Think Before You Administer: Is Routine Benzodiazepine Premedication Before Endoscopy in Adults Necessary?

See Article, p 741 Sedation and analgesia are commonly administered during gastrointestinal endoscopy to promote patient comfort and cooperation and to facilitate performing the procedure. Propofol has become the standard of care because it allows for rapid recovery without residual psychomotor effects as well as optimizes patient satisfaction and procedure acceptability. Usually, a low dose of midazolam and/or opioid (eg, fentanyl) are combined with propofol. Midazolam is typically administered before transferring the patient to the procedure room to reduce anxiety. In addition, midazolam may reduce propofol dose requirements and the risk of propofol-induced cardiorespiratory complications. Opioids not only provide analgesia but also reduce propofol dosing and patient movement.1 In this issue of Anesthesia & Analgesia, Gurunathan et al2 assessed the effects of adding midazolam (0.04 mg/kg, up to 5 mg) to a propofol and opioid sedation regimen on the quality of recovery after colonoscopy. The investigators used the Postoperative Quality of Recovery Scale (PostopQRS)—a validated multidimensional survey-based tool—to assess postoperative recovery in the domains of physiological, emotive, nociceptive, cognitive, and activities of daily living. The study found that there was no difference between the midazolam and placebo groups in the cognitive domain of PostopQRS on postprocedure day 3. Also, the overall recovery, including nociceptive, emotive, and activities of daily living, was similar with or without midazolam. With respect to immediate recovery, there were no differences in the time to eye opening and hospital discharge. Finally, patient and endoscopist satisfaction were similar between groups. Because there appeared to be neither benefits nor harm, the authors concluded that the anesthesiologist should determine the use of midazolam. So, what should an anesthesiologist do? The outcomes of the present study could readily support the view that midazolam should not be used at all for colonoscopy because no benefit was shown. Alternatively, midazolam use is justified as there was lack of demonstrated harm. A nuanced answer, then, requires one to address the following questions based on a critical evaluation of the literature: Does midazolam premedication in adults provide significant benefits such as reduced anxiety, amnesia, improved cardiorespiratory status, or improved patient satisfaction? Alternatively, does it increase the incidence of cardiorespiratory depression and airway obstruction either during or after short procedures? Does it hinder recovery and delay discharge home after short ambulatory procedures? Does low-dose midazolam cause any potential harm in specific at-risk populations such as the elderly and the obese with obstructive sleep apnea or hypoventilation syndrome? The primary outcome of the study by Gurunathan et al2 was to assess for cognitive impairment—a subject that has received increasing attention in recent years—because it is associated with increased morbidity and mortality.3 Although not observed in this study, benzodiazepine administration has been shown to increase emergence delirium and cognitive dysfunction.3 The lack of cognitive dysfunction observed in the present study may be due to its limitation of substantial study subject attrition for cognitive assessment, possibly because cognitively impaired patients were reluctant to undergo evaluation. The differences in the findings of this study and previous reports may also be due to the variability in tools used to assess postoperative cognitive dysfunction.4 The present study did not report differences between younger and older patients, which might have provided interesting insights on the differing effects of midazolam based on age, as reported in others studies. A recent study observed that even a 2 mg intravenous dose of midazolam in older adults is associated with measurable deficits by cognitive testing and functional imaging changes.5 Another study in colonoscopy patients reported that the administration of adjuvants (eg, fentanyl or midazolam) with propofol resulted in better operating conditions and shorter endoscopy time but also increased the time to achieve cognitive recovery.6 This was especially pronounced in patients who took preoperative psychotropic medications or received >2 mg midazolam, who tended to display more clinically significant cognitive decline.6 It should be noted that antidepressant use has been on the rise for decades; currently 1 in 8 adolescents and adults take them, but this increases to 1 in 5 adults over 60 years.7 Additionally, older patients are at a higher baseline risk of postoperative cognitive dysfunction which, when coupled with higher rates of use of psychotropic medications, may further increase the risk of postoperative cognitive dysfunction in this population. It is possible that midazolam may have minimal effects on cognitive function after a diagnostic endoscopic procedure compared with a therapeutic endoscopic procedure requiring general anesthesia or a more invasive surgical procedure. It is also important to note that subjects with prior cognitive decline or memory impairment were excluded from the present study by Gurunathan et al,2 and thus the effects of midazolam on postprocedure cognitive function in this patient population remain unknown. However, with the aging world population, the prevalence of patients with cognitive impairment presenting for diagnostic and therapeutic gastrointestinal endoscopies will increase. While in a rigorous study, even subtle cognitive impairment can be identified by experienced research personnel, it may be overlooked in a busy endoscopic practice. Given these unknowns and the lack of any clinical benefits, routine use of midazolam seems unjustified. The current study observed that both groups approached 100% patient satisfaction whether or not midazolam was administered. In contrast, a previous study found higher patient satisfaction for fentanyl/propofol sedation for colonoscopy if preprocedure midazolam was included.8 Another recent study noted patient preference for propofol sedation over midazolam/fentanyl sedation.9 Thus, it appears that propofol, rather than midazolam, is the main contributor to patient satisfaction. It is generally assumed that midazolam premedication would reduce preprocedure anxiety and, thus, improve patient satisfaction. However, several studies have concluded that benzodiazepine premedication may not necessarily reduce anxiety or improve patient experience.10,11 In the present study by Gurunathan et al,2 the midazolam group did require lower propofol doses, which may exert clinical benefits such as reduced occurrence of hypotension or airway obstruction, but these were not reported. Given this, we cannot draw any conclusions from this study on this end point. Nevertheless, combining propofol, midazolam, and fentanyl would have at least additive cardiorespiratory depressive effects and, thus, may increase the potential cardiorespiratory complications. Several studies have shown evidence of dose-dependent airway obstruction under propofol sedation,12–14 and a recent study comparing obstruction patterns with different sedatives showed higher levels of obstruction with midazolam compared to either propofol or dexmedetomidine.15 The evidence-based guidelines of the Society of Anesthesia and Sleep Medicine concluded that patients with obstructive sleep apnea may be at an increased risk for adverse respiratory events after propofol sedation.16 This risk is further exacerbated with depressant drugs with additive effects such as midazolam, opioids, ketamine, and dexmedetomidine. Therefore, it was recommended that benzodiazepine sedation should be used with caution.16 In the present study by Gurunathan et al,2 the midazolam group had delayed recovery in the level of consciousness and response to verbal command at 15 minutes, but this resolved after 1 hour. While a seemingly minor difference, a 15-minute delay in recovery for each patient may significantly reduce the efficiency of a busy endoscopy facility. Thus, the addition of preprocedure midazolam should be avoided if one anticipates rapid postprocedure discharge. Additionally, preoperative midazolam reliably causes anterograde amnesia and thus has been reported to impair patients’ recall of their interactions with their anesthesiologist.17 While this does not worsen clinical outcomes, such gaps in memory may mask from patients the integral role of anesthesiologists in their care and undermine this doctor–patient relationship. Although not assessed in this study by Gurunathan et al,2 the need for addition of opioids to propofol sedation for colonoscopy is also questionable. Addition of an opioid to propofol has been reported to improve conditions of esophageal instrumentation during upper endoscopy.18 In contrast to upper gastrointestinal endoscopy where insertion of endoscope may cause retching, the insertion of the colonoscope is usually well tolerated, and the stimulating portions of the procedure relate to gas insufflation and scope manipulation around colic flexures can be adequately managed with propofol alone. Because gas insufflation and abdominal compression during colonoscopy have been shown to increase the risk of aspiration,19 preservation of airway reflexes would be beneficial, which would be achieved by avoiding deep sedation that is more likely with combinations of midazolam, fentanyl, and propofol. In summary, given the modest advantages and significant potential adverse effects of a benzodiazepine, its routine use before endoscopy should be avoided. This recommendation is in line with the enhanced recovery protocols published by several prominent professional societies emphasizing avoidance of routine benzodiazepine premedication. Finally, future studies should more formally assess the differential effects of each component of the propofol–opioid–midazolam combinations in procedures of varying invasiveness and in subpopulations at higher risk for complications. DISCLOSURES Name: John C. Alexander, MD, MBA. Contribution: This author helped write and revise the manuscript. Conflicts of Interest: J. C. Alexander has received honoraria from Pacira Pharmaceuticals. Name: Girish P. Joshi, MBBS, MD, FFARCSI. Contribution: This author helped write and revise the manuscript. Conflicts of Interest: G. P. Joshi has received honoraria from Baxter Pharmaceuticals and Pacira Pharmaceuticals. This manuscript was handled by: Jean-Francois Pittet, MD.