV 1 or the PPO D l CO less than 60% predicted and
less than 10 mL/kg/min or less than 35% predicted], preoperative or postoperative pulmonary rehabilitation is recommended.
Risk Modifications
Smoking Cessation
The most obvious risk modification available to patients is smoking cessation (see “ Patients Who Smoke ,” earlier). Interestingly, there had been a concern that there would be increased secretions and reactivity in the immediate withdrawal period. In the most relevant study of the timing of cessation before thoracotomy, Barrera and coworkers found no difference in mortality or morbidity based on the duration of cessation and thus found no evidence of a paradoxical increase in complications among recent quitters. Smoking cessation is always to be encouraged before thoracotomy.
Surgical Approach
The surgical approach has implications for postoperative outcome. Based on the STS database of over 1000 patients undergoing lobectomy, a thoracoscopic approach had improved outcomes, including less AF, need for transfusion, and prolonged air leaks and shorter length of stay. Similar findings have recently been found in a propensity score-matched study in which video-assisted thoracoscopic lobectomy was associated with a lower incidence of pulmonary complications (1.1% versus 12.1%), a shorter length of stay, and shorter operation time than lobectomy via thoracotomy. Studies are now also showing that video-assisted thoracoscopic lobectomy offers similar long-term outcomes: in a study of patients with clinical stage I non–small cell lung cancer, video-assisted thoracoscopic lobectomy offered similar overall survival and disease-free survival as lobectomy by thoracotomy.
Exercise Training (For Clinical Exercise Testing, see Chapter 26 )
It is interesting that 
is a modifiable measurement. Preoperative interventions can improve exercise capacity in lung resection candidates. For COPD patients about to undergo lobectomy, Stefanelli and colleagues found that “high-intensity training” improved 
by 20%, an improvement that persisted for at least 2 months postoperatively. Static measures of lung function (D l CO and FEV 1 ) were unchanged by the training and decreased after surgery, as expected. In another study of 27 COPD patients facing lung resection, pulmonary rehabilitation improved 
(from 13 to 19 mL/kg/min) , arterial P o 2 , and even FEV 1 to resectable levels, and subsequent resection was tolerated with only a 15% short-term morbidity.
Even if rehabilitation was not initiated before resection, there is evidence that postoperative exercise function can be improved with pulmonary rehabilitation after lung resection. Of course, merely improving the 
does not guarantee better outcomes. The hope is that better function will allow a safer perioperative course in marginal patients and will improve overall postoperative function. However, the current literature is not yet sufficient to reach this conclusion. Certainly there is no hint of harm from pulmonary rehabilitation in this population, and the therapy matches recommendations for those in the general population with impaired pulmonary function.
Intraoperative Management
There is increasing recognition that intraoperative management can affect postoperative outcomes. Lung-protective ventilation, specifically smaller tidal volumes, use of positive end-expiratory pressure, and periodic recruitment maneuvers have significantly reduced postoperative pulmonary complications, need for reintubation, and length of stay in elective abdominal surgery. The extension of these protective maneuvers to thoracic surgery is not simple. Pulmonary surgery usually benefits from one-lung ventilation, with the lung on the operative side deflated. However, because the pulmonary circulation is not routinely occluded, there is an obligatory pulmonary shunt and a real chance of hypoxemia during surgery. Traditionally, large tidal volumes, no positive end-expiratory pressure, and high fractional concentration of oxygen in inspired gas were used to combat hypoxemia. However, there is now evidence that such management can cause postoperative pulmonary complications. In a study designed to test different ventilatory strategies during one-lung ventilation for lung resection surgery, a protective lung ventilatory strategy was associated with significantly less postoperative lung dysfunction ( Fig. 27-6 ).
