Undiagnosed obstructive sleep apnea can prove dangerous to surgery patients coming out of sedation, but one program seeks to reduce the risk.

All people who undergo surgery are at risk for complications. However, this risk is increased if the person who is having the operation also has an underlying medical problem, such as cardiac or pulmonary disease. Patients may be especially vulnerable if this underlying condition is undiagnosed. Much of preoperative health assessment has focused on detecting signs of heart and lung disease prior to surgery. However, there has been little emphasis placed on sleep disordered breathing in general and obstructive sleep apnea (OSA) in particular.¹

At Barnes-Jewish Hospital in St Louis, medical professionals have become concerned about undiagnosed OSA in recent years. Several patients with no known medical problems experienced unexpected postoperative respiratory complications. In some instances, on retrospective review, undiagnosed OSA was considered to be a major contributing factor.

As a result of this concern, the Perioperative OSA Project was initiated as a collaborative venture among Washington University physicians and Barnes-Jewish Hospital staff. Its mission was to evaluate the extent of OSA among surgical patients and to institute processes to improve perioperative safety for patients with OSA. This initiative has had the generous backing of the Barnes-Jewish Foundation.

Coincident with our local initiatives, the American Society of Anesthesiologists (ASA) commissioned a national, multidisciplinary task force of sleep medicine specialists, intensive care physicians, surgeons, respiratory therapists, and anesthesiologists to develop guidelines for the perioperative management of patients with diagnosed and suspected OSA. These guidelines were adopted at the ASA’s national meeting in Atlanta in October of last year and published in the journal Anesthesiology in May. The guidelines establish new standards for diagnosing, monitoring, and treating patients with OSA in the perioperative period.²

OSA: At Risk in the Perioperative Period
OSA afflicts an estimated 2% to 4% of the US adult population, but it is estimated to be 80% to 90% undiagnosed.^3,4 Typically, people with this disorder are aroused repeatedly from deep sleep by hypoxemia and hypercapnia, which occur during episodes of apnea. Such arousals are protective as they allow breathing to resume (Figure 1, page 54). However, residual effects of general anesthesia, opioid analgesics, and sedative agents may blunt this protective arousal mechanism, potentially resulting in respiratory arrest. Until recently, there has been no universally accepted approach to diagnosing OSA and optimizing the care of patients with OSA in the perioperative period.

Patients undergoing surgical procedures often are given narcotic analgesia postoperatively. With limited awareness of OSA, many of these patients are sent home with their analgesics the same day of their surgery to a completely unmonitored environment.⁵ Those patients who do stay in hospital are often sent to a regular ward with minimal monitoring. As Figure 1 on page 58 shows, should an obstructive sleep apnea then occur, the consequences could be serious as the patient’s body may not arouse enough to resume breathing and no one may notice the interruption in breathing.

Furthermore, in addition to the specific perioperative risks of respiratory compromise, OSA is associated with other diseases, which may in themselves increase perioperative morbidity. Poorly controlled hypertension, right heart failure, type 2 diabetes, pulmonary hypertension, coronary artery disease, arrhythmias, and stroke are examples. The increase in obesity is creating an epidemic of OSA; 50% of morbidly obese patients have OSA. In Missouri, 22% of adults are obese.⁶

The Perioperative OSA Project
At Barnes-Jewish Hospital, no protocol for the screening of surgical patients for OSA or guidelines for the perioperative management of patients with OSA existed prior to the Perioperative OSA Project. Postoperative pulse oximetry also was unavailable outside of the intensive care units and high-dependency units, and there were no remote alarm systems to alert staff on the surgical wards when patients had episodes of apnea or oxygen desaturation. We, therefore, felt that it was a safety priority to identify surgical patients with OSA who may be at highest risk for postoperative respiratory compromise and to institute system changes to protect those who were most vulnerable.

Preoperative Process
Our first aim was to try to quantify how many adult surgical patients had OSA, both diagnosed or undiagnosed. The prevalence of OSA among surgical patients at Barnes-Jewish Hospital—or, for that matter, any hospital—is unknown.

In our Center for Preoperative Assessment and Planning, patients routinely undergo comprehensive evaluations prior to their upcoming surgeries. To these evaluations, we added the ARES screening questionnaire that assigned a risk level for OSA: no risk, low risk, moderate risk, and high risk. The questionnaire incorporated known risk factors for OSA, as well as symptoms that are associated with OSA. Risk factors included male gender, obesity, age greater than 40, a family history of OSA, neck circumference greater than 17 inches for men and 16 inches for women, a recessed chin, smoking, alcohol consumption, and upper-airway abnormalities. Symptoms included loud snoring, excessive daytime sleepiness, and periods of apnea while sleeping.²

A definitive diagnosis of OSA is usually made only in conjunction with a formal sleep laboratory study. However, with the volume of patients we were screening, laboratory sleep studies would have been impractical. In general, with the high prevalence of OSA among adult Americans, laboratory sleep studies may not be feasible for confirming all the diagnoses.

Many portable devices have been developed that someday may allow greater use of home diagnosis methods for OSA. A study of one such device, the ARES Unicorder, showed that it compared favorably with formal sleep studies in its ability to diagnose OSA.⁷ The device is also one of the few such products that the US Food and Drug Administration has approved for the home diagnosis of OSA. We, therefore, chose to offer all the patients who were identified as being at high risk of having OSA in our screening questionnaire a Unicorder to take home and use for 2 nights of sleep prior to their surgeries.

The device works through an internal computer chip that stores continuous data on the patient’s oxygen saturation, pulse rate, airflow, head position, and snoring decibel level. When the patient returns the device on the day of surgery, a member of our team downloads the information to obtain a report detailing the respiratory disturbance index (RDI) and apnea/ hypopnea index (AHI)—two measures of OSA severity. This report is then sent to the patient’s primary care physician, and, if the measures recorded by the Unicorder indicate that the patient likely has OSA, we suggest that the primary care physician refer the patient to a sleep laboratory for a formal assessment of OSA.

The ASA recommends risk stratification for patients with OSA. This may be useful to individualize the risk of perioperative complications. Table 1 (above) shows the scoring system. Based on the severity of OSA, the invasiveness of the surgery, and the requirement for postoperative analgesics, a calculated score distinguishes among patients at low, increased, and significantly increased risk for perioperative complications. This scoring system has yet to be validated and should be used as an adjunct to clinical judgment.

In addition to offering an ARES Unicorder to patients who screen high risk for OSA, we flag them as being at risk for OSA with various markers in purple. The preoperative assessment sheet is stamped in purple with the words “OSA Risk” where this can be seen easily by the attending anesthesiologist and surgeon. In addition, a purple sheet is inserted into the patient’s chart stating that the patient is at risk for OSA. This sheet includes the ASA perioperative risk-stratification score for OSA and the phone number for sleep-medicine clinic and respiratory therapy consultations. Finally, a purple wristband stating “OSA Risk” is placed in the chart to be worn on the patient’s wrist on the day of the surgery. A list of all the patients who screen “high risk” for OSA, as well as those with an established OSA diagnosis, is faxed to the respiratory therapy department, the sleep medicine team, and the surgical placement coordinators.

Intraoperative Care
When patients are admitted for their surgeries, the purple wristbands with the “OSA Risk” warnings are placed on their wrists. Anesthesiologists have been encouraged to tailor their anesthetic plan accordingly. Modifications could include awake fiber-optic intubation, a bias toward a regional anesthetic technique, the use of shorter-acting drugs, and an attempt to use less sedatives and opioid analgesics. Recent literature confirms that patients with OSA are at increased risk for difficult mask ventilation and difficult tracheal intubation.⁸

Postoperative Care
New order forms at Barnes-Jewish Hospital are designed to streamline care for patients with known or suspected OSA. The most important interventions to increase safety for patients with OSA occur in their postoperative care. When these patients are recovering in the postanesthesia care unit (PACU), unless contraindicated, they are placed in the nonsupine position, sitting or on their sides. For those with OSA, the severity of the apneas is usually worse in the supine position. For patients who use CPAP or noninvasive positive pressure ventilation (NIPPV) at home, respiratory therapists will attach a CPAP or NIPPV device with the home settings to the patient as soon as practical. A recent paper showed that for patients who have undergone major abdominal surgery, the empirical use of CPAP may decrease the likelihood of respiratory compromise and the need for reintubation.⁹ Staff in the PACU try to observe patients while they are asleep. They also make three separate recordings of oxygen saturation, 15 minutes apart, without supplementary oxygen. If patients have periods of apnea or oxygen desaturation less than 90%, a physician is called and a venous or arterial blood gas is sent. If the blood gas reveals CO₂ retention, patients are kept in the PACU until this resolves, or they are sent to an intensive care or high-dependency unit for further management.

When patients with known or suspected OSA are discharged from the PACU and transported to a surgical ward, an appropriately trained staff member accompanies them. They should not lie supine during transport. Supplementary oxygen should be administered, and continuous pulse oximetry should be monitored. The nurse who accepts the patient on the ward must acknowledge that the patient has known or suspected OSA.

The most important change in the way patients with known or suspected OSA are cared for in the perioperative period has occurred on the surgical wards. In particular, the medical engineering department has connected continuous pulse oximeters to alarms in the central nursing station. The next step will be to upgrade the central telemetry system so that continuous pulse oximetry with high-level alarms is integrated. Besides pulse oximetry, other important changes have been implemented. Patients are encouraged to remain nonsupine. Vital signs are documented without supplementary oxygen. Patients are kept on supplementary oxygen until the peripheral oxygen saturation is persistently greater than 90%, even during sleep. If patients have apneas or episodes of desaturation, the nurse calls the ward physician and a blood gas may be sent to check for CO₂ retention. If patients use CPAP or NIPPV at home, this is continued on the wards. The sleep medicine service and respiratory therapists may help with the management of individual patients. Patients may be referred to an intensive care or high-dependency unit when the ward staff has concerns about possible respiratory compromise.

Conclusion
Preliminary results of our screening study suggest alarmingly that more than 20% of adult patients undergoing surgery at Barnes-Jewish Hospital may have moderate to severe OSA. In the vast majority of patients—about 75%—there has been no known previous OSA diagnosis. Awareness about OSA has escalated impressively at our institution, and within the space of a year we have revolutionized the perioperative care of these patients. We believe that we are achieving two important goals. We are now identifying patients with OSA where this diagnosis was often previously unrecognized. We have also instituted evidence-based system changes in the management of patients with known and suspected OSA. The change will hopefully impact meaningfully on patient safety and save lives of vulnerable patients at Barnes-Jewish Hospital.

Kevin Finkel, MD, is s resident in the Department of Anesthesiology at Washington University School of Medicine, St Louis. Leif Saager, MD, is a research fellow for the same department. Colleen Becker, RN, MSN, CCRN, is the director of perioperative services at Barnes-Jewish Hospital in St Louis. Heidi Tymkew, MHS, is a research coordinator for the Department of Anesthesiology. Michael Avidan, MBBCh, FCA, is deputy director of clinical research, anesthesiology and intensive care, at the Washington University School of Medicine.

References
1. Task Force on Perioperative Management of Patients with Obstructive Sleep Apnea. Practice guidelines for the perioperative management of patients with obstructive sleep apnea. Anesthesiology. 2006; 104:1081-93.

2. Young T, Palta M, Dempsey J, et al. The occurrence of sleep disordered breathing among middle-aged adults. N Engl J Med. 1993; 328:1230-1235.

3. Young T, Evan L, Finn L, et al. Estimation of the clinically diagnosed proportion of sleep apnea syndrome in middle age men and women. Sleep. 1997; 20:705-706.

4. Westbrook PR, Levendowski DJ, Cvetinovic M, et al. Description and validation of the apnea risk evaluation system: a novel method to diagnose sleep apnea- hypopnea in the home. Chest. 2005; 128:2166-2175.

5. Kim JA, Lee JJ. Preoperative predictors of difficult intubation in patients with obstructive sleep apnea syndrome. Can J Anesth. 2006; 53:393-397.

6. Squadrone V, Coha M, Cerutti E, et al. Continuous positive airway pressure for treatment of postoperative hypoxemia. JAMA. 2005; 293:589-595.