These authors also developed a multifactorial risk factor index for predicting postoperative respiratory failure. Postoperative respiratory failure occurred in <0.01 % of low -risk patients (who had a combined risk factor score <8), 0.8 % of medium -risk patients (score 8–12) and 6.8 % of high-risk patients. Very high-risk patients (score >25) had a 40 % incidence of postoperative respiratory failure.
More recently, an index for the risk of re-intubation was developed from a study of 34,000 patients. This comprised: ASA grade (III or IV: 3 points), emergency surgery (3 points), high-risk surgery (2 points), congestive cardiac failure (2 points) and COPD (1 point). The risk of re-intubation was significantly higher in patients with a score of 7–11 (5.9 %) than in patients with a score of 0 (0.1 %, p < 0.001) [5].
These large trials provide insights into the multiplicity of factors that contribute to PPCs. In addition, respiratory risk indices are valuable research tools for investigating strategies to decrease their incidence. However, because of their low sensitivity, these indices have limited utility in the clinical management of individual patients. For example, it is unlikely that a patient would be denied essential surgery even if they had a high predicted risk of PPCs (i.e. more than 10 %). Similarly, a high predicted risk of re-intubation (i.e. about 6 %) is too insensitive to guide extubation and for the same reason, has limited utility in determining the allocation of intensive care resources. More research is needed to determine whether the sensitivity of risk indices for PPCs in high-risk patients can be improved by the addition of other variables, such as (for example) respiratory investigations or functional testing.
10.3 Patient-related Risk Factors for PPCs
While it should be no surprise that age and type of surgery are the most important risk factors for PPCs, the relative contribution of some other patient risk factors may be less widely appreciated.
10.3.1 Pulmonary Risk Factors
10.3.1.1 Chronic Obstructive Lung Disease
Chronic obstructive pulmonary disease (COPD) approximately doubles the risk of PPCs overall [4]. In one study, more than a third of patients with severe COPD (FEV1 ≤1.2 L and FEV1/FVC <75 %) who underwent major non-cardiothoracic surgery had major PPCs and nearly half of them died within 2 years of surgery [6]. However, well controlled mild to moderate asthma does not significantly increase the risk of PPCs [7].
10.3.1.2 Interstitial Lung Disease
The influence of other lung diseases on PPCs is much less well studied. A recent study showed that interstitial lung disease (which was mostly idiopathic pulmonary fibrosis) also approximately doubled the risk of PPCs. This risk was further increased in males (OR 1.8) and in patients with a low BMI (OR 2.5). However, in common with other studies, surgical factors (such as operative site and emergency procedures) had a much greater influence on the risk of PPCs [8].
10.3.1.3 Pulmonary Hypertension
Severe pulmonary hypertension has only relatively recently been recognised as a significant risk factor for postoperative respiratory failure (~25 %) and perioperative mortality. In a study of 145 patients undergoing major non-cardiac surgery with a mean pulmonary arterial systolic pressure (PAPs) of 68 mmHg, 16 patients (11 %) developed congestive heart failure and 5 patients (3.5 %) died from right ventricular (RV) failure [9]. In another study of 62 patients with more severe pulmonary hypertension (mean PAPs of 79 mmHg) undergoing similar surgery, mortality was 9.7 % vs. 0 % for case-matched controls [10]. Most of these deaths occurred due to acute RV failure.
While routine preoperative screening for pulmonary hypertension is not advocated, these studies highlight its clinical significance and the importance of preserving RV function in these patients perioperatively [11].
10.3.2 Non-pulmonary Risk Factors for PPCs
There are other important, but non-modifiable non-pulmonary patient risk factors that increase the risk of PPCs up to threefold. These include an ASA classification of >2, congestive cardiac failure and functional dependence [4].
10.3.2.1 Obesity
Contrary to popular perception, obesity is not strongly associated with an increased risk of PPCs. For example, in a recent review of 20 papers, obesity did not predict an increased risk of unplanned admission following day surgery [12].
Although morbid obesity does increase the risk of several postoperative complications (such as wound infections), it does not significantly increase the risk of postoperative respiratory failure [13, 14]. This is surprising, given the high incidence of obstructive sleep apnoea (OSA) in this population, which is more than 70 % in patients undergoing bariatric surgery [15].
However, super obese patients (BMI >50 kg/m2) have an increased risk of obesity hypoventilation syndrome, heart failure and pulmonary hypertension. They also have an increased risk of perioperative complications from bariatric surgery compared with less obese patients, including airway difficulties, venous thromboembolism, prolonged hospitalisation and death [16].
Hence while day surgery appears to be well tolerated in patients with a BMI <40 kg/m2, this may not be the case for super obese patients.
10.3.2.2 Obstructive Sleep Apnoea
Obstructive sleep apnoea is characterised by periods of partial or complete upper airway collapse and obstruction during normal sleep, which is aggravated by the respiratory depressant effects of general anaesthetic agents, sedatives and opiates [17]. Several studies have shown that OSA increases the risk of PPCs, especially hypoxaemia and airway difficulties [7, 18]. However, there remains considerable clinical uncertainty (and controversy) regarding the optimal perioperative management of OSA.
Recently published guidelines from the American Society of Anesthesiologists recommend that the decision to perform day surgery in patients with OSA should be based on its severity, the invasiveness of the surgery and the likely requirement for postoperative opioids [19]. However, these guidelines have not been validated clinically, are largely based on expert opinion and are poorly supported by the available literature.
For example, severity of OSA has not been proven to increase the risk of perioperative complications. In a prospective study of 797 patients undergoing bariatric surgery at the Mayo Clinic (all of whom had preoperative sleep studies), severity of OSA was not associated with an increased rate of any postoperative complications [20]. In an earlier study from the same institution, OSA was also not found to be a significant risk factor for unplanned admissions (or other adverse events) in patients undergoing outpatient non-ENT surgery [21]. However, OSA had been identified preoperatively in all patients in these two studies, so the perioperative risk of undiagnosed or untreated OSA was not examined. It is also possible that patients diagnosed with severe OSA may have been subject to selection bias and may have received different management (e.g. more intensive monitoring or lower opiate doses).
While the above studies examined the perioperative impact of OSA in previously diagnosed patients, Stierer et al. [22] assessed its prevalence and significance in previously undiagnosed patients. A screening survey for OSA (the Maislin Index Score) was administered to 3553 patients having day surgery at Johns Hopkins Hospital. One hundred and three patients (4.8 %) were identified as having a high propensity for OSA, which had not been previously diagnosed in three quarters of these cases. The survey results were not disclosed to their anesthesiologists. Patients with a high propensity for OSA did not have an increased rate of unplanned hospital admission, however they did have an increased incidence of airway difficulties and a greater requirement for postoperative oxygen. In addition, these patients were more likely to be on antihypertensive medications and to receive intraoperative pressors and beta-blockers, which is consistent with the increased incidence of hypertension and cardiac rhythm disturbances found in OSA patients [23].
Although there is good evidence that careful patient management in experienced centres mitigates the potential risks of OSA in patients undergoing outpatient surgery, these findings are not generalisable to less well resourced facilities, such as small private hospitals and stand -alone day-surgical units. This is an important distinction, because inadequate postoperative monitoring (and excessive opioid administration) have been implicated in the deaths of patients with OSA following relatively minor surgical procedures [24].
10.3.3 Why Assessing the Risk of PPC May (or May Not) Be Useful
10.3.3.1 Informed Patient Consent
The preoperative identification of patients who are at increased risk of PPCs facilitates informed consent for surgery, and gives patients an opportunity to discuss the limits of therapy that they are willing to receive. For example, some patients may not be willing to undergo surgery if it entailed a significant risk of prolonged postoperative ventilation. However, as discussed above, because of their poor sensitivity, the available risk indices have limited clinical utility in predicting this complication.
10.3.3.2 Targeting Interventions in High-Risk Patients
Another reason that may be advocated for assessing the risk of PPCs preoperatively is that preventative strategies can be implemented in high-risk patients. However, strategies that have been proven to be effective in decreasing PPCs (such as lung expansion techniques) are cheap, safe and easily implemented, hence they should be used in all patients undergoing major surgery, regardless of their level of baseline risk.
10.4 Preoperative Respiratory Investigations
Preoperative respiratory investigations have a very limited role in the assessment of patients undergoing major non- cardiothoracic surgery, which is primarily based on clinical history and examination. This is because routine preoperative screening investigations are poor predictors of PPCs.
For example, spirometry is no better than clinical assessment by history and examination in predicting the risk of PPCs and is therefore not recommended as a routine screening test [4]. In contrast to patients undergoing lung resections, there is also no lower limit of FEV1 that reliably precludes non-thoracic surgery.
Similarly, in the absence of other risk factors, a routine preoperative chest X-ray is not indicated in patients less than 70 years of age, because it mostly detects pre-existing, chronic conditions. Routine preoperative CXR screening has also not been shown to significantly alter perioperative management or improve outcome [4, 25].
Cardiopulmonary exercise (CPEX) testing is a non-invasive method of assessing combined cardiovascular and pulmonary function. Unlike other screening tests, the anaerobic threshold that is derived from CPEX testing may provide a quantitative means of assessing “fitness for surgery,” as it has been shown to independently predict morbidity, mortality and duration of hospitalisation after major abdominal surgery [26]. In a trial currently underway in London, perioperative care guided by CPEX testing is being compared with standard management in patients undergoing colorectal surgery.
10.4.1 Preoperative Strategies to Decrease PPCs
10.4.1.1 Smoking Cessation
Cigarette smokers have a higher incidence of mortality and major cardiac and respiratory complications than non-smokers [27, 28].
Smokers are commonly advised to stop smoking at least 4–8 weeks preoperatively, as this has been thought to decrease the incidence of PPCs. However, stopping smoking within this period of time has also been suggested to increase PPCs, possibly due to increased sputum production [29]. More recently, a meta-analysis failed to demonstrate any difference in postoperative complications between patients who ceased smoking within 8 weeks of surgery and those who did not [30]. By contrast, patients who stop smoking at least 1 year preoperatively have the same perioperative mortality as non-smokers and also have a lower risk of cardiac and respiratory complications than active smokers [21].
While ceasing smoking within 8 weeks of surgery may do little to mitigate the risk of PPCs, patients should still be encouraged to do so, as it decreases the incidence of wound infections [31] and may promote long term abstinence.
10.4.1.2 Preoperative Steroids for Reactive Airways Disease
While well-controlled reactive airways disease (asthma and COPD) does not significantly increase the risk of PPCs, this is not the case for poorly controlled disease. In a study of patients with reversible, but poorly controlled reactive airways disease (FEV1 <70 % of predicted and not on chronic bronchodilator therapy) who receive salbutamol preoperatively, there was a high incidence of bronchospasm following intubation if it was given either preoperatively for 5 days (7/9 patients) or 10 min before induction (8/10 patients). Of note, intraoperative bronchospasm still occurred even though spirometry improved following salbutamol therapy in these patients. However, in a third group of patients who received oral methylprednisolone (40 mg) for 5 days preoperatively in addition to salbutamol, intraoperative bronchospasm was significantly decreased compared to those patients who received salbutamol alone (1 / 10 patients, p < 0.001) [32].
While there have been concerns that a short course of high-dose steroids may increase the risk of other surgical complications such as wound infections, several studies have shown that this does not occur.
10.5 Intraoperative Strategies to Decrease PPCs
10.5.1 Protective Lung Ventilation
There is recent evidence that the long established anaesthetic practice of mechanical ventilation with high tidal volumes (TV) of 7–10 ml/kg and low, or zero positive end-expired pressure (PEEP) should be re-evaluated in favour of “protective” lung ventilation strategies that minimise airway pressures and prevents atelectasis.
10.5.1.1 Background
The adverse effects of general anaesthesia and mechanical ventilation on the lung are well recognised. In particular, atelectasis rapidly occurs in the dependent regions of the lung. This is largely caused by compression of the lungs due to changes in the shape of the chest and the position and function of the diaphragm.
High inspired oxygen concentrations also contributes to alveolar collapse through absorption atelectasis [33]. Atelectasis from any cause increases intrapulmonary shunt and promotes bacterial growth, which in turn contributes to the development of postoperative pneumonia [34].
Traditionally, the rationale for mechanically ventilating patients with high tidal volumes during anaesthesia was to overcome atelectasis. However, this may cause lung injury by over-distending alveoli (volutrauma) in non-dependent regions of the lung, which receive a greater proportion of the tidal volume than dependent regions during mechanical ventilation. Volutrauma damages endothelial and other cells in the lung, which causes both local injury and damage to other organs through the release of systemic inflammatory mediators [35].
The most effective strategies for reducing atelectasis are avoiding high concentrations of inspired oxygen, the use of PEEP and recruitment manoeuvres.
While 100 % inspired oxygen rapidly promotes atelectasis during anaesthesia, this can be mitigated by even a small reduction in inspired oxygen concentration. For example, pre-oxygenation with 80 % oxygen causes much less atelectasis than 100 % oxygen (0.8 % vs. 6.8 % respectively), although it also decreases the time to oxygen desaturation [36]. Positive end-expired pressure (PEEP) should be routinely applied when breathing 100 % oxygen [37], as it reliably prevents atelectasis occurring.
While PEEP helps prevents atelectasis during general anaesthesia, it does not reliably re-expand areas of lung that have already collapsed. Atelectasis also occurs rapidly if PEEP is interrupted even briefly. Atelectasis can be effectively treated and prevented with the application of PEEP after a lung recruitment manoeuvre [33].
A recruitment manoeuvre is the delivery of a brief period of high inspiratory airway pressure. While a recruitment manoeuvre can completely reverse atelectasis [38], sufficiently high inflating pressures need to be delivered. For example, 20 cm H2O has almost no effect, but 40 cm H2O sustained for 7–8 s is completely effective [39]. While prolonged recruitment manoeuvres can cause haemodynamic compromise, repeated brief recruitment manoeuvres have not been shown to cause either adverse cardiovascular effects or lung damage [40].
Related posts:
Defining Postoperative Quality of Recovery
Perioperative Management of the Diabetic Patient
Perioperative Beta-Blockade, the Pros and Cons. The Story of Beta-Blockade and Cardiac Protection
Prehabilitation
Enhanced Recovery for Colorectal Surgery
Lung Ultrasound in Anaesthesia and Critical Care Medicine
Full access? Get Clinical Tree
