Abstract
Obesity is a major health problem that is associated with significant comorbidities including Obstructive Sleep Apnea (OSA). OSA is a chronic condition characterized by partial or complete collapse of the upper airway during sleep. Laparoscopic bariatric surgery is one of the fastest growing surgical procedures in the United States to promote weight loss in morbidly obese patients. The number of patients undergoing bariatric surgery that have OSA is estimated to be >70%. However, the majority of these patients are neither diagnosed nor appropriately screened for OSA. The purpose of this paper is to discuss evidenced-based strategies to identify patients with OSA and to reduce postoperative respiratory dysfunction in obese patients undergoing laparoscopic bariatric surgery.
Introduction
OSA is one of many significant health implications of obesity. OSA is a chronic condition characterized by partial or complete collapse of the upper airway during sleep. 3 Frequently, patients with OSA experience periods of apnea or hypopnea. Apnea in OSA patients is defined as a period of total cessation of airflow for 10 seconds or more despite repeated attempts at respiration against a closed glottis. 3 Hypopnea is defined as a 50% reduction in airflow or a decrease in airflow sufficient enough to cause a 4% drop in arterial oxygen saturation. 3 Additionally, patients with OSA snore and have excessive daytime symptoms such as sleepiness, neurocognitive impairment, and morning headaches. Inevitably, OSA may lead to hypoxemia, hypercapnia, pulmonary and systemic vasoconstriction, and secondary polycythemia. Therefore, patients with OSA have an increased sensitivity to the respiratory depressant effects of opioids, anxiolytics, and general anesthetics, which may contribute to prolonged length of hospital stay.3,4
Patients with OSA may present with several anesthetic challenges, such as difficult mask ventilation, tracheal intubation, extubation, and adequate postoperative pain management without respiratory compromise. According to Mulgrew et al., 5 (p157) “symptomatic obstructive sleep apnea is a common, underdiagnosed condition that occurs in 4% of men and 2% of women.” In fact, the prevalence of OSA in the surgical population is much higher than the prevalence of OSA in the general population. 6 An estimated 70–80% of bariatric surgical patients have OSA, with a great majority of OSA patients neither diagnosed nor appropriately screened. 6 Chung et al. 6 note that nearly 80% of men and 93% of women with moderate to severe sleep apnea are undiagnosed. Patients with untreated OSA are known to have increased “incidence of difficult intubation, postoperative complications, ICU admissions, and longer duration of hospital stay.” 6 Additionally, untreated OSA patients have an estimated average lifespan of approximately 58 years, which is approximately 20 years less than the average lifespan of the general public. 7 The first step in preventing postoperative complications due to OSA is to identify patients with OSA. 6 Identification of patients with OSA is paramount in preventing postoperative complications.
The purpose of this paper is to provide a thorough review of the anesthesia research in order to determine the best practices for assessment and management of the patient with OSA undergoing laparoscopic bariatric surgery. Specifically, anesthetic strategies to manage and reduce postoperative respiratory dysfunction in patients with OSA will be explored.
Review of Literature
Preoperative management
In an ideal world, an obese patient with signs and symptoms of OSA would be followed by a primary care provider and thoroughly evaluated for the disorder. This would include a thorough interview, physical assessment, and a polysomnography, which is also known as a sleep study. Based on the results of the polysomnography, a diagnosis of OSA is confirmed and the patient is subsequently prescribed a continuous positive airway pressure (CPAP) device and regime. Although the polysomnography is the diagnostic standard for OSA, it is time consuming, costly, and can be difficult to schedule due to long waiting lists. 6 As a result, the sleep study is often not obtained preoperatively.
There are several additional screening tools that can be used to identify patients who are likely to have OSA. One of these screening tools is the Berlin questionnaire, 8 which addresses three items: presence and frequency of snoring behavior, waketime sleepiness or fatigue, and history of obesity or hypertension. Symptoms in any two of the three domains that are persistent and frequent place a patient at high risk for OSA. 8 While the Berlin questionnaire is an effective tool for identifying patients with OSA in the primary care setting, 8 it has yet to be validated as a screening tool for OSA in surgical patients. 6
In a study performed by Chung et al., 6 the concise and easy to use “STOP” questionnaire was developed and validated to screen for OSA in surgical patients. The four-item STOP questionnaire is a self-report, forced-choice (yes/no), paper-and-pencil scale that takes approximately 1 min to complete (see Table 1). In the STOP questionnaire, answering yes to two or more questions indicates a high risk for OSA and answering yes to fewer than two questions indicates a low risk for OSA.
Questionnaire reprinted with permission from Wolters Kluwer. 6
The sensitivity of the STOP questionnaire for detection of moderate to severe OSA was found to be 65.6% with a specificity of 60%, a positive predictive value (PPV) of 78.4%, and a negative predictive value (NPV) of 44%. 6 The acronym BANG (B—BMI > 35; A—age >50; N—neck circumference >40 cm; G—gender male) was later added to identify certain clinical characteristics to the STOP questionnaire to achieve an even higher level of sensitivity. 6 Therefore, the STOP-BANG questionnaire has been identified as a quick and concise means for an anesthesia provider to determine whether a patient has OSA and is at risk for potential complications (see Table 2). In the STOP-BANG questionnaire, answering yes to three or more items indicates a high risk of OSA, and answering yes to fewer than three items indicates a low risk for OSA.
Questionnaire reprinted with permission from Wolters Kluwer. 6
As stated previously, preoperative preparation is an extremely important step toward minimizing postoperative pulmonary risk. Preparation should include preoperative CPAP, noninvasive positive-airway pressure (NIPPV) or bilevel positive airway pressure (BiPAP), preoperative medications, and weight loss. 9 Preoperative patient compliance with CPAP should not be underestimated. Stierer and Collop 10 found that patients with OSA who were compliant with preoperative CPAP use had fewer postoperative complications when compared to those who were noncompliant with preoperative CPAP use. Furthermore, patients who have had corrective airway surgery (i.e., uvulopalotopharyngoplasty) should still be considered at risk for OSA complications unless a normal sleep study has been obtained and there has been a cessation of OSA symptoms. 9
Intraoperative management
The intraoperative choices made by anesthesia providers for patients with OSA can minimize the potential of postoperative respiratory compromise. First and foremost, patients who are known to have OSA, or who are suspected of having OSA, may have a difficult airway and should be managed according to the difficult airway algorithm developed by the American Society of Anesthesiologists (ASA). As most anesthesia providers have experienced, securing the airway is not the only difficult task when managing a morbidly obese patient with OSA; ventilating the morbidly obese patient with OSA can be equally as difficult, if not more so.
The anesthesia providers' primary goal in ventilatory management is preventing or minimizing atelectasis. 11 This is extremely important because atelectasis often occurs after induction of anesthesia in the nonobese population. Obese patients have a greater propensity of developing atelectasis than the nonobese because of decreased functional residual capacity (FRC) and decreased chest wall and lung compliance. 11 Additionally, there is an increased occurrence of postoperative atelectasis associated with laparoscopic bariatric surgery due to the induced pneumoperitonium that causes compression of basil lung regions. 11
In 2009, Talab et al. 11 conducted a random, prospective, double blind, controlled study of intraoperative ventilatory strategies for preventing pulmonary atelectasis in obese patients undergoing laparoscopic bariatric surgery. The study involved 66 obese adults with a BMI of 30–50 kg/m2. The participants were aged between 20 and 50 years old and scheduled to undergo laparoscopic bariatric surgery. All participants received the same preoperative medications of metoclopramide, ranitidine, and oral lorazepam 1 h before induction. The participants were preoxygenated with 100% FIO2 for 3–5 min, followed by the same weight-based dose of propofol, fentanyl, and rocuronium. Anesthesia was maintained with 2% sevoflurane, fentanyl, and rocuronium. All patients were ventilated with volume control ventilation, 50% FIO2, and a tidal volume of 8–10 ml/kg, based on ideal body weight. The end-tidal carbon dioxide was maintained between 32 and 36 by adjusting the respiratory rate. Pneumoperitonium was achieved with abdominal insufflation of carbon dioxide, with intraabdominal pressure maintained for the duration of the procedure between 11 to 15 mmHg.
The patients were randomly assigned to one of three groups: Zero End Expiratory Pressure group (ZEEP), Positive End Expiratory Pressure (PEEP)-5 group, and the PEEP-10 group. All participants were given a vital capacity maneuver (VCM) after intubation, which was maintained for 7–8 s. Afterwards, the participants were placed on the ventilator with zero PEEP, PEEP-5, and PEEP-10, depending on group assignment. Partial pressure of oxygen (PaO2) was sampled before and after surgery via arterial blood gas sample. A chest computed tomographic (CT) scan was obtained on each participant on admission to the hospital and immediately after discharge from the postanesthesia care unit (PACU) in order to assess the presence and degree of atelectasis. The radiologist was made aware of the study protocol but was blinded to patient group assignments. The degree of atelectasis was then classified into four types.
There were no statistically significant differences between the three groups regarding age, sex, ASA classification, duration of surgery, or BMI. 11 There was a significant decrease of postoperative A-a gradient in the PEEP-10 group compared with both the ZEEP and PEEP-5 group. While in the PACU, five patients in the ZEEP group required 100% FIO2 via a nonrebreathing 02 mask, whereas only three patients in the PEEP-5 group and one patient in the PEEP-10 group required 100% FIO2 via a nonrebreathing O2 mask. The study also concluded that within the first 48 h postoperatively, none of the patients in the PEEP-10 group had significant episodes of desaturation, chest infection, or bronchospasm as opposed to the ZEEP group and PEEP-5 group, which had four and three patients respectively. Postoperatively, the PEEP-10 group had four patients with segmental and lobar atelectasis when compared to the ZEEP and PEEP-5 group, which had 14 and 9 patients respectively. When comparing the atelectasis scores, there were no significant differences found between the ZEEP and PEEP-5 groups.
In conclusion, Talab et al. 11 concluded that a VCM followed by 10 cm H20 of PEEP was associated with better oxygenation during and after surgery, as well as lower atelectasis scores based on CT scans done approximately 2 h postoperatively. Another study conducted by Coussa et al. 12 supports the conclusion of Talab et al. 11 that the application of PEEP (10 cm H20) is effective in preventing atelectasis in the morbidly obese patient population.
Immediate postoperative management
Postoperative lung function
In the past, it has been a concern that the use of CPAP immediately postoperative can disrupt suture lines after laparoscopic Roux-en-Y gastric bypass surgery (RYGBP). However, several recent studies claim that postoperative use of CPAP does not disrupt the anastomosis.13,14 According to Neligan et al., the application of CPAP immediately after extubation can be beneficial. 14 Neligan et al. 14 conducted a prospective, observer-blinded, randomized control study in order to evaluate whether morbidly obese patients with OSA lose significant lung volume in the first hour postoperatively, if the application of CPAP immediately after extubation preserves lung volume, and whether CPAP continued for 1 h postoperatively recovers lost lung volume. Forty morbidly obese participants with OSA undergoing laparoscopic bariatric surgery were randomly divided into two groups. Baseline pulmonary function tests were obtained via spirometry and compared to figures obtained postoperatively. The anesthesia care was standardized for all participants in the study. The Boussignac group (BG) participants were extubated and immediately attached to the Boussignac CPAP system, which is a portable facemask system that is attached to oxygen cylinders delivering 25 L/m. This system was confirmed to deliver approximately 10 cm/H20 of CPAP. The standard group (SG) participants were extubated and placed on 4–6 L/min of oxygen via nasal cannula.
The investigators found no significant differences in the baseline pulmonary function tests obtained preoperatively, and all participants had a statistically significant reduction in pulmonary function test values from baseline. 14 However, the FEV1 in the SG participants was reduced by 50% from baseline as opposed to 27% in the BG participants. The researchers demonstrated that bariatric patients with OSA lose significant lung volume immediately after extubation, which can be minimized with the immediate application of CPAP after extubation. The delay in administration of CPAP until arrival in the PACU, which is approximately 30 min later, is inadequate and does not restore lost lung volume.
Reducing the use of opioids
Postoperative respiratory dysfunction readily occurs in obese patients with OSA secondary to the respiratory depressive effects of opioids, sedatives, and anesthetics. Therefore, these medications should be restricted perioperatively in this patient population. 15 In order to avoid hypoxic episodes in obese patients with OSA, there is an increased likelihood of the need for postoperative mechanical ventilation secondary to the increased sensitivity of the respiratory depressant effect of opioids. 16 In reviewing the literature, several research articles reported that the perioperative administration of dexmedetomidine (precedex) was successful in restricting the use of opioids, sedatives, and anesthetics, thus reducing postoperative respiratory dysfunction.
Dexmedetomidine is a potent alpha-2 adrenergic receptor agonist that has hypnotic, sedative, anxiolytic, sympatholytic, and analgesic properties, but does not produce significant respiratory depression. 17 In addition, it decreases the minimal alveolar concentration for volatile anesthetics from 35% to 50%. 3 The hypnotic effect produced from dexmedetomidine is largely due to depressing the function of the locus coeruleus (a supraspinal site), which evidence suggests controls sleep, attention, memory, analgesia, and autonomic function. 3 The analgesia-sparing effect produced from dexmedetomidine is via central actions in the dorsal horn of the spinal cord. 18
Dexmedetomidine has a rapid onset of action (<5 min) and peak effect occurs ∼15 min after intravenous administration. 3 The cytochrome P-450 system of the liver metabolizes dexmedetomidine and is eliminated by the kidneys. 19 With the patient monitored, a 1 mcg/kg loading infusion (not a bolus) must be given over 10 min followed by a continuous infusion of 0.2 to 0.7 mcg/kg/h. 3 Chalikonda 19 (p105) explains, “Large intravenous bolus doses can result in hypertension [from activation of alpha-2 receptors on vascular smooth muscle] and bradycardia, but because it [dex] exhibits a biphasic pharmacokinetic profile it will then cause hypotension from vasodilation as the serum concentration declines. Therefore, administering a loading infusion of of dexmedetomidine over ten minutes followed by a continuous infusion is recommended.”
In 2007, Dholakia et al. 20 conducted a retrospective study of 71 patients examining the association between postoperative narcotic use after bariatric surgery, duration of hospital stay, and perioperative dexmedetomidine administration. The study group consisted of 34 patients that had undergone either laparoscopic gastric bypass or gastric band surgery after a new clinical protocol was instituted in 2005 at one hospital. Patients who were taking prescriptive narcotics preoperatively or who had experienced major surgical complications contributing to increased pain or length of hospital stay were excluded from this study.
The protocol consisted of administrating a dexmedetomidine infusion starting 30 min before the anticipated completion of surgery to all patients undergoing laparoscopic bariatric surgery. A loading dose of 1 mcg/kg IV over 10 min was administered followed by an infusion at 0.2–0.7 mcg/kg/h until the end of surgery; the infusion was discontinued in the PACU. 20 The control group consisted of 37 patients that had also undergone either laparoscopic gastric bypass or gastric band surgery prior to the institution of the new clinical protocol, thus they did not receive dexmedetomidine. Dholakia et al. 20 reported that the laparoscopic gastric bypass patients in the study group received fewer narcotics during their hospitalization (an average of 66 mg morphine in dexmedetomidine group compared to 130 mg in control group), met discharge criteria on postop day 1 more often (i.e., 61% patients in dexmedetomidine group compared to 21% patients in control group), and were discharged home sooner than patients in the control group (1.4 days in dexmedetomidine group compared to 1.9 days in control group). However, in the gastric band patients, the requirement for narcotic pain medication and the duration of hospitalization were similar in each group.
According to this study, the benefits of administering dexmedetomidine to patients undergoing bariatric surgery include adequate pain control with a narcotic-sparing effect. 20 In a population of patients with a high prevalence of OSA, this may have important safety ramifications. In fact, other studies in nonbariatric18,21 and bariatric16,17 surgery patients have confirmed this finding.
In 2007, Bakhamees et al. 16 conducted a study that evaluated the effect of dexmedetomidine on anesthetic requirements during surgery, hemodynamics, recovery profile, and on morphine use in the postoperative period. The study consisted of 80 morbidly obese patients scheduled for laparoscopic RYGBP surgery that were equally randomized into two groups. Forty patients received dexmedetomidine 0.8 mcg/kg bolus followed by 0.4 mcg/kg/h infusion, which was discontinued at the removal of the laparoscopy ports. The other 40 patients received normal saline in the same volume and rate. 16 A propofol infusion of 10 mg/kg/h was used to maintain anesthesia, which was titrated to maintain a bispectral index (BIS) level between 40 and 60. When comparing the dexmedetomidine group to the control group, the study concluded that mean arterial blood pressure and heart rate, the total amount of propofol required to maintain the target BIS level, and the total amount of intraoperative fentanyl required to maintain hemodynamics were significantly lower in the dexmedetomidine group. The recovery profile, which consisted of the duration to spontaneous respiration, adequate respiration, and safe extubation, was significantly shorter in the dexmedetomidine group compared to the saline group. 16 In addition, patient pain scores, the total amount of PCA morphine used, blood pressure, and heart rate were significantly lower in the dexmedetomidine group while in the PACU.
In 2008, Tufanogullari et al. 17 conducted a study consisting of 80 morbidly obese patients that evaluated the effect of dexmedetomidine on both early and late recovery after laparoscopic bariatric surgery. The patients were randomly assigned to one of four groups. 17 Group 1 received a saline infusion, group 2 received a 0.2 mcg/kg/h dexmedetomidine infusion, group 3 received a 0.4 mcg/kg/h dexmedetomidine infusion, and group 4 received a 0.8 mcg/kg/h dexmedetomidine infusion. 17 No loading dose was administered in any of the dexmedetomidine groups. The results of particular interest include a reduced average end-tidal desflurane concentration in groups 2, 3, and 4 by 19, 20, and 22% respectively, and significantly less fentanyl was administered in the PACU for patients in groups 2, 3, and 4 versus the control group (113 ± 85 mcg, 108 ± 67 mcg, and 120 ± 78 mcg vs. 187 ± 99 mcg respectively, p < 0.05). 17
Minimizing the use of anesthetic agents
As stated earlier, the restrictive use of anesthetic medications perioperatively is beneficial in reducing postoperative respiratory dysfunction in obese patients with OSA. Bispectral index monitoring (BIS) is one of several monitors currently available that estimate the depth of anesthesia through analysis of brain-wave monitoring. A BIS value between 40 and 60 is an appropriate level for general anesthesia. The use of BIS monitoring during anesthesia for bariatric surgery allows the anesthesia provider to titrate anesthetic agents based on the patient's need, therefore decreasing the overall anesthetic consumption. This results in a more rapid emergence from anesthesia with minimal respiratory adverse effects. For instance, Ibraheim et al. 22 conducted a study of 30 morbidly obese patients equally divided into two groups that had undergone laparoscopic gastric banding surgery in order to assess whether titrating sevoflurane based on BIS monitoring would shorten recovery time. Sevoflurane was titrated to maintain a BIS value between 40 and 60 during surgery, and then titrated to maintain a BIS value of between 60 and 70 during the last 15 min of the procedure. The patients monitored with the BIS had awakening and extubation times that were significantly shorter (p < 0.05) than those monitored without the BIS. The awakening time was 6.80 ± 2.14 min in the BIS group compared to 8.66 ± 2.69 min in the non-BIS group. The extubation time was 9.26 ± 2.01 min in the BIS group compared to 11.8 ± 2.9 min in the non-BIS group. The study concluded that BIS monitoring during anesthesia provides statistically significant reductions in recovery times and that “a rapid emergence from anesthesia with minimal respiratory and cardiovascular adverse effects is important in morbidly obese patients who have a high prevalence of cardiovascular disease and are at risk for respiratory complications.” 22 (p827)
The pharmacokinetics of many drugs used in anesthesia including inhalation agents are different in obese patients compared to the nonobese patient. In particular, the blood–gas coefficients of inhalation agents, which reflects the solubility of the agent in the blood, is more important in this patient population. When an inhalation agent has a high blood–gas partition coefficient, the onset of effect is slower and recovery is slower, and when an inhalation agent has a low blood–gas coefficient, the onset of effect is faster and recovery is faster. The blood–gas partition coefficient of desflurane is approximately 30% less than sevoflurane, and given these facts, one could theoretically ascertain that desflurane would allow for a more rapid emergence in obese patients when compared to sevoflurane. 23 Nonetheless, there is still some degree of uncertainty of significantly better recovery profiles between desflurane and sevoflurane.
In 2007, La Colla et al. 4 compared the pharmacokinetics and pharmacodynamics of desflurane and sevoflurane in a random prospective study of 28 obese patients undergoing laparoscopic bypass surgery without any type of premedication. Fourteen patients were placed in one of two groups, and standard monitors, including BIS, were insured. Sevoflurane 2% and desflurane 6% were administered to the participants according to group assignment. Upon cessation of the volatile anesthetic, the end-tidal concentrations of the anethestics were recorded for the first 5 min. The researchers also recorded the time from cessation of anesthetic drugs to eye opening on verbal command, squeezing the observer's hand, extubation, verbalizing name and date of birth, and discharge from the PACU. The authors concluded that desflurane provides significantly faster onset and offset than sevoflurane in morbidly obese patients who were not premedicated. 24
Interestingly, Arain et al. 25 conducted a randomized prospective blinded study that examined the emergence profiles of 40 morbidly obese patients. The researchers concluded that, with careful drug titration in the morbidly obese population, no significant differences between sevoflurane and desflurane profiles could be determined. Mantouvalou et al. 26 supported these findings in a study of 206 morbidly obese patients that underwent laparoscopic gastric banding. Mantouvalou et al. 26 concluded that the only difference between patients who received either anesthetic was that those who had received sevoflurane had lower oxygen saturation values upon arrival to the PACU (94 ± 2%) as opposed to the desflurane group (97 ± 4%). This was not the case upon discharge from the PACU, however, Otherwise, there were no other statistical differences in emergence and recovery profiles between the two groups if the anesthetic concentrations were carefully titrated based on hemodynamic and BIS values. Finally, De Baerdemaeker et al. 27 reported similar results in a randomized prospective blinded study of PACU observers who compared the recovery profiles of 50 morbidly obese patients that received either sevoflurane or desflurane during surgery in conjunction with a remifentanyl infusion. They reported no clinically significant differences in emergence profiles between the sevoflurane and the desflurane recipients with the exception of postop nausea duration being shorter in the sevoflurane group.
Conclusion
In summary, decisions made throughout the perioperative period regarding the care of patients with OSA undergoing laparoscopic bariatric surgery can minimize the likelihood of pulmonary complications. Based on the review of literature, the preoperative identification of patients with OSA is of utmost importance. Once patients with OSA have been identified, clinical decisions focus on reducing the risks of pulmonary complications. The STOP-BANG tool is a concise and valid means of identifying patients with OSA. Additionally, a vital capacity maneuver (VCM) conducted immediately after intubation, coupled with intraoperative ventilation with 10 cm PEEP, has proven to reduce the occurrence of pulmonary atelectasis in patients with OSA.
Furthermore, the application of CPAP immediately after extubation has proven to preserve preoperative lung volumes. The opioid sparing effects of dexmedetomidine in the OSA patient population is extremely beneficial and should be considered as an adjunct in the plan of care. Additionally, careful titration of anesthesia with the use of BIS monitoring in OSA patients has demonstrated positive outcomes in recovery profiles regardless of the inhalation anesthetic agent used intraoperatively. In conclusion, the incorporation of the multiple evidenced-based strategies may reduce the risk of pulmonary dysfunction, reduce recovery room time, and reduce length of hospital stay in patients with OSA who present for laparoscopic bariatric surgery.
Footnotes
Disclosure Statement
No competing financial interests exist.
