DISORDERS OF THE RESPIRATORY AND IMMUNE SYSTEMS

General Preoperative and Preprocedure Considerations

Although little may seem to have changed in the preoperative preparation of patients with respiratory disease, this impression is not true. Major changes in drug therapy have occurred, and appreciation has increased regarding the effects of smoking and sleep apnea on perioperative and chronic care.^{[502] [503] [504] [505] [506] [507] [508] [509] [510] [511] [512] [513] [514] [515]} (Preoperative and preprocedure identification and perioperative care of patients with sleep apnea are discussed in the earlier section on obesity.)

The main purpose of preoperative testing is to identify patients at risk for perioperative complications and to institute appropriate perioperative therapy. Preoperative assessment can also establish baseline function and the feasibility of surgical intervention. Whereas numerous investigators have used pulmonary function tests to define inoperability or high-risk versus low-risk groups for pulmonary complications, few have been able to demonstrate that the performance of any specific preoperative or intraoperative measure reliably decreases perioperative pulmonary morbidity or mortality. Because routine preoperative pulmonary testing and care are discussed extensively in Chapter 26 and Chapter 75 , the current discussion is limited to an assessment of the effectiveness of this type of care.

In fact, few randomized prospective studies indicate an outcome benefit of preoperative preparation; a larger cohort has implied a benefit of intraoperative and postoperative pain therapy techniques^{[516] [517] [518]} (usually planned preoperatively; see Chapter 25 ). Stein and Cassara^[519] randomly allocated 48 patients to undergo preoperative therapy (cessation of smoking, administration of antibiotics for purulent sputum, and use of bronchodilating drugs, postural drainage, chest physiotherapy, and ultrasonic nebulizer) or no preoperative therapy. The no-treatment group had a mortality of 16% and morbidity of 60%, as opposed to 0% and 20%, respectively, for the treatment group. In addition, the treatment group spent an average of 12 postoperative days in the hospital as compared with 24 days for the 21 survivors in the no-treatment group.

1086

Collins and colleagues^[520] prospectively examined the benefits of preoperative antibiotics, perioperative chest physiotherapy and therapy with bronchodilating drugs, and routine postoperative analgesia (morphine) on postoperative respiratory complications in patients with COPD. Of these therapies, only preoperative treatment with antibiotics had a beneficial effect.

Warner and coworkers^[521] collected data retrospectively about smoking history and prospectively (concurrently) about pulmonary complications for 200 patients undergoing CABG surgery. These investigators documented that 8 weeks or more of smoking cessation was associated with a 66% reduction in postoperative pulmonary complications. Smokers who stopped for less than 8 weeks actually had an increase (from 33% for current smokers to 57.1% for recent quitters) in the rate of one or more of the six complications surveyed: purulent sputum with pyrexia; need for respiratory therapy care; bronchospasm requiring therapy; pleural effusion or pneumothorax (or both) necessitating drainage; segmental pulmonary collapse, as confirmed by radiography; or pneumonia necessitating antibiotic therapy. Others have found that both shorter and longer periods of cessation of smoking were needed before achieving cardiovascular^{[522] [523] [524]} and hematologic benefit.^[525] Of note, Bluman and associates^[526] performed a retrospective chart review of 410 patients undergoing noncardiac surgery at a VA hospital. Current smoking was associated with a nearly sixfold increase in risk for a postoperative pulmonary complication. Reduction in smoking within 1 month of surgery was not associated with a decreased risk of postoperative pulmonary complications. Nakagawa and coauthors also reported higher pulmonary complication rates in patients undergoing pulmonary surgery who quit within 4 weeks of surgery than in current smokers or those who had stopped smoking for more than 4 weeks.^[527] The fact that anesthesiologists rarely see their patients 4 weeks or more before surgery presents a dilemma: if one is unable to advise the patient to stop smoking 8 weeks or more before surgery, is it preferable for the patient to continue smoking? Perhaps these data will further support the implementation of preoperative and preprocedure assessment clinics in which anesthesiologists are able to advise and counsel patients about risk reduction. When Skolnick and coworkers^[515] studied 602 children prospectively, exposure to passive smoking (as measured by urinary cotinine, the major metabolite of nicotine) correlated directly with airway complications. Children with the least exposure to passive smoke had the fewest complications.

Celli and associates^[528] performed a randomized prospective controlled trial of intermittent positive-pressure breathing (IPPB) versus incentive spirometry and deep-breathing exercises in 81 patients undergoing abdominal surgery. The groups exposed to a respiratory therapist (regardless of the treatment given) had more than a 50% lower incidence of clinical complications (30% to 33% versus 88%) and shorter hospital stays than the control group did. Thus, this third prospective study indicates that outcome improves when there is any concern about lung function on the part of someone knowledgeable in maneuvers designed to clear lung secretions.

Bartlett and coworkers^[529] randomly assigned 150 patients undergoing extensive laparotomy to one of two groups. One group received preoperative instruction in and postoperative use (10 times per hour) of incentive spirometry. The other group received similar medical care but no incentive spirometry. Only 7 of 75 patients using incentive spirometry had postoperative pulmonary complications, as opposed to 19 of 75 in the control group. However, other studies have not shown a benefit for specific treatments or have been too contaminated with bias to have a clear result emerge. Lyager and colleagues^[530] randomly assigned 103 patients undergoing biliary or gastric surgery to receive either incentive spirometry with preoperative and postoperative chest physiotherapy or only preoperative and postoperative chest physiotherapy. No difference in the postoperative course or pulmonary complications was found between the two groups. Other studies have shown a specific benefit (i.e., above that provided by routine care) for chest physiotherapy and IPPB. These studies are usually poorly controlled, not randomized, or retrospective in design (or any combination of the three); these deficiencies probably substantially bias the results toward finding a benefit in reducing postoperative pulmonary complications.^{[531] [532] [533] [534]} Although randomized prospective studies showed no benefit or actual harm from chest physiotherapy and IPPB on the resolution of pneumonia^{[535] [536]} or postoperative pulmonary complications,^{[520] [528] [533] [536] [537] [538] [539]} the four studies cited earlier^{[519] [520] [528] [529]} and numerous retrospective studies^{[521] [531] [532] [533]} strongly suggest that preoperative evaluation and treatment of patients with pulmonary disease actually decrease perioperative respiratory complications, even if only by causing a change in anesthetic techniques.^[516]

Recent meta-analyses have suggested a benefit of anesthetic and pain management with respect to respiratory outcomes. Rodgers and colleagues reviewed 141 trials involving 9559 patients who had been randomized to receive neuraxial blockade or general anesthesia. Overall mortality was significantly lower in the neuraxial blockade group (2.1% versus 3.1%). The relative risk of pneumonia in the neuraxial group was 0.61 (CI, 0.48 to 0.81), and the relative risk of respiratory depression was 0.41 (CI, 0.23 to 0.73).

Not all studies demonstrate beneficial effects of pretreatment. In afebrile outpatient ASA I and II children with no lung disease or findings who underwent non-cavitary, nonairway surgery for under 3 hours, neither albuterol nor ipratropium premedication decreased adverse events.^[540]

Evaluation of dyspnea is especially useful and thus warrants discussion here (for a review of the specific pulmonary function tests that identify high-risk groups, see Chapter 26 ). Boushy and coworkers^[539] found that grades of preoperative dyspnea correlated with postoperative survival. (Grades of respiratory dyspnea are provided in Table 27-41 .) Mittman^[541] demonstrated an increased risk of death after thoracic surgery from 8% in patients without dyspnea to 56% in patients who were dyspneic. Similarly, Reichel^[542] found that no patients died after pneumonectomy if they were able to complete a preoperative treadmill test for 4 minutes at the rate of 2 mph on level ground. Other studies have found that the history and physical examination of an asthmatic subject can also predict the need for hospitalization.^{[504] [543]} Wong and

1087

Arozullah and associates developed the first validated multifactorial risk index for postoperative respiratory failure, defined as mechanical ventilation for more than 48 hours after surgery or reintubation and mechanical ventilation after postoperative extubation.^[547] In a prospective cohort study of 181,000 male veterans as part of The National Veterans Administration Surgical Quality Improvement Program, seven factors independently predicted risk ( Table 27-43 ). With increasing numbers of risk factors present, the rate of complications increased from 0.5% (class 1) to 26.6% (class 4). Arozullah and colleagues subsequently developed a risk index for postoperative pneumonia by using data on 160,805 patients undergoing major noncardiac surgery and validated the index by using data on an additional 155,266 patients. ^[548] Patients were divided into five risk classes by using risk index scores ( Table 27-44 ). Pneumonia rates were 0.2% in those with 0 to 15 risk points, 1.2% in those with 16 to 25 risk points, 4.0% in those with 26 to 40 risk points, 9.4% in those with 41 to 55 risk points, and 15.3% in those with more than 55 risk points.

From our perspective, the important information and conditions to search for during the history taking and physical examination are as follows:

1. Dyspnea (see Table 27-41 ).
2. Coughing and production of sputum. Sputum, if present, should be Gram-stained and cultured, and appropriate antibiotic treatment should be instituted.

---

1088

3. Recent respiratory infection. Viral respiratory infections affect respiratory function by giving rise to increased airflow obstruction that may persist for as long as 5 weeks.^{[549] [550]} These infections also adversely affect the respiratory mechanisms responding to bacteria. However, whether the incidence of complications in normal children is lessened by waiting 5 weeks until the symptoms of respiratory infection have disappeared is open to question.^{[551] [552] [553] [554] [555] [556]} Elwood and colleagues recently studied afebrile outpatient tertiary care children (aged 2 months to 18 years, N = 109) without lung disease or findings who underwent noncavitary, nonairway surgery for under 3 hours and did not find any association between adverse events and either upper respiratory tract infection (URI) within 6 weeks (n = 76) or URI within 7 days (n = 21).^[540] Therefore, the risk and benefits of delaying surgery must be balanced in patients with URI if they are afebrile.
4. Hemoptysis.
5. Wheezing and previous use of bronchodilating drugs and corticosteroids (systemic or inhaled). Wheezing often suggests potentially reversible airway obstruction but is a notoriously poor indicator of the degree of obstruction. In addition, not all wheezing is caused by bronchospasm. Cardiac and other pulmonary causes must be differentiated from asthma.^[557] Drug therapy for asthma should be ascertained and optimized if possible.^[558] Asthmatics have a fourfold increase in perioperative respiratory complications. Holleman and Simel^[504] found that auscultated audible wheezing on nonforced expiration (along with a 70-pack-year history of smoking and patient sensation of wheezing) was a reliable sign of severe airflow limitation.
6. Pulmonary complications from previous surgery. Prolonged endotracheal intubation after surgery can be required by many conditions, most notably respiratory and neuromuscular disorders.
7. A history of smoking. The incidence of respiratory complications is higher in tobacco smokers than nonsmokers.^{[521] [531] [532] [559] [560]}
8. Age, general history of the patient, and any other significant physical findings. Although other disease conditions probably increase a patient's respiratory risk, the documentation on this hypothesis is barely adequate.^[544] Old age definitely increases respiratory and cardiac risk,^{[521] [530] [547] [548] [559] [561] [562]} although it is less important in ambulatory surgery.^[563] In a multivariate analysis, McAlister and colleagues reported that age older than 65 years was associated with an odds ratio of 1.8 for pulmonary complications.^[560] Cardiovascular history and examination are obviously important in themselves but are also especially helpful in revealing signs of pulmonary hypertension, such as right ventricular lift (i.e., lift over the lower part of the sternum), fixed and widely split second heart sound, and S₄ gallop at the left sternal border.
9. Breathing frequency and form. Pursed lips, cyanosis, and the use of accessory muscles should be noted. Vocal cord dysfunction occurs commonly from both psychosocial illness and an attempt to increase end-expiratory pressure.^[564]
10. Body habitus:
1. Abnormalities of the chest wall, trauma, kyphoscoliosis with restrictive lung disease. Development of a barrel chest is a late manifestation of obstructive lung disease.
2. Obesity. There is controversy regarding the influence of obesity on postoperative pulmonary complications. In an analysis of 12 studies conducted between 1968 and 2000, obesity was associated with a relative risk of 1.3 for postoperative pulmonary complications, but this risk was not clinically significant.^[565] However, a weight 30% over ideal doubles the incidence of respiratory complications according to one author (Amalraj S, personal communication).^{[90] [91] [92] [93] [94] [95] [106] [107] [108] [110] [111] [112] [113] [114] [535] [559]}
3. OSA is associated with increased pulmonary complications, primarily hypercapnia and hypoxia,^{[110] [111] [112] [118] [566]} as well as a threefold increased rate of unplanned ICU stay.

---

1089

11. Adequacy of the upper airway, presence of tracheal deviation, ease of facemask application, ease of endotracheal intubation.
1. McAlister and coauthors reported that a maximal laryngeal height of 4 cm or less was associated with an odds ratio of 2.0 for pulmonary complications after nonthoracic surgery.^[560]
12. Presence of rales, rhonchi, wheezing (especially on nonforced expiration), diaphragmatic excursion, and air movement and the ratio of expiratory to inspiratory time.^[504]
13. Site of proposed surgery. Upper abdominal surgery increases the incidence of perioperative pulmonary complications.^{[516] [519] [532] [544] [545] [567]}
14. Surgery performed on an emergency basis. Wong and colleagues^[544] found that surgery performed on an emergency basis increased the risk of perioperative pneumonia 13-fold in patients with severe COPD.

Chapter 25 and Chapter 26 review the value of chest radiographs and pulmonary function tests in identifying patients with preoperative pulmonary disease, as well as which tests should be ordered and for whom. A recent study suggests that clinical evaluation is more accurate than abnormal spirometry. Warner and colleagues used a matched control design to study smokers undergoing abdominal surgery and found no significant differences in the incidence of prolonged intubation, pneumonia, or prolonged ICU stay in those with abnormal findings on spirometry.^[568] Any patient who will require postoperative ventricular support should be considered for preoperative testing. Although we have yet to find the ideal pulmonary function test that guarantees success or that predicts a poor outcome that could be improved by therapy, the following conditions are relatively good predictors of complications associated with resection of a lung: a maximum breathing capacity of less than 40 or 50 L/min/70 kg, an FEV₁ /FVC ratio of less than 40% of predicted, an FEV₁ of less than 1.2 L, or a maximum midexpiratory flow rate of less than 50 L/min.^{[531] [534] [541] [542] [543] [544] [545] [569] [570] [571]}

To these conditions is now added the diffusing capacity of the lung for carbon monoxide (DLCO).^{[391] [570] [571]} DLCO decreases with fixed obstructive lung disease (i.e., emphysema) but remains normal in bronchospastic disease. In one study, when preoperative DLCO was less than 60% of predicted, the mortality rate was 25% and the pulmonary morbidity rate was 45%, whereas a DLCO of 100% or greater of predicted was associated with a 0% mortality rate and a 11% pulmonary morbidity rate. A subsequent study confirmed these findings.^[571] Although we have always tried to reduce laboratory studies, they may prove useful in a patient with COPD who requires pulmonary surgery. Further analysis is necessary to determine whether the finding of a normal DLCO may allow expansion of the usual spirometric criteria to permit successful pulmonary resection in patients who are currently excluded as candidates for such operations. We conduct such tests on any patient with dyspnea of grade II or higher or any patient who shows significant abnormalities or risk on the first 13 factors listed earlier. Wong and coworkers^[544] found that the Shapiro score (see Table 27-42 ) and ASA physical status independently predicted adverse outcomes after surgery in patients with severe obstructive pulmonary disease. Although this score depends heavily on abnormal pulmonary function test results for grading risk, abnormalities in the cardiovascular, neuromuscular, and central nervous systems, plus restrictions in ambulation, can also increase the risk score. A Shapiro score of 5 points or more was associated with a 17-fold higher risk of perioperative death and a 14-fold higher risk of post-operative bronchospasm after nonthoracic surgery.^[544]

Despite the lack of definitive data establishing the efficacy of preoperative pulmonary testing and therapy, we recommend the following approach:

1. Eradicate acute infections and suppress chronic infections by using appropriate diagnostic measures and antibiotic treatment.
2. Relieve bronchospasm by using inhaled corticosteroids and bronchodilating drugs and document such relief with measurements of FEV₁ (also see Chapter 26 and Chapter 75 ).
3. In patients with bronchial asthma, consider the administration of corticosteroids beginning at least 48 hours before surgery in patients with a significant history of bronchospastic disease to achieve the maximal effect at the time of surgery. No randomized data support such an approach, but studies have not shown an increase in the risk of respiratory infection or wound complications.^{[572] [573]}
4. Institute measures to improve sputum clearance and familiarize the patient with respiratory therapy equipment (incentive spirometry) and postural drainage maneuvers. Initiate practice coughing and deep-breathing exercises (also see Chapter 26 and Chapter 75 ).
5. Treat uncompensated right ventricular heart failure with digoxin, diuretics, oxygen, and drugs that decrease pulmonary vascular resistance (e.g., hydralazine). ^[574]
6. The use of low-dose heparin prophylactically to decrease the incidence of venous thrombosis (and pulmonary emboli) has been the standard.^{[478] [479] [480] [575]} More recently, the use of low-molecular-weight heparin and intermittent compression stockings has become standard for high-risk patients.^{[576] [577]}
7. Identify and treat suspected or diagnosed sleep apnea with CPAP or BiPAP or other measures as indicated earlier (in the section "Obesity").
8. Encourage reduction or cessation of smoking at least 4 weeks and preferably 8 weeks or more before surgery.^[521] Although the debate about cessation of smoking includes more than pulmonary risk, cardiovascular, hematologic, and aspiration risk has also not been shown to be responsive to short-term cessation.^{[521] [522] [523] [524] [525]} Perhaps the benefit of using a major life event, an operation, to promote cessation of smoking is worth the increased short-term risk that encouraging cessation within a day or two of surgery would entail; the latter hypothesis remains to be tested. Even young people who smoke only a half to a pack of cigarettes per day exhibit abnormalities in respiratory function.^[578] Avoiding passive smoking also appears beneficial.^[515]