Sleep Apnea and Narcotic Postoperative Pain Medication: A Morbidity and Mortality Risk
by Ann Lofsky, MD
Sleep apnea has long been a concern for anesthesiologists because of its association with difficult intubation and recovery room complications. Claim reviews at The Doctors Company (TDC) have revealed a number of recent cases of "unexplained" postoperative cardiopulmonary arrests occurring in hospitalized patients. All of these patients received parenteral narcotic pain medication and were ultimately diagnosed with sleep apnea. An Anesthesia Advisory panel was convened in Napa, California, in March 2001, to discuss 8 of these claims. TDC believes that the trend exhibited by these cases is sufficiently alarming to present the following panel summary with the hope that this cause of major morbidity and mortality may be avoided.
The following are composite cases incorporating details from the 8 cases reviewed:
What is sleep apnea syndrome?
Apnea is defined as the absence of airflow at the nose and mouth for longer than 10 seconds.1, 2 Sleep apnea syndrome (SAS) is described as the occurrence of more than 30 apnea episodes over a 7-hour period of nocturnal sleep.1,2,3,4 The majority of sleep apnea cases are obstructive as opposed to central. The upper airway becomes impeded while respiratory effort and diaphragmatic contraction still occur. In the absence of formal sleep studies, the diagnosis can be made clinically by interviewing patients and their sleeping partners. Clinical signs include loud snoring, observed apneic episodes, and excessive daytime somnolence.3
This is a common disorder. It is estimated that its prevalence in the middle-aged population is 1-4%, although it can occur at any age.4,5 The age range seen in the 8 cases reviewed ranged from 3 to 53 years. SAS patients are predominately male. Affected females are more likely to be postmenopausal or morbidly obese.5 In the cases we reviewed, 6 of the 8 patients were male, and both females were morbidly obese. Obesity itself is a risk factor for SAS, especially if it affects the neck.5
Why are SAS patients more prone to significant episodes of airway obstruction?
The cause of upper airway obstruction in SAS patients was long thought to be the tongue, but fluoroscopy during sleep has shown this is rarely the case. The oropharynx itself is the most collapsible segment of the upper airway and the most likely site of obstruction.2 Physiologic studies have demonstrated that patients with SAS have narrowed upper airways to start with, so they are more susceptible than other patients to drugs or anesthetics that suppress pharyngeal muscle tone.6,7
In normal, awake patients, there is a phasic activity of the pharyngeal muscles that contracts them immediately before inspiration, helping to resist the negative pressure generated by the diaphragm and keeping the airway from collapsing.2,8 This phasic pharyngeal contraction is markedly reduced both by REM sleep and by narcotic pain administration.4,5 Patients with sleep apnea appear to be much more sensitive than normal individuals, even to minimal levels of sedation.5 The increased sensitivity of their hypoglossal nerves to low doses of anesthesia has been well described.9
This increased tendency to airway obstruction can occur out of proportion to the level of sedation the narcotics achieve.9 This would explain why many of the patients complained of significant pain shortly before falling asleep and developing obstructive episodes. Apneic episodes can occur with all routes of narcotic administration. In the 8 cases reviewed, 2 patients received narcotics intravenously by PCA, 1 spinally, 1 epidurally, and 4 by intravenous or intramuscular injections. The increased sensitivity of SAS patients to narcotics may be impossible to reverse with antagonists. One well-documented case in the literature described an obese patient who became comatose after a minimal dose of meperidine given intramuscularly as a premedication. Neither the airway obstruction nor the obtundation could be satisfactorily reversed with naloxone.7 When the patient had spontaneously recovered, his physicians ordered a sleep study, which confirmed the diagnosis of sleep apnea. This patient had a documented decrease in arterial pO2 to 30 mm Hg spontaneously during sleep associated with multiple PVCs and episodes of sinus bradycardia.7 Had this not been picked up intraoperatively, it might have happened unmonitored on the ward.
One of the first problems in planning the care of SAS patients is identifying them. The vast majority of surgical patients do not have preoperative sleep studies. Four of the cases we reviewed did carry the diagnosis of sleep apnea preoperatively, and 2 slept with CPAP (continuous positive airway pressure) devices. Although many authors have described methods of clinically diagnosing this syndrome,3,4,5,7,10 anesthesiologists and surgeons are not routinely trained to elicit this type of information.
The diagnosis of sleep apnea should be suspected in patients with the classical body habitus of obesity with a short, thick neck.7 A history of loud snoring as reported by the sleeping partner is a cardinal sign mentioned in almost all case reviews.3,4,5,7,10 One study suggested that snoring be considered significant if:
Although these signs were mentioned by a number of authors, anesthesiologists do not routinely ask patients about them. Perhaps not all patients exhibiting these symptoms will have sleep apnea if evaluated by formal sleep studies, but it might be safer to treat them as if they did until proven otherwise.
Panel members were struck by the fact that all of the cases reviewed could have been prevented by audible pulse oximeter monitoring on the ward. Although this is not currently the standard of care, the apparent high incidence of respiratory obstruction in SAS patients should make this a strong consideration, especially if narcotic analgesics are to be used.
Literature review suggests that the traditional nursing measurement of respiratory rate is ineffective in SAS patients for a number of reasons. Critical hypoxemia occurred during sporadic episodes of obstruction with continued respiratory effort that could be difficult to differentiate from effective respiration.10 These episodes were only rarely associated with slow respiratory rates (fewer than 10 breaths per minute), making rate alone an insensitive indicator of compromise.2,8
The prophylactic use of supplemental oxygen in SAS patients is controversial. The respiratory drive during apneic periods appears to be hypoxia. Carbon dioxide retention does not occur to a significant degree during obstructive episodes, while oxygen partial pressure falls quite rapidly.3,5 In SAS patients, the incidence and duration of apneic periods have been reported to increase with supplemental oxygen use, presumably because of removal of the hypoxic drive.8 If oxygen is to be used, pulse oximetry would again appear to be prudent.
Children with SAS secondary to adenotonsillar hypertrophy have been shown to have a decreased ventilatory response to CO2 when compared with controls. This has been suggested as an explanation for their propensity for respiratory compromise. Children with sleep apnea syndrome, like adults, usually present with histories of snoring, apneic episodes, and daytime hypersomnolence, and they are more likely than controls to be obese.11
The administration of narcotic pain medication in patients with sleep apnea syndrome needs to be closely monitored. One problem observed by the panelists was that pain medication orders for any given patient might be written by different individuals (surgeon, anesthesiologist, or primary care practitioner), not all of whom may be aware of the diagnosis of SAS. Clearly, wider understanding of this syndrome is crucial, and suggestions included red-flagging the charts of these patients to warn of the risks of narcotic usage.
Patients who use CPAP devices at home, as did 2 of the patients in the cases reviewed, should be advised to bring the mask into the hospital for use postoperatively.
As pain is treated more aggressively, the tragic complication of respiratory arrest in patients with sleep apnea syndrome may be seen more frequently. The panelists unanimously suggested that anesthesiologists need to be alert to signs of SAS and should consider being aggressive in asking questions that would help identify patients at risk.
Patients suspected of having SAS need to be handled differently than routine postoperative patients--especially with painful procedures that require significant narcotic pain medication postoperatively. Anesthesiologists may be able to help get this information to surgeons and internists who might also be writing parenteral narcotic orders for these patients. The panelists also hope hospitals can find ways to monitor SAS patients for oxygen saturation more effectively. Although the clinical outcomes in the claims reviewed were devastating, they suggest that further cases may be preventable.
Dr Lofsky is an anesthesiologist in private practice in Santa Monica, California, and is a member of The Doctors Company Board of Directors.
Anesthesiology Advisory Panel Members
(all TDC insureds and board-certified anesthesiologists):
Malcolm Donnell, MD, Texas; Greg Lind, MD, Montana; Ann Lofsky, MD, California; Don Peterson, MD, Washington; Geraldine Pulito, MD, Oregon; Ron Stevens, MD, Colorado; Tod Tolan, MD, Oregon; Dean Walker, MD, California; Elliott Wohlner, MD, Colorado.