Completion of upper endoscopic procedures despite paradoxical reaction to midazolam: a role for flumazenil?

THE AMERICAN JOURNAL OF GASTROENTEROLOGY © 2000 by Am. Coll. of Gastroenterology Published by Elsevier Science Inc.

Vol. 95, No. 3, 2000 ISSN 0002-9270/00/$20.00 PII S0002-9270(99)00925-9

Completion of Upper Endoscopic Procedures Despite Paradoxical Reaction to Midazolam: A Role for Flumazenil? Scott A. Fulton, M.D., and Kevin D. Mullen, M.B., F.R.C.P.I Department of Internal Medicine, Division of Gastroenterology, Case Western Reserve University at Metrohealth Medical Center, Cleveland, Ohio

ABSTRACT Paradoxical excitation after benzodiazepine administration is well described. Although it is relatively uncommon, its occurrence can severely impede or even prevent the performance of upper endoscopy. We describe three cases in which paradoxical reactions to midazolam responded so well to flumazenil administration that the procedure was successfully completed in each instance. We review the limited literature on this topic and suggest that flumazenil may have greater utility in the management of this particular problem than is considered at present. (Am J Gastroenterol 2000;95:809 – 811. © 2000 by Am. Coll. of Gastroenterology)

INTRODUCTION Benzodiazepines, particularly midazolam, are the drugs of choice for conscious sedation in endoscopy (1). Rarely, with increasing dosage, patients become restless, agitated, and at times violent, making endoscopic evaluation more difficult, if not impossible. These “paradoxical reactions” are well described in the literature (2, 3). In the past, these reactions were managed primarily by increasing the dosage of the medication (worsening the response, at times) in an attempt to complete the procedure on an uncooperative patient. Frequently the procedure had to be abandoned altogether. In addition, prolonged recovery time from the administered medication was commonly observed. Other modes of management of the problem have included increased narcotic dosing, as well as physostigmine (4, 5), which have seen limited success. In alcoholic patients, the use of droperidol in conjunction with other sedatives has apparently reduced the incidence of this reaction (6). In recent years, flumazenil has been the drug of choice for reversal in benzodiazepine overdose, but data on its use in reversal of paradoxical reactions is limited. We describe 3 cases in which flumazenil was given with good response for paradoxical reactions, and briefly review the literature on this phenomenon.

CASE REPORTS Case 1 A 62-yr-old man with a long history of alcohol abuse presented with hematemesis and melena. After a 2-U blood transfusion, his vital signs were stable and Hct was 31. Blood alcohol was negative and the patient had no signs of alcohol withdrawal syndrome. After premedication with meperidine 12.5 mg and midazolam 5 mg (titrated slowly to sedation), the patient seemed ready for endoscopy. However, any attempts at intubation were met with wild arm swinging and head shaking. Undisturbed, the patient remained well sedated. Droperidol was given incrementally i.v. up to 5 mg, with no improvement in cooperation over the next 10 min. Suspecting a possible paradoxical reaction to midazolam, flumazenil 0.5 mg was given i.v. The patient remained sedated. Within 1–2 min he was easily intubated, and a smooth upper endoscopy was performed. A clean-based duodenal ulcer was seen. The patient had no recollection of the procedure and was discharged within 2 days. Blood pressure remained normal, and oxygen saturation ⬎94% throughout the entire episode, as well as the examination. Case 2 A 58-yr-old woman presented for outpatient upper endoscopy for evaluation of chronic upper abdominal pain. Long-term alcohol abuse was suspected but not proven. Premedication consisted of meperidine 12.5 mg i.v., droperidol 5.0 mg i.v., and midazolam 7.5 mg i.v. Despite this, the patient remained restless and somewhat agitated. Any attempts to communicate with the patient, or efforts to intubate, caused more agitation. Vital signs and blood oxygenation remained normal, as defined by pulse oximetry readings of ⬎ 94%. Flumazenil 0.5 mg i.v. was given, which immediately calmed the patient. Upper endoscopy was performed without any difficulty, and revealed severe distal esophagitis, and chronic gastritis. The patient took ⬎3 h to fully emerge from sedation and had total amnesia for the procedure.

810

Fulton and Mullen

Case 3 A previously healthy 26-yr-old man was admitted for hematemesis after initially having multiple vomiting episodes with clear gastric contents. The patient had no prior alcohol history but reported that a previous upper endoscopy attempted 2 yr ago for a similar episode was unsuccessful. His vital signs were stable and there were no other significant medical problems. Before endoscopy, premedication given included meperidine 25 mg i.v., midazolam 10 mg i.v. (slowly titrated), and droperidol 5.0 mg i.v. The patient appeared to be well sedated but kept spitting out his mouthpiece. Attempts at replacing this were unsuccessful, and the patient began to violently shake his arms and legs at the slightest touch. Vital signs remained normal, and oximetry readings ⬎ 94%. General anesthesia was briefly contemplated but, after flumazenil 0.5 mg i.v. was given, the patient allowed insertion of the mouthpiece while sedated. Upper endoscopy was immediately accomplished without difficulty and revealed a nonbleeding Mallory-Weiss tear. The patient had no recollection of the procedure.

DISCUSSION Benzodiazepines are the agents used most frequently for conscious sedation during endoscopic procedures. These drugs have significant advantages over other sedative and anxiolytic drugs in that they produce these effects without causing major respiratory depression. They have no narcotic actions, and patients remain arousable even after high doses. In low doses, anxiolytic and anticonvulsive action is found; at higher doses, muscle relaxation and amnesic properties predominate. Their mode of action is that of an agonist to the central benzodiazepine receptor (7). Midazolam has become a popular sedative for outpatient procedures because of its rapid onset of action and short half-life. Although many of the properties are favorable, a number of adverse reactions such as oversedation and paradoxical excitation can occur, which complicate procedures significantly. Paradoxical reactions are essentially responses that are contrary to the desired effect of the medication. With regard to benzodiazepines, these phenomena have been reported as early as 1960 (2), with descriptions ranging from increased talkativeness, to depression, disinhibition, excitement (sexual and otherwise), hostility, and rage (2, 3, 9, 10 –13). In a review of outpatient dental practices, the overall incidence of adverse reactions to benzodiazepines was reported as 29% (3). The specific incidence of pronounced adverse psychological reactions ranges from 1% to 9% (8). In some instances patients can become dangerous to themselves and to others in the room, making it necessary to abort the procedures and attempt to reverse the medication. It has been suggested that the personality of the individual, genetic factors, female gender, degree of patient apprehensiveness, and chronic alcoholism may be predictive of a paradoxical/ adverse reaction (2, 3, 5, 10). The mechanism of these

AJG – Vol. 95, No. 3, 2000

reactions is unknown, but has been ascribed to a disinhibition or release of “anxiety-bound” hostility. In both children and adults, there is the suggestion that this is dose-dependent, but this is by no means universal. A significant proportion of so-called paradoxical reactions may, in fact, represent an “appropriate” response to oversedation. As described by McCloy and Pearson, “oversedation can lead to isolation of the patient from verbal and outside stimuli, inhibiting cooperation” (14). Attempts at insertion of the endoscope into the mouth and pharynx are perceived as threatening, potentially obstructing the airway; and, uncontrollably, the patient attempts to remove the threat. This may explain why paradoxical reactions are more frequently observed with upper endoscopies, higher dosages of medications, and darkened endoscopy areas. Imidazodiazepines, a new class of benzodiazepines that were developed in 1981 demonstrated the ability to antagonize central actions of other benzodiazepines (15). Of these, only flumazenil has received approval by the United States Food and Drug Administration because of its efficacy and favorable side effect profile in safely reversing sedative effects (7, 15). Since its release, flumazenil has been employed primarily as a reversal agent in patients with hemodynamic or respiratory compromise after premedication. Interestingly, the use of flumazenil is not well described for paradoxical reactions. To date, there are four case reports in the English literature describing the use of flumazenil in this setting. Table 1 is a breakdown of these cases along with the current reported patients. In all cases, even though the half-life of flumazenil is approximately 50 min, repeated dosages were unnecessary, and the patients had no recollection of the previous events (7). Of note, in only one of the cases did the patient receive flumazenil during the procedure, allowing comfortable completion (10). In the others, either the procedure was completed with an uncooperative patient, or it was terminated before administration of the drug. Regarding retention of amnesia after administration of flumazenil, one author commented that this suggested that paradoxical reactions might be mediated at a site that is different from the site for sedative or amnesic responses (11). One of the most interesting aspects of the effect of flumazenil on our patients was the lack of complete reversal of sedation with flumazenil. Whether this occurred because of the other agents given as part of the premedication regimen is unknown, as none of our patients received midazolam alone. However, this offers one explanation for the beneficial effect of flumazenil in our patients. The distinguishing feature of the cases presented in this article is that, not only was the aggressive behavior reversed, but the patients remained sedated and were able to cooperate fully with the procedure, as reported by Rodrigo (10). This phenomenon has not, to our knowledge, been previously reported in relation to upper endoscopy. This observation is important in the setting of endoscopy, when issues such as preparation for procedures, loss of time from work, staff costs, etc. surface

AJG – March, 2000

Flumazenil for Paradoxical Midazolam Reaction

811

Table 1. Reports of Reversal of Paradoxical Reactions to Midazolam With Flumazenil Midazolam Dose Procedure/ Setting

Flumazenil Dose/ When Administered

49/female

12 mg i.v./70 min oral surgery

0.5 mg/during procedure

27/male 71/male

7 mg i.v. cystoscopy 4 mg i.v. EGD

0.3 mg/post procedure 0.2 mg/during procedure

11/male 62/male

17.5 mg p.o. i.v. placement 5 mg i.v. EGD

58/female

7.5 mg i.v. EGD

26/male

10 mg i.v. EGD

0.15 mg/post procedure 0.5 mg i.v./during procedure 0.5 mg i.v./during procedure 0.5 mg i.v./during procedure

Age (yr)/Sex

because of the need to reschedule aborted procedures. More important, when the nature of the procedure is emergent, methods to improve cooperation are essential. The primary message and reasons for presenting these cases are to offer other possible alternatives for managing this difficult situation. Regardless of how one defines this reaction, the use of flumazenil may be helpful in selected patients provided that 1) inadequate sedation and hypoxemia have been ruled out as causes for poor cooperation, and 2) it has been determined that the patient does not habitually use benzodiazepines. It is likely that, with an improved understanding of the mechanisms involved in the paradoxical response, better methods for patient management in this setting can be developed. Reprint requests and correspondence: Kevin D. Mullen, M.B., F.R.C.P.I., Division of Gastroenterology, 6th Floor Bellgreve, Metrohealth Medical Center, 2500 Metrohealth Drive, Cleveland, OH 44109-1998. Received Feb. 6, 1998; accepted Dec. 17, 1998.

REFERENCES 1. Bell GD. Review article: Premedication and intravenous sedation for upper gastrointestinal endoscopy. Aliment Pharmacol Therap 1990;4:103–22. 2. Tobin JM, Lewis N. New psychotherapeutic agent chlordiazepoxide. JAMA 1960;174:1242–9. 3. Litchfield NB. Complications of intravenous diazepam. Adverse psychological reactions. Anesth Prog 1980;27:175– 83.

Outcome

Reference

calmed patient, completed procedure, retained amnesia calmed patient, retained amnesia calmed patient, retained amnesia, procedure aborted calmed patient, retained amnesia calmed patient, procedure completed, retained amnesia calmed patient, procedure completed, retained amnesia calmed patient, procedure completed, retained amnesia

10 11 12 13 current series current series current series

4. Caldwell CB, Gross JB. Physostigmine reversal of midazolam-induced sedation. Anesthesiology 1982;57:125–7. 5. Garber JG, Ominski AJ, Orkin FK, et al. Physostigmineatropine solution fails to reverse diazepam sedation. Anesth Analg 1980;59:58 – 60. 6. Wilcox CM, Forsmark CE, Cello JP. Utility of droperidol for conscious sedation in gastrointestinal endoscopic procedures. Gastrointest Endosc 1990;36:112–5. 7. Amrein R, Hetzel W, Bonetti EP, et al. Clinical pharmacology of Dormicum (midazolam), and Anexate (flumazenil). Resuscitation 1988;16(suppl):S5–27. 8. Physician’s desk reference. Montvale, NJ: Medical Economics, 1997:2324 –7. 9. Greenblatt DJ, Shader RI. Benzodiazepines (first of two parts). N Engl J Med 1974 291:1011–5. 10. Rodrigo CR. Flumazenil reverses paradoxical reaction with midazolam. Anesth Prog 1991;38:65– 8. 11. Thurston TA, Williams CG, Foshee SL. Reversal of a paradoxical reaction to midazolam with flumazenil. Anesth Analg 1996;83:192. 12. Honan, VJ. Paradoxical reaction to midazolam and control with flumazenil. Gastrointest Endosc 1994;40:86 – 8. 13. Thakker P, Gallagher TM. Flumazenil reverses paradoxical reaction to midazolam in a child. Anaesth Intens Care 1996; 24:505–7. 14. McCloy RF, Pearson RC. Which agent and how to deliver it? A review of benzodiazepine sedation and its reversal in endoscopy. Scand J Gastroenterol 1990;25(suppl 179):7– 11. 15. Ricou B, Forster A, Bruckner A, et al. Clinical evaluation of a specific benzodiazepine antagonist (RO 15–1788). Studies in elderly patients after regional anesthesia under benzodiazepine sedation. Br J Anaesth 1986;58:1005–11.