Chapter 27

Anesthesia for Thoracic Surgery

Thoracic surgery is greatly facilitated by the contribution of anesthesia care, which can isolate the lungs and create a quiet surgical field. Although this procedure is an advanced technique of airway management, it has been in existence almost as long as tracheal intubation itself. In 1928, Guedel, Magill, Waters, and other pioneers first achieved closed endotracheal anesthesia. The treatment of tuberculosis and empyema, however, required isolation of the infected lung, and in 1931 Joseph Gale and Ralph Waters first described “closed endobronchial anesthesia.” Bronchial blockade for selective ventilation and lung isolation was reported by Magill in 1936. In 1950, Björk and Carlens are credited with the first use of a double lumen endotracheal tube for thoracic surgery, the bronchospirometric double-lumen tube, which Carlens described the year before.1,2 Double-lumen tubes evolved in design, and the use of the Fogarty embolectomy catheter for bronchial blockade gave way in 1982 to the Univent tube invented by Inoue and colleagues.

Development of airway devices continues, and anesthetists caring for patients undergoing thoracic surgery must be skilled at insertion, maintenance, and monitoring of these devices for proper function. However, the process of lung isolation facilitates the surgical procedure but compounds a central concern in thoracic anesthesia: maintaining effective gas exchange in the face of ventilation and perfusion mismatch. General anesthesia creates atelectasis, which is compounded by muscle relaxants and lateral positioning, but ventilation and perfusion are further mismatched when the thorax is opened, and ceasing ventilation of one lung is the final insult. The cardiac output persists, but only one lung is ventilated. Fortunately, physiologic processes combat the inherent shunt, and anesthetic management is geared toward supporting those processes while fostering oxygenation through various ventilation modalities. The general incidence of hypoxemia is 5% to 10% during one lung ventilation (OLV).³

Besides managing a complex tracheal tube, the anesthetist must also be cognizant of the effects of underlying disease as it relates to management of ventilation (bullous disease contraindicating nitrous oxide [N₂O]), interactions with anesthetic drugs (small cell carcinoma being associated with myasthenic syndrome), and even concerns about oxygen toxicity (in patients treated with bleomycin and other chemotherapeutics).

Preoperative Preparation

Bronchogenic carcinoma is the leading cause of cancer deaths in the United States.⁴ Because lung cancer is most often discovered only once symptomatic (i.e., advanced), even with aggressive multimodal treatment, prognosis is poor. A review of patients with non–small cell carcinoma found a 5-year survival rate of less than 7%.⁵ However, resection of affected lung tissue offers a better prognosis than radiation and chemotherapy, and so cancer is a common indication for lung surgery. There is a strong association between lung cancer and chronic obstructive pulmonary disease (COPD), such that the incidence of lung cancer is four times higher among COPD sufferers than it is in the general population, and underlying COPD almost doubles the 3-year mortality of lung cancer.^6–8 Because many patients presenting for lung surgery will have complex underlying pathology, evaluating respiratory function and predicting postresection function are crucial to anticipating the patient’s intraoperative and postoperative care.

The surgical risk assessment for patients in need of pulmonary resection surgery focuses on the risk of potential postoperative complications and whether postoperative pulmonary function will be sufficient to allow reasonable quality of life. Between 20% and 30% of patients with lung cancer are found to be surgical candidates,⁹ and almost 40% of them are disqualified based on poor lung function alone.¹⁰ The anesthetic risk assessment specific to pulmonary surgery focuses on how the underlying pathology will challenge the maintenance of adequate gas exchange and general homeostasis under OLV and the potential for postoperative respiratory failure to make weaning and extubation difficult. However, mortality from unresected carcinomas is sufficiently high that the risks of postoperative complications would need to be extraordinarily high before they would preclude surgery. COPD is a disease requiring years of development. Considering the aging nature of our population and the increase in the incidence of obesity, it is not surprising that lung resection surgeries are now being performed on more patients who have end-stage COPD, morbid obesity, or who are of advanced age.^11,12 Evidence shows that these patients can be treated safely. The American College of Chest Physicians does not set a maximal age cutoff for pulmonary surgical candidacy.¹³ Changes in surgical techniques, especially the use of video-assisted thoracoscopic surgery, have markedly decreased the incidence of postoperative pulmonary complications and will necessitate a reevaluation of the testing required for preoperative assessment.¹⁴ For smaller lung resections, minimal preoperative evaluations of cardiac disease, gas exchange, and oxygenation may suffice. Given the large physiologic changes that occur after pneumonectomy, complete pulmonary function testing, as well as cardiac testing, may still be reasonable in these patients.

Fear of creating pulmonary insufficiency by lung resection is an important concern, and numerous studies have attempted to determine the lowest limit of pulmonary function that will allow surgery to be safely performed. Research findings are limited in their ability to predict particular complications, though. Studies performed to predict postoperative pulmonary complications after lung resection demonstrate that patients develop both pulmonary and cardiac complications such as dysrhythmias, myocardial infarction, pulmonary embolism, pneumonia, and empyema. These complications influence the duration of mechanical ventilation and outcome; however, none of these complications can be predicted by preoperative studies of pulmonary function. Box 27-1 presents some commonly used preoperative assessment criteria for lung resection.

BOX 27-1 Summary of Preoperative Factors That Predict Postoperative Complications

Factors That Characterize Average Risk

• FEV₁ > 2 L or 80% of predicted

• PPO FEV₁ > 80% of predicted

• PPO FEV₁ + PPO DLCO both > 40%

• O₂max > 15 mL/kg/min

• Ability to climb 3 flights of stairs

Factors That Characterize Elevated Risk

• FEV₁ <2 L or < 40% of predicted

• PPO FEV₁ < 40% of predicted

• PPO DLCO < 40% of predicted

• PPO product, (FEV₁ × DLCO) < 1650

• O₂max < 10 mL/kg/min

• Inability to climb 1 flight of stairs

• Oxygen desaturation > 4% during exercise

DLCO, Carbon monoxide diffusion in the lung; FEV₁, forced expiratory volume in 1 second; PPO, predicted postoperative; O₂max, maximum volume of oxygen utilization.

History and Physical Examination

Cancer patients who undergo lung resection typically have a history of multiple risk factors and signs of respiratory disease. Risk factors include cigarette smoking, air pollution, and industrial chemical exposure. The smoking history that is so common to these patients should also lead to suspicion of ischemic heart disease or hypertension. Therefore, patients must be evaluated for exertional dyspnea, productive cough, hemoptysis, cyanosis, poor exercise tolerance, and chest pain.

The presence of ischemic cardiac disease that is severe, unstable, or associated with dysrhythmias should indicate consultation with a cardiologist to help mitigate risk of cardiac complications. Because severe COPD may significantly limit physical activity, a provocative dobutamine or other type of cardiac stress test may be helpful to identify coronary insufficiency.¹⁵ The use of perioperative beta blockade is controversial, and nonselective agents may be particularly detrimental if they inhibit bronchodilation. However, cardioselective beta-blockers have been found beneficial in COPD patients undergoing vascular surgery.¹⁶ Beta blockade may be considered to reduce cardiac risk,¹⁷ or at the least, patients currently taking beta-blockers should continue their regimen throughout the surgical encounter. The surgical plan must be carefully considered for patients with coronary diffusion defects that show greater than 20% reversibility. In cases of high-risk cardiac disease, lung surgery may be delayed for 6 weeks to allow coronary artery bypass first.¹⁵ With the popularity of off-bypass coronary revascularization, the two procedures can more easily be performed in a single surgical encounter.¹⁸

Lung cancer patients should be assessed for effects of the primary tumor, as well as effects of secondary pathologies and side-effects of therapy (Table 27-1). Supine dyspnea may result from COPD or compression of the airway by a mediastinal mass. A high index of suspicion should also be maintained for hormonal abnormalities, because many lung tumors cause paraneoplastic syndromes characterized by secretion of endocrine-like substances such as adrenocorticotropic hormone,¹⁹ antidiuretic hormone,²⁰ serotonin, parathyroid-like hormone,²¹ and insulin-like growth factor,²² causing a variety of metabolic abnormalities.²³ Cushing disease may lead to metabolic alkalosis, hypokalemia, and hyperglycemia.^24,25 Hypercalcemia occurs in 10% to 25% of lung cancer patients and is related to parathyroid-like hormone, increased calcitriol, or overactivity of osteoclasts.²⁶ The finding of hypercalcemia carries a very poor prognosis. Clinical signs may include polyuria, polydipsia, confusion, vomiting, abdominal cramps, bradycardia, and mental status changes.

TABLE 27-1

Anesthetic Considerations in Lung Cancer Patients: “The 4 Ms”

Preoperative Parameter	Clinical Assessment
Mass effects	Obstructive pneumonia, lung abscess, superior vena cava syndrome, tracheobronchial distortion, Pancoast syndrome, recurrent laryngeal nerve or phrenic nerve paresis, chest wall or mediastinal extension
Metabolic effects	Lambert-Eaton syndrome, hypercalcemia, hyponatremia, Cushing syndrome, syndrome of inappropriate antidiuretic hormone
Metastases	Particularly to brain, bone, liver, and adrenal glands
Medications	Chemotherapy-induced lung changes bleomycin = pulmonary toxicity, doxorubicin = cardiac toxicity, cisplatin = renal toxicity

Modified from Slinger PP, Johnston MR. Preoperative assessment and management. In: Kaplan JA, Slinger PD, eds. Thoracic Anesthesia. 3rd ed. Philadelphia: Churchill Livingstone; 2003.

Neuroendocrine tumors comprise 20% of lung cancers, and 5% of them produce carcinoid syndrome.²⁶ In those patients, histamine-stimulating drugs and adrenergics may precipitate the flushing, hypotension, and tachyarrhythmias related to serotonin release. Paraneoplastic neurologic syndromes represent autoimmune dysfunctions associated with cancers; 1% to 2% of patients with small cell cancer develop Lambert-Eaton myasthenic syndrome (LEMS).²⁷ In this syndrome, antibodies are formed against the voltage-gated calcium channels,²⁸ causing weakness and sensitivity to nondepolarizing muscle relaxants. The usual presentation is first with autonomic dysfunction (80% of patients with LEMS) such as orthostatic hypotension, and then weakness that progresses upward from the legs.²⁹ In contrast to myasthenia gravis, because LEMS involves dysfunction of the calcium channels, repetitive stimulation or activity improves function (as more acetylcholine is mobilized), and anticholinesterase drugs are not an effective treatment. Treatment is with immunoglobulin, corticosteroids, or with 4-diaminophridine (DAP), which opens potassium channels and increases calcium in the nerve terminal. Other autoimmune channelopathies can affect voltage-gated potassium channels and prolong acetylcholine release, causing a myotonia, or they can affect autonomic ganglia, causing orthostatic hypotension and arrhythmias.³⁰

Physical examination findings in COPD are commonly barrel chest deformity, accessory muscle use or paradoxical breathing movement, pursed-lip breathing, and tympanic percussion notes on the chest. Auscultation reveals rhonchi or wheezing. Signs of cor pulmonale include jugular vein distention or peripheral edema, split S₂ heart sound, pulmonary or tricuspid valve insufficiency murmurs, and rales auscultated over the lungs.³¹ Nutritional status, commonly compromised in patients with cancer, is also important to note because hypoalbuminemia and malnutrition are associated with increased postoperative complications such as pneumonia.³² Box 27-2 lists important elements of the preoperative evaluation.

BOX 27-2 Preoperative Evaluation of Patients for Pulmonary Surgery

1. Evaluate comorbidities:

• Smoking-related complications

• Cor pulmonale

• Effects of paraneoplasms

• Cardiovascular disease (unstable angina, MI within 6 weeks, or significant dysrhythmias = high risk of cardiac complications)

• Treatment side effects (particularly from cytotoxic drugs and radiation)

2. ECG and chest x-ray for signs of cardiovascular dysfunction and effects of lung pathology

3. Laboratory assessment of electrolytes, blood count, albumin, and renal function indicators

4. Lung function testing: 80-40-15 rule:

• FEV₁ and DLCO > 80% predicted = no additional testing needed; if < 80 or dyspnea present, diffusing capacity and postoperative function should be predicted

• PPO FEV₁ and DLCO < 40% predicted = increased risk; exercise testing should be evaluated

• O₂max < 15 mL/kg/min = increased risk

5. For COPD, consider blood gas and response to bronchodilators.

COPD, Chronic obstructive pulmonary disease; DLCO, carbon monoxide diffusion in the lung; ECG, electrocardiogram; FEV₁, forced expiratory volume in 1 second; MI, myocardial infarction; PPO, predicted postoperative; O₂max, maximum volume of oxygen utilization.

Diagnostic Data

Chest Radiograph

The chest radiograph should be obtained to assess for associated disease and complications of COPD (tumor infringement on airway or vascular structures, bullous disease, congestive heart failure, or pneumothorax). The radiograph does not provide abundant information regarding the degree of COPD, but findings characteristic of COPD include hyperinflation, increased anteroposterior diameter, and diaphragm flattening. Bullae of emphysema may be present, and infection or pleural effusions may be noted preoperatively and treated to improve the postoperative course. The locations of masses can be identified. In some patients, it can be ascertained whether lesions compress mediastinal structures, cause tracheal shift, or invade the airway. This information is important to predict whether intubation will be difficult, whether induction of anesthesia could cause collapse of the airway, or whether surgical dissection may be difficult and potentially involve excessive bleeding. Evidence of increased pulmonary vascular resistance (PVR) resulting from compression of the vascular bed increases the likelihood of right ventricular failure and worsens the prognosis after lung resection. Relevant signs include prominent pulmonary arteries with rapid tapering of the vasculature, and a widened right heart border.

Electrocardiogram

Assessment of the electrocardiogram (ECG) is useful to assess signs of right ventricular hypertrophy. In such a case, the ECG shows a tall R wave in V₁ (greater than 6 mm), R/S ratio greater than 1 in lead V₁ and a ratio less than 1 in V₆, along with a right-axis deviation and diminished amplitude limb leads.33–35 Right atrial hypertrophy causes the initial component of a biphasic P wave in lead V₁ to be larger than the second component. Strain characteristics such as S-T segment depression and T-wave inversion, as well as incomplete or complete right bundle branch block, may be observed. The ECG has excellent specificity for identifying left ventricular hypertrophy (LVH), but less sensitivity for detecting it.^36,37 Therefore, if ECG criteria are inconclusive, echocardiography may be helpful to further elucidate the status of the right ventricle. Findings of pathologic Q waves and evidence of left ventricular hypertrophy preoperatively correlates with an increased incidence of postoperative ischemia and infarction.³⁸

Echocardiogram

Expected echocardiographic findings are increased thickness of the right ventricular free wall, chamber enlargement, pulmonary artery systolic hypertension, septal shift, tricuspic regurgitation, and decreased right ventricular ejection fraction. Increased PVR and right ventricular strain cause concern in patients undergoing pneumonectomy or extensive partial resection because of the added resistance produced by clamping the vasculature of one lung in the surgical procedure. Echocardiography is the best initial tool for assessing pulmonary hypertension, but additional studies with pulmonary angiography, ventilation-perfusion scintigraphy, computed tomography, and magnetic resonance imaging also may be used for more in-depth evaluation.³⁹

Laboratory Assessment

Measurement of preoperative room air arterial blood gases should be considered for patients with COPD, and is useful in guiding the weaning of O₂ and ventilation postoperatively. Carbon dioxide (CO₂) retention with an arterial partial pressure (Paco₂) greater than 45 mmHg is an indicator of poor ventilatory function. However, hypercapnia is not a reliable predictor of increased risk of perioperative pulmonary complications.⁴⁰ Preoperative hypoxemia (Spo₂ less than 90%) and particularly desaturation during exercise may be predictive of increased complications after thoracic surgery;^41,42 however, in general, blood gas analysis is not a reliable tool in predicting postoperative pulmonary complications,⁴³ and the correlation between desaturation during exercise and postoperative complications is not a consistent finding.⁴⁴

Hypoalbuminemia is the most common laboratory finding, which serves as an important predictor of pulmonary complications. Numerous studies have demonstrated increases in postoperative pulmonary complications among patients with low serum albumin levels (generally less than 3.6 g/dL).45–48 This factor increases risk as much as 2.5 times,⁴⁷ and albumin level maintenance is a measured factor in the American College of Surgeons’ National Surgical Quality Improvement Program (NSQIP).⁴⁹ The blood urea nitrogen is also a factor identified by NSQIP data as a predictive factor for pulmonary complications when the level is greater than 22 mg/dL.^47,48

Other laboratory analysis of interest includes renal function indicators (particularly for patients treated with nephrotoxic drugs, such as methotrexate, gemcitabine, cisplatin),50,51 sodium (related to syndrome of inappropriate antidiuretic hormone),²³ and calcium (due to parathyroid hormone-like protein).^21,52

Pulmonary Function Tests

Patients presenting for lung resection should undergo pulmonary function testing to assess for airflow limitation, diffusion defect, and cardiopulmonary reserve.⁵³ Assessment should include the response to bronchodilators for patients who demonstrate obstructive disease (see Figure 26-19). In that case, assessment of spirometry should be based upon values obtained postbronchodilator therapy, because these would represent the patient’s potential lung function once optimized on medications. No single measurement provides the overall risk assessment. For example, although the forced expired volume in 1 second (FEV₁) is the most prevalent spirometric measurement, one case series of 100 thoracic surgery patients with very low FEV₁ values (less than 35%) demonstrated a low rate of mortality and ventilator dependence (but patients did show a prolonged duration of hospitalization and air leak).⁵⁴ In another series of 109 elderly patients, stair-climbing ability was better correlated (inverse relationship) to postoperative cardiopulmonary complications than was forced vital capacity (FVC) or predicted postoperative (PPO) FEV₁.⁵⁵ Therefore, a multimodal approach must be taken, considering airflow (PPO FEV₁), parenchymal function (carbon dioxide diffusion in the lung [DLCO]), and cardiopulmonary reserve (O₂max).The general cutoff points indicating increased risk among these parameters is below 40% for predicted postoperative FEV₁ and DLCO, and 15 mL/kg/min for O₂max.

It should be noted that guidelines to assess surgical candidacy are not intended to dictate candidacy for anesthesia. If indications are that the patient is a candidate for surgery, it is less likely that there will be anesthetic-specific concerns about pulmonary function that would override the surgical decision, particularly since many of these surgeries are performed to treat cancer. That notwithstanding, it is helpful for the anesthetist to understand the patient’s risk stratification to plan for the level of ventilatory support required during and after surgery. Assessment must consider multiple variables of function. Similar to standards set by the American Thoracic Society and the European Respiratory Society, the American College of Chest Physicians (ACCP) proposes the following assessment of risk factors for patients with lung cancer undergoing lung surgery.⁴⁰ Preoperative FEV₁ greater than 80% of predicted value (or greater than 2 L for pneumonectomy or greater than 1.5 L for lobectomy) indicates average risk, and no further assessment of lung function is required. The diffusing capacity for carbon monoxide (DLCO) should be assessed if diffuse parenchymal disease or dyspnea on exertion is noted. If the FEV₁ or DLCO is less than 80% of predicted, then the predicted postoperative FEV₁ and DLCO are assessed.⁴⁰ This is accomplished either through radionucleotide scanning or mathematically, based on the proportion of total lung that will remain after the planned resection. The PPO FEV₁ can be calculated by multiplying the current FEV₁ by the fraction of functioning lung or the fraction of lung segments that will remain after surgery.¹⁵ For high-risk patients, more detailed assessment via radionuclide scanning, computed tomography (CT) scanning, or magnetic resonance imaging (MRI) is advisable. Predicted postoperative values of PPO FEV₁ greater than 40% of predicted for the patient indicate average risk. Values less than 30% of predicted indicate increased risk, and values in between warrant exercise testing to assess oxygen consumption (O₂max). A O₂max less than 15 mL/kg/min indicates high risk, whereas a value greater than 15 indicates average risk.⁴⁰ The European Respiratory Society places cardiopulmonary reserve (O₂max) more prominently in the assessment and considers high-risk cutoffs as being less than 30% for PPO FEV₁ and DLCO, and less than 10 mL/kg/min for O₂max.⁵³ As related to anesthetic planning, average-risk patients (e.g., PPO FEV₁ greater than 40%) are likely to be extubated immediately after surgery. High-risk patients (e.g., PPO FEV₁ less than 30%) have a higher likelihood of requiring some degree of postoperative ventilation. Planning for intermediate-risk patients (e.g., PPO FEV₁, 30%-40%) is further individualized, based upon other assessment parameters.

Ventilation-Perfusion Assessment

When the preoperative lung function tests indicate that the patient is at increased risk for perioperative complications, split lung function tests of ventilation and perfusion are valuable in the prediction of postresection lung function.53,56 Removal of a diseased portion of lung may not decrease overall lung function; in fact, it may improve it. The extent of pulmonary surgery has been found to correlate inversely with the intraoperative Pao₂, where patients undergoing pneumonectomy had higher Pao₂ than those undergoing lobectomy, which in turn were higher than those undergoing segmentectomy.³ This paradox is related to the corresponding amount of perfusion of the diseased lung. With larger, central tumors (such as would require a pneumonectomy), perfusion, and thus shunting under OLV, is diminished in comparison to a more peripheral lesion requiring a limited resection. Likewise, the results of perfusion scanning can provide a prediction of the degree of hypoxemia under OLV, because the degree of perfusion to the operative lung is proportional to the degree of potential shunt produced when ventilation to that lung ceases.³

Ventilation can be measured by having the patient inhale one vital capacity breath of a radioisotope and measuring isotope counts with multiple scanners placed over the chest wall. Radioisotope injected intravenously and imaged shows the distribution of perfusion to all areas of the lung. After determining function in various areas of the lung, calculations can then estimate postresection function by multiplying current function by the fraction of functioning lung that will remain postoperatively. Calculations based on segmental lung regions may help predict outcomes for patients undergoing lung volume reduction surgery. This procedure of removing emphysematous portions of lung to improve overall lung function has proved efficacious and particularly beneficial in allowing resection of cancerous lung tissue from patients in whom overall lung function studies would have contraindicated surgery.⁴⁰ Lung volume reduction surgery is most useful in patients with heterogeneous emphysema (particularly when the emphysematous lobe is also the one containing the tumor), where removal of a lung segment or lobe will result in better pulmonary function overall. Incidentally, this effect is appreciated more often with upper rather than lower lobectomy, in which patients with a low preoperative FEV₁ tend to demonstrate improvement in the FEV₁ after upper lobectomy.⁵⁷

Dynamic MRI and quantitative CT are newer modalities used to determine postresection pulmonary function.58,59 Dynamic MRI traces movement of oxygen or sulfur hexafluoride to reflect diffusing capacity.⁶⁰ Quantitative CT is intended to provide more specific data than global measurements such as FEV₁. CT can be used to quantify low attenuation (emphysematous) areas of lung to determine both overall proportion and regional distribution of disease. Results are comparable to FEV₁ in predicting obstruction.⁶¹

Diffusion Capacity

Diffusion capacity of the lung tests the lung’s ability to allow transport of gas across the alveolar-capillary membrane. Because it is difficult to measure the diffusing capacity of oxygen, carbon monoxide is used, in which the patient inhales a small amount of carbon monoxide, holds the breath for 10 seconds, exhales, and the amount of carbon monoxide in the exhaled breath is measured. After subtracting the amount of carbon monoxide that should be expired with dead space air, the amount exhaled provides an indicator of the diffusion of gases in the lung. A carbon monoxide diffusing capacity of the lungs (DLCO) less than 40% of predicted has been associated with increased complications after pulmonary surgery. However, DLCO has been found to have good specificity but low sensitivity as an independent measurement. The product of the predicted values for DLCO and FEV₁ may provide better reliability than single measures. This measurement, called the predicted postoperative product, was found by Pierce to be less than 1650 in 75% of those who died.⁶²

Cardiopulmonary Reserve

Maximal oxygen consumption (O₂max) during exercise testing is also assessed as a strong predictor of outcomes.31,40,63 A O₂max less than 10 mL/kg/min (or 40% of predicted) is associated with increased mortality, whereas a O₂max greater than 20 mL/kg/min is a favorable finding.⁶⁴ These values may be roughly estimated by evaluating the patient’s physical ability, in which the ability to climb five flights of stairs suggests O₂max greater than 20 mL/kg/min, and the inability to climb one flight of stairs suggests O₂max less than 10 mL/kg/min.⁶⁵

Patient Optimization

Aggressive treatment of acute or reversible components of respiratory disease greatly decreases the risk of postoperative complications. Treatable preoperative conditions include infections, excess bronchial secretions, bronchospasm, dehydration, electrolyte imbalance, cigarette smoking, alcohol abuse, and malnutrition (Figure 27-1).

FIGURE 27-1 A full, aggressive preoperative respiratory preparation regimen entails a five-pronged attack. (1) Require the patient to stop smoking. (2) Dilate the airways. (3) Loosen secretions. (4) Remove secretions. (5) Increase patient participation. IS, Incentive spirometry; PT, chest physiotherapy. (From Wilson WC, Benumof JL. Anesthesia for thoracic surgery. In: Miller RD, et al, eds. Miller’s Anesthesia. 6th ed. Philadelphia: Churchill Livingstone; 2005.)

Smoking is not only a major risk factor for chronic lung disease, but it is a strong predictor of perioperative complications, as well. Among patients undergoing noncardiac surgery, pulmonary complications occurred in 22% of smokers, 13% of past smokers, and only 5% of nonsmokers.⁶⁶ In a study of lung cancer patients, 87% were smokers, and the smokers had 1.5% mortality, whereas nonsmokers had only 0.4% mortality. Smokers in that study demonstrated twice the rate of complications as did nonsmokers.⁶⁷ The correlation of complications with the duration of smoking history can be performed through the pack-year index (the product of the packs per day smoked times the years of smoking at that rate). Patients with greater than a 20 pack-year history have demonstrated increased incidence of complications compared with those who have a more modest smoking history.^68,69

Smoking cessation may reduce postoperative complications; however, the timing of this intervention is important. Most recent research indicates that smoking cessation for less than 4 weeks prior to surgery does not alter risk of complications at all.66,70,71 Some data suggests that short-term smoking cessation (less than 1 month) may cause increases in mucus production, which may increase complications;^66,72 however, research findings are equivocal on the concept of increasing complications.^71,72 Carboxyhemoglobin levels have shown a significant decrease that occurs rapidly after smoking cessation, and improvement of nasal mucociliary clearance occurs within 1 to 2 weeks; however, the implications of these findings on postoperative complications are unclear.^73,74 Reduction in complication rates occur in proportion to the amount of time after quitting, with a threshold of at least 4 weeks to observe improvement, with even more improvement noted after 8 weeks.^70,71 Only in one very large study was a trend toward slight reduction in complications noticed with less than 1 month of smoking cessation, and complications further decreased in proportion to the total duration of cessation.⁶⁷ Smoking cessation counseling and intervention is a widely recommended strategy for medical management of COPD patients; however, owing to the urgent nature of treating pulmonary carcinoma, delaying surgery to allow for an adequate period of smoking cessation is an impractical goal.

Monitoring Plan

The purpose of monitoring during thoracic surgery is the quick recognition of sudden and severe changes in ventilation and hemodynamics that can accompany positioning, one-lung ventilation (OLV), and surgical manipulation of the airway and thoracic structures. Standard monitors, according to American Association of Nurse Anesthetists (AANA) guidelines, should be used.⁷⁵ An airway pressure monitor helps detect changes in airway compliance and assists in identifying the proper placement of double-lumen tubes (DLTs). Capnography is useful for determining the adequacy of ventilation when one lung is deflated, and also useful for detecting abrupt changes in cardiac output, which may accompany positioning or surgical manipulation. Considering preexisting ventilation derangements and changes with lateral position, the gradient of end-tidal to arterial CO₂ may be wider than the typical 5 mmHg. There is evidence that even if the gradient is determined early, it may not remain the same throughout the case. One-lung ventilation is a good application for transcutaneous CO₂ monitoring where that modality is available.

Electrocardiogram

All patients require continuous monitoring of the ECG. Typically, anesthetists monitor a limb lead (II) for easy rhythm recognition, and a precordial lead to add sensitivity for detection of ischemia. A landmark article by London identified V₅ as the precordial lead that would add the most sensitivity to ischemia detection.⁷⁶ That article noted that monitoring a combination of leads II and V₅ would help detect more than 85% of ischemia. A more recent study, which accounted for both the changing pattern of ischemia over time (onset vs. peak), as well as comparison to preoperative values, found lead V₄ to be the most sensitive for detecting ischemia, followed by V₅.³⁸ In practice, concerns for positioning, surgical site preparation, and access to the skin electrode may result in the precordial lead not being placed in the precise location needed. Various studies have found ischemic evidence to be frequent in leads V₃ through V₆.^38,77,78 A consistent finding is that monitoring a combination of leads is significantly more effective than monitoring a single lead.^38,76,79 A second intraoperative lead monitored in the V₄ to V₅ position or in whatever anterolateral position provides the most isoelectric S-T segment is desirable.³⁸

Arterial Pressure Monitoring

Arterial blood pressure monitoring instantly identifies acute hypotension with surgical manipulation. It also facilitates sampling of arterial blood for gas analysis. For thoracotomies, the arterial cannula is generally placed in the dependent arm, where it is more easily stabilized. For mediastinoscopy, the arterial monitoring site is selected to provide indication of innominate artery occlusion (see Mediastinoscopy later in chapter).

Central Venous Pressure Monitoring

Central venous pressure (CVP) monitoring is not required for routine thoracotomies but may be indicated if the patient’s volume status is unclear or if large fluid shifts are anticipated. In complex cases, a CVP line may help manage fluid status. Increased filling pressures (CVP or pulmonary capillary wedge) have been associated with greater lung injury and prolonged mechanical ventilation after complex pulmonary surgery.⁸⁰ In addition, a large-bored CVP line can provide access for rapid infusion and an access site if transvenous pacing or pulmonary artery pressure monitoring become necessary.

The CVP line can be inserted via the external or internal jugular veins or the subclavian veins. An external jugular line is more easily kinked in the lateral position. One should remain alert to the possibility of pneumothorax with the insertion of central lines. A pneumothorax on the ventilated (nonoperative) side can lead to severe hypoxemia during OLV. If a subclavian puncture is planned, the insertion site should be on the same side as the planned thoracotomy.

Cardiac Performance Monitoring

Pulmonary artery pressure monitoring is intended to provide estimation of left ventricular pressures and guide the support of cardiac performance with fluids and cardiovascular drugs. In spite of its past popularity, pulmonary artery catheterization has not been demonstrated to improve patient outcomes in either cardiac or noncardiac surgery,⁸¹ nor is it helpful in predicting postoperative complications.^53,82 There have even been suggestions that right heart catheterization may promote cardiac complications.⁸³ If pulmonary artery monitoring is deemed useful, the anesthetist must be cautious in interpreting values in pulmonary disease because the normal correlation of right and left ventricular pressures may be disturbed. The use of pulmonary artery catheters has specific limitations during OLV. Lung pathology or hypoxic pulmonary vasoconstriction may alter the resistance in pulmonary vessels and reduce the correlation between pulmonary artery occlusion pressure and left ventricular pressure.⁸⁴ More than 90% of pulmonary artery catheters float into the right lung.⁸⁵ During right thoracotomy, then, the catheter will likely be in the nondependent, collapsed lung and give a false low reading for cardiac output. Finally, care must be taken to ensure that a pulmonary artery catheter is not situated in a vessel that will be clamped during the course of lung resection. A better evaluation of cardiac filling, contractility, and valvular performance would come by way of echocardiography. Preoperative echocardiography is indicated when suspicion exists of valvular disease, outflow tract obstruction, ventricular dysfunction, or pulmonary hypertension.^17,53

Lateral Decubitus Position

The position most commonly chosen for surgical exposure during thoracotomy is the lateral decubitus position. A roll is placed beneath the torso just caudal to the axilla to prevent compression of the neurovascular bundle and forward rotation of the humeral head. It is important to note that this commonly called “axillary roll” is better considered an “axillary support roll,” because positioning it in the axilla may cause neurovascular compression. Hyperabduction of the arms is prevented to keep the brachial plexus from stretching against the humeral head. Arms can be separately padded and extended forward with armboards. Strategies for supporting the nondependent arm may include a pillow between the arms, a padded Mayo stand (which provides good access to intravenous or arterial lines in the dependent arm), or specially made double armboards. Pulse oximetry or frequent palpation of the radial pulse ensures the integrity of circulation to the hand.

The head is supported on pillows to maintain alignment of the head and neck with the spine. Lateral flexion of the neck can cause compression of the jugular veins or vertebral arteries, compromising cerebral circulation. The dependent ear can be compressed by the weight of the head. Careful padding or use of a foam doughnut relieves this pressure, but care must be taken to prevent corneal abrasion and retinal ischemia by avoiding pressure on the eyes.

Other pressure points of concern in the lateral position include the peroneal nerve in the area of the fibular head of the dependent leg and the femoral head of the nondependent leg if a stabilizing strap is placed over the patient.

Physiology of the Lateral Decubitus Position

Positional changes and changes in chest-wall integrity produce significant alterations in ventilation and perfusion of the lungs during thoracic surgery.

Upright Position

The distribution of perfusion in the lungs depends on gravity in relation to the level of the heart and on pressures transmitted through alveoli. In a spontaneously breathing, upright patient, perfusion increases from the apex to the base of the lung (Figure 27-2). Ventilation also increases from apex to base, based on the relative compliance of alveoli. Owing to downward traction from gravity, pleural pressure is most negative at the apex of the lung, and this keeps alveoli distended (Figure 27-3). Dependent alveoli are less distended and therefore more compliant (i.e., can expand by a greater volume for a given pressure change because they are starting at a lower resting volume). Therefore, most of a tidal breath is distributed to the dependent alveoli (Figure 27-4). The increase in both ventilation and perfusion from apex to base is not parallel, and is certainly more complexly arranged than in neatly divided zones. However, the general increase in both from top to bottom results in efficient gas exchange.

FIGURE 27-2 The lung is divided into three zones according to the relative magnitudes of the pulmonary arterial, venous, and alveolar pressures (Pa, Pv, and PA, respectively). In zone 1, alveolar exceeds arterial pressure, so the collapsible vessels are held closed and there is no flow. In zone 2, arterial exceeds alveolar pressure, but alveolar exceeds venous pressure. Under these conditions, a constriction occurs at the downstream end of each collapsible vessel, and the pressure inside the vessel at this point is equal to alveolar pressure, so the pressure gradient causing flow is arterial-alveolar. This gradient increases linearly with distance down the lung, and therefore so does blood flow. In zone 3, venous exceeds alveolar pressure, and the collapsible vessels are held open. Here, the pressure gradient causing flow is arteriovenous, and there is constant perfusion of alveoli. (From West JB. Explanation of the uneven distribution of blood flow in the lung, based on the pressures affecting capillaries. In: West JB. Respiratory Physiology: The Essentials. 8th ed. Philadelphia: Lippincott Williams & Wilkins; 2008.)

FIGURE 27-3 Schematic diagram of the lung within the chest wall, showing the tendency of the lung to assume a globular shape because of its viscoelastic nature. The tendency of the top of the lung to collapse inward creates a relatively negative pressure at the apex, and the tendency of the bottom of the lung to spread outward creates a relatively positive pressure at the base. Therefore pleural pressure (Ppl) increases by 0.25 cm H₂O per centimeter of lung dependency. (From Triantafillou AN, et al. Physiology of the lateral decubitus position, the open chest, and one-lung ventilation. In: Kaplan JA, Slinger PD, eds. Thoracic Anesthesia. 3rd ed. Philadelphia: Churchill Livingstone; 2003.)

FIGURE 27-4 Due to gravity, alveoli in the most dependent portions of lung are compressed by the tissue above, while those in the most nondependent portions are suspended open. At end-inspiration and end-expiration, pressure inside all alveoli is equalized with atmospheric; however, an increasing gradient of pleural pressure toward dependent areas creates a gradient in transpulmonary pressure at rest. Although the change in transpulmonary pressure during inspiration is equal for all, alveoli in dependent regions are on the steep portion of the compliance curve and therefore receive the largest share of the tidal volume. (From Triantafillou AN, et al. Physiology of the lateral decubitus position, the open chest, and one-lung ventilation. In: Kaplan JA, Slinger PD, eds. Thoracic Anesthesia. 3rd ed. Philadelphia: Churchill Livingstone; 2003.)

Awake Lateral Position

Less vertical distance is present to cause differences in the intrapleural pressure and blood pressure gradients in the lateral position (Figure 27-5). Abdominal contents displace the diaphragm in a cephalad direction on the dependent side. Starting from a higher position in the thorax, the dependent hemidiaphragm can contract further. During inspiration, therefore, contraction of the diaphragm causes more of the tidal volume (V_T ) to fill the dependent lung. Because perfusion is dependent upon gravity, perfusion in the lateral position is also greatest in the dependent lung (Figure 27-6). Overall, the relationship of greater ventilation and perfusion in the dependent lung is unchanged, and gas exchange remains efficient.

FIGURE 27-5 Schematic representation of the effects of gravity on the distribution of pulmonary ventilation and blood flow in the lateral decubitus position. The vertical gradient in the lateral decubitus position is less than in the upright position; consequently, blood flow in zones 2 and 3 is less. Nevertheless, pulmonary blood flow increases with lung dependency and is greater in the dependent lung than in the nondependent lung. PA, Alveolar pressure; Ppa, pulmonary arterial pressure; Ppv, pulmonary venous pressure. (From Triantafillou AN, et al. Physiology of the lateral decubitus position, the open chest, and one-lung ventilation. In: Kaplan JA, Slinger PD, eds. Thoracic Anesthesia. 3rd ed. Philadelphia: Churchill Livingstone; 2003.)

FIGURE 27-6 A, An increasing gradient of intrapleural pressure from top to bottom of lungs creates different resting volumes of alveoli and creates variation in regional compliance as described in Figure 27-4. B, In the lateral decubitus position, gravity-related effects translate to greater compliance of the dependent lung and less of the nondependent lung. Therefore, in the spontaneously breathing person, more ventilation is delivered to the dependent lung. Ppl, Intrapleural pressure; P, Transpulmonary pressure; V, alveolar volume. (From Triantafillou AN, et al. Physiology of the lateral decubitus position, the open chest, and one-lung ventilation. In: Kaplan JA, Slinger PD, eds. Thoracic Anesthesia. 3rd ed. Philadelphia: Churchill Livingstone; 2003.)

Anesthetized Lateral Position, Chest Closed, with Spontaneous Ventilation

A change in the distribution of ventilation is seen with the induction of anesthesia, even when spontaneous respiration is maintained. Functional residual capacity (FRC) decreases almost immediately with the induction of anesthesia. The weight of the mediastinum and the cephalad displacement of the diaphragm by abdominal contents further decrease FRC in the dependent lung and reduce the proportion of the favorable zone 3 area. Lower volumes in each lung shift their place on the compliance curve. The lungs are less compliant when they are either at a very high volume (distended alveoli) or a very low volume (atelectasis). In the anesthetized patient, the nondependent lung moves from a flat, noncompliant portion of the compliance curve to a more compliant position. Although anesthesia results in a net loss of FRC, the relative proportion of FRC in the nondependent lung increases in contrast to the dependent lung.⁸⁶ As the dependent lung loses FRC, its volume becomes so low as to decrease its compliance. It shifts to a less compliant, flatter portion of the curve (Figure 27-7). Ventilation is therefore preferentially distributed to the nondependent lung, whereas gravity-dependent blood flow preferentially goes to the dependent lung, resulting in a mismatch of ventilation and perfusion.

FIGURE 27-7 Diagram showing A, regional compliance of lungs in the awake patient in lateral decubitus position and B, regional compliance of lungs in the anesthetized patient in lateral decubitus position. Induction of anesthesia causes a loss of lung volume in both lungs, with the nondependent lung moving from a flat, noncompliant portion of the pressure-volume curve to a steep, compliant portion. The dependent lung moves from the highly compliant portion toward the lower flattening of the curve. In some areas of the dependent lung, resting alveolar volume is so low that atelectasis and airway closure impede inflation. Therefore, in the anesthetized patient in lateral decubitus position, more tidal ventilation shifts toward the nondependent lung (where there is the least perfusion), creating ventilation-perfusion mismatch. P, Transpulmonary pressure; V, alveolar volume. (From Triantafillou AN, et al. Physiology of the lateral decubitus position, the open chest, and one-lung ventilation. In: Kaplan JA, Slinger PD, eds. Thoracic Anesthesia. 3rd ed. Philadelphia: Churchill Livingstone; 2003.)

Anesthetized, Paralyzed, Mechanically Ventilated

Under mechanical ventilation, the diaphragm no longer contributes to ventilation of the lower lung, and FRC further declines because the compression from abdominal viscera is no longer counteracted by the force of the contracting diaphragm (Figure 27-8). With the initiation of mechanical ventilation and the deletion of effect of the contracting diaphragm, ventilation further shifts to follow the path of least resistance, favoring the nondependent lung. The ventilation-perfusion relationship further deteriorates. The addition of positive end-expiratory pressure (PEEP) to mechanical ventilation may help restore FRC and improve the ventilation-perfusion ratio.