Endoscopy sedation risk factors

Does increased BMI equate to an ascendancy of risk during procedural sedation? Intuitively the answer is yes, but this is not backed by extensive data from the literature. Large database-driven studies addressing cardiopulmonary complications during gastrointestinal (GI) endoscopy did not contain BMI in their analyses. However other measures such as the American Society of Anesthesiologists (ASA) physical classification score have been found to represent a risk for cardiopulmonary events. In a retrospective study using the Clinical Outcomes Research Initiative database, over 324,737 procedures were reviewed. Multiple logistic regression found that the ASA class was a significant predictor for cardiopulmonary unplanned events. Patients with an ASA class 3 designation exhibited a risk for cardiopulmonary events that was 1.8 times that for the ASA class 1 (odds ratio [OR] 1.8; 95% confidence interval [CI]:1.6–2.0). The risk essentially doubles for each subsequent ASA class thereafter. It should also be emphasized that the calculation of the BMI is simple but has important limitations. For example, extremely muscular patients may fall into the obesity range, whereas patients who have a normal BMI may be nutritionally depleted with a decrease in lean body mass.

Data equating increasing BMI to cardiopulmonary events during endoscopy are available in small case series. A case series of 82 patients undergoing diagnostic upper endoscopy found that a BMI greater than 28 was an independent risk factor for hypoxemia. Qadeer and colleagues found that an increasing BMI was an independent risk factor for hypoxemia in a cohort of 79 subjects undergoing a variety of ambulatory outpatient procedures including upper endoscopy, colonoscopy, endoscopic retrograde cholangiopancreatography (ERCP) and endoscopic ultrasonography (EUS). Other risk factors for hypoxemia included age greater than 60 years and meperidine dose in a study by Dhariwal and colleagues.

The American Society of Anesthesiologists Task Force on Sedation and Analgesia by Nonanesthesiologists stated that there is insufficient published evidence to evaluate the relationship between sedation analgesia outcomes and the performance of a preprocedure patient evaluation. Expert opinion suggested that there is evidence that some pre-existing medical conditions may be related to adverse outcomes in patients receiving either moderate or deep sedation. One of these risk factors is significant obesity, especially involving the neck and facial structures. Similarly, the American Gastroenterological Association Institute review of endoscopic sedation recommended the use of an anesthesia professional for patients with morbid obesity. However, this is an expert opinion only, and no evidence was provided for this recommendation. The American Society for Gastrointestinal Endoscopy recommended that a history of sleep apnea, snoring, or stridor should be reviewed as part of the preprocedure assessment.

The presence of obstructive sleep apnea (OSA) has been touted as a risk factor for cardiopulmonary complications for procedural sedation. Khiani and colleagues used the Berlin Questionnaire, a validated tool for the diagnosis of OSA to determine the incidence of hypoxemia (pulse oximetry <92%) in 233 subjects undergoing elective outpatient upper endoscopy and colonoscopy with targeted moderate sedation using a combination of an opioid and benzodiazepine. Exclusion criteria included a previous diagnosis of OSA, congestive heart failure, chronic obstructive pulmonary disease, long-term narcotic use, haloperidol use or history of substance abuse, and lung disease requiring home oxygen. At the start of the procedure, 2 L of oxygen via nasal cannula was given to all patients. Of the endoscopic procedures performed, 60 (25.8%) patients underwent esophagogastroduodenoscopy (EGD); 155 (66.5%) underwent colonoscopy, and 18 (7.7%) patients underwent both procedures. Thirty-nine percent of the subjects were deemed to be at high risk for OSA by the questionnaire. The high-risk group was significantly older (60.4 plus or minus14.5 vs 55.2 plus or minus 15.2 years, P <.005) and exhibited a higher BMI (30.8 plus or minus 5.6 vs 25.9 plus or minus 4.7 kg/m ² , P <.001). There was no difference between the groups in terms of age, gender, ethnicity, or mean medication dosage. There was no significant difference in the rate of transient hypoxia between the high- and low-risk groups (OR 1.48; 95% CI 0.58–3.80). Important caveats can be gleaned from this study:

• 
  

  There was no stratification between upper endoscopy and colonoscopy. It is felt that orally intubated subjects are at higher risk for cardiopulmonary events.

  
  
• 
  

  This was a relatively healthy cohort of patients, and in the setting of relatively brief procedures targeting moderate sedation, there appears to be a minimal risk in subjects felt to be at risk for OSA.

Does snoring during moderate sedation predict OSA? Sharara and colleagues studied 131 subjects undergoing elective ambulatory colonoscopy. Each completed a detailed sleep questionnaire and physical examination focused on detecting OSA (BMI, neck circumference, and the presence of craniofacial abnormalities). Subjects who snored during colonoscopy in the left lateral decubitus position for at least 10 seconds were referred for nocturnal polysomnography. Gender- and BMI-matched patients who did not snore served as controls. All subjects who snored during colonoscopy exhibited a significantly higher Mallampati score and greater BMI and neck circumference and were more likely to report daytime sleepiness than controls. Portable nocturnal polysomnography diagnosed OSA in all snoring subjects compared with only 4 of 18 controls ( P <.001).

Sedation options

Numerous case series have been published involving elective upper endoscopy and the preoperative and postoperative assessments of bariatric surgery patients. Frequently, the medications administered were the combination of the opioid and a benzodiazepine. Unfortunately, cardiopulmonary endpoints and patient tolerance were not that the focus of these articles. For example, Catalano and colleagues performed endoscopic balloon dilation of anastomotic stenosis in 26 consecutive patients with a prior Roux-en-Y gastric bypass for morbid obesity. The patients received moderate sedation with the combination of meperidine and midazolam. Although no specific cardiopulmonary endpoints were described, no procedure related complications were seen during the course of the dilating. To the author’s knowledge, there is no prospective trial comparing the safety of monitored anesthesia care (MAC) with endoscopist-directed sedation in subjects undergoing GI endoscopy. The ASA closed claims project compared MAC with general anesthesia for injury- and liability-associated claims. Respiratory depression after an absolute or relative overdose of medication was the most common cause for adverse events in the MAC group. Nearly half of the MAC claims were judged to be preventable with better monitoring and improve vigilance.

One-hundred subjects undergoing routine preoperative assessment before bariatric surgery received either monitored anesthesia care with propofol versus a combination of narcotic and benzodiazepine administered by a surgical endoscopist. There were no significant differences between the two sedation arms with respect to subjective assessment of recovery in postprocedure recall. This study was hampered by the fact that it was not randomized, and a phone survey was used 1 to 3 months following the procedure.

Is monitored anesthesia care without risk? Coté and colleagues studied 231 consecutive patients undergoing advanced procedures such as ERCP and endoscopic ultrasound with anesthesiologist-directed propofol sedation. A previously validated screening tool for OSA known as the snore, tired, obstruction, pressure -BMI, age, neck, gender (STOP -BANG) (SB) questionnaire was used to identify patients at high risk for OSA. This screening tool uses questions such as the occurrence of snoring as well as physiologic measurements such as a BMI. A score greater than or equal to 3 out of 8 is indicative of an increased risk for OSA. The incidence airway maneuvers such as a chin lift, placement of nasopharyngeal airway, bag mask ventilation, and endotracheal intubation were tallied between the SB-positive and SB-negative subjects. Other significant adverse events such as hypoxemia, hypotension apnea, and early procedure termination due to a cardiovascular event were tallied. SB-positive subjects were more likely to exhibit hypoxemia and require an airway intervention when compared with SB-negative subjects. This difference remained significant after adjusting for an ASA physiologic class 3 or higher. This study shows that a risk stratification for OSA can be helpful in identifying patients who may require airway maneuvers during anesthesiologist-directed propofol sedation for advanced endoscopic procedures. Further data are needed regarding gastroenterologist-directed sedation for procedures not requiring deep sedation, as well as those requiring general anesthesia.

Are there any alternative forms of procedural sedation that may benefit the obese patient? Dexmedetomidine is a highly selective α ₂ adrenal receptor antagonist that possesses hypnotic, sedative, anxiolytic, and analgesic properties. A unique aspect of this medication is that it does not produce significant respiratory depression and hence may be of advantage in obese subjects undergoing endoscopic procedures. Disadvantages of this medication include the fact that a loading dose coupled with an infusion is necessary. Moreover, it is not released for use by nonanesthesiologists. The use of this agent during operative procedures has been shown to be of benefit. A case report described the utility of this medication in a morbidly obese patient undergoing a Roux-en- y gastric bypass. Eighty ASA 2 to 3 patients underwent laparoscopic bariatric surgery with the use of dexmedetomidine. In this randomized double-blind placebo-controlled dose ranging study, dexmedetomidine was found to lead to a significantly faster emergence from anesthesia with no significant difference in the intraoperative hemodynamic parameters.

In a prospective randomized trial, patients undergoing elective outpatient colonoscopy were randomized to 4 sedation regimens including dexmedetomidine, meperidine with midazolam, and fentanyl on demand. Primary outcomes included cardiorespiratory parameters, as well as recovery time. The study was terminated early due to the side effects stemming from dexmedetomidine. A randomized trial compared the safety and efficacy of dexmedetomidine versus midazolam for 50 ASA 2 to 3 adults undergoing elective upper endoscopy. There were no difference between the 2 sedation arms with respect to heart rate and mean arterial pressure. Patient satisfaction was significantly higher in the dexmedetomidine group. A small randomized trial compared the combination of propofol to dexmedetomidine for elective ERCP. Dexmedetomidine was not was found not to be as effective as the combination of fentanyl and propofol for conscious sedation. As in previous endoscopic studies, it was also found to be limited due to hemodynamic instability.

Another alternative to sedated procedures would be transnasal small-caliber endoscopes. Alami and colleagues used this method in 25 morbidly obese subjects undergoing a preoperative upper endoscopy. The average BMI was 47 (range 38–69 kg/m ² ). Sixty-eight percent of the patients had a history of OSA. No sedation was required for 23 of the patients. Pathology was identified in 14 of the 25 patients including Barrett esophagus, gastric polyps, esophageal varices, and gastric ulcers. Biopsies were successful in 12 out of 12 patients. No data were presented regarding hemodynamic parameters or patient tolerance.