Suspected remimazolam formulation-associated peri-induction anaphylaxis: two cases highlighting diagnostic complexity - Two case reports -
Article information
Abstract
Background
Reports of remimazolam formulation-associated anaphylaxis are increasing as its use expands. Between 2023 and 2025, we identified two severe reactions among 1,592 patients who received remimazolam at our institution.
Case
A 65-year-old man with end-stage renal disease developed profound hypotension without cutaneous signs after induction with 10 mg of remimazolam and peri-induction cefazedone administration. Surgery was cancelled due to persistent hypotension. Serum tryptase at 6 h was 38.0 μg/L; however, delayed skin testing was negative. Subsequent propofol-based anesthesia without cefazedone re-administration was uneventful. A 50-year-old woman developed hypotension and facial flushing within 2 min of receiving 10 mg of remimazolam. Although serum tryptase levels remained within the normal range, delayed skin testing for remimazolam was positive.
Conclusions
Diagnosis of remimazolam formulation-associated hypersensitivity may remain inconclusive when based on a single diagnostic test. Accurate evaluation requires integrating timed tryptase, delayed skin testing, and exposure history.
Remimazolam is an ultra-short-acting benzodiazepine sedative-anesthetic that was first approved in Japan for general anesthesia in 2020 [1] and has since been used for both procedural sedation and general anesthesia [2,3]. Remimazolam is characterized by rapid onset and recovery, a relatively favorable hemodynamic profile compared with propofol in anesthesia studies, and reversibility with flumazenil [2-4].
Although remimazolam is generally considered to have a favorable safety profile, severe hypersensitivity reactions, including anaphylaxis, have been reported following its administration [5,6]. Recent case reports and reviews suggest that remimazolam formulation-associated anaphylaxis may exhibit heterogeneous clinical and laboratory findings, which can complicate diagnostic evaluation [5,6].
From January 2023 to December 2025, remimazolam was administered to 1,592 patients at our institution, among whom two severe reactions were identified. We report these two cases of severe peri-induction anaphylaxis associated with the remimazolam formulation, highlighting the diagnostic complexity of suspected remimazolam formulation-associated hypersensitivity.
CASE REPORT
This case report was prepared in accordance with the CAse REports (CARE) guidelines. This study was approved by the Institutional Review Board of Inje University Haeundae Paik Hospital (No. HPIRB 2025-11-010), and the requirement for informed consent was waived due to the retrospective nature of the analysis.
Case 1
A 65-year-old man (160.5 cm, 59.3 kg) was scheduled for umbilical hernia repair. His medical history included diabetes mellitus, hypertension, and end-stage renal disease on peritoneal dialysis. He had undergone general anesthesia twice within the past 2 years without peri-anesthetic complications. Previous anesthetic agents included glycopyrrolate, lidocaine, remifentanil, fentanyl, propofol, rocuronium, desflurane, and sugammadex. Cefazedone had been administered during another surgical procedure 1 year before the index event without any adverse reaction. He had no known drug allergies.
Preoperative chest radiography showed bilateral pleural effusions. Transthoracic echocardiography performed 3 months earlier revealed a left ventricular ejection fraction of 54% with regional wall motion abnormalities in the left circumflex artery territory. Preoperative laboratory tests showed anemia (hemoglobin: 9.7 g/dl) and severe renal dysfunction (blood urea nitrogen/creatinine: 42.2/12.58 mg/dl). The patient was classified as American Society of Anesthesiologists physical status III.
Total intravenous anesthesia with remimazolam besylate (Byfavo®, Hana Pharm Co., Ltd.) and remifentanil was planned. Standard monitoring was applied. Baseline vital signs were blood pressure 143/64 mmHg, heart rate 53 beats/min, and oxygen saturation 95%. Glycopyrrolate 0.2 mg was administered intravenously, followed by preoxygenation. Continuous infusions of remifentanil (0.14 μg/kg/min) and remimazolam (6 mg/kg/h) were initiated for induction. After a cumulative remimazolam dose of 10 mg, the patient lost consciousness. Rocuronium 30 mg was administered intravenously, followed by endotracheal intubation. Concurrently, prophylactic cefazedone 2 g diluted in 100 ml normal saline was initiated via intravenous infusion.
Immediately after intubation, the patient developed profound hypotension (37/21 mmHg) with a heart rate of 55 beats/min, while oxygen saturation remained at 98%. No prominent cutaneous manifestations were observed, and there was no evidence of respiratory manifestations, such as bronchospasm or wheezing. End-tidal CO2 decreased to approximately 27 mmHg. A call for assistance was requested. All anesthetic agents and antibiotics being administered were discontinued, and resuscitation with vasopressors and intravenous fluids was initiated. A radial arterial catheter was placed for invasive monitoring. A 12-lead electrocardiogram revealed no significant abnormalities. Arterial blood gas analysis, obtained approximately 20 min after symptom onset, showed the following results: pH 7.145, PaCO2 67.7 mmHg, PaO2 106 mmHg, base excess −6.5 mmol/L, and lactate 2.7 mmol/L.
Because cutaneous and respiratory symptoms were not prominent and the patient had underlying cardiac abnormalities with reduced cardiopulmonary reserve, anaphylaxis was not initially suspected. Hemodynamic resuscitation was continued with continuous infusions of epinephrine, norepinephrine, and dobutamine, in addition to 1,500 ml of intravenous fluids. However, mean arterial pressure could not be maintained above 50 mmHg, and the surgery was cancelled. Endotracheal extubation was performed after partial improvement was confirmed on repeat arterial blood gas analysis (at approximately 50 min after symptom onset: pH 7.255, PaCO2 45.2 mmHg, PaO2 496 mmHg, base excess −6.9 mmol/L, and lactate 3.1 mmol/L) prior to transfer. The patient was admitted to the intensive care unit (ICU) for close monitoring.
Six hours after the event, serum tryptase was measured, and dexamethasone 5 mg was administered intravenously. Approximately 9 h after the event, blood pressure normalized without vasopressor support. The patient was closely monitored in the ICU for 24 h. Serum tryptase, reported 4 days later, was elevated at 38.0 μg/L (institutional reference range < 11.0 μg/L). On repeat testing 6 days later, the level had decreased to 13.8 μg/L. The patient was discharged 6 days after the event without sequelae and referred to the allergy specialist for further evaluation. At 4 weeks post-event, skin prick and intradermal testing for cefazedone, glycopyrrolate, remifentanil, rocuronium, remimazolam, and sugammadex were all negative.
At the rescheduled surgery 1 month later, remimazolam was replaced with propofol, and cefazedone was not re-administered. The patient was receiving ceftazidime and ciprofloxacin for peritonitis at that time; therefore, no additional prophylactic antibiotic was given. Rocuronium was safely re-administered as part of an otherwise similar anesthetic regimen without any adverse reaction, supporting its low likelihood as the causative agent despite the inherent limitations of skin testing. Although cefazedone had been tolerated during a prior anesthetic exposure and yielded negative skin test results, it could not be definitively excluded, as it had been administered at the time of the index event and was not re-administered during the rescheduled surgery. Based on the temporal relationship and prior uneventful anesthetic exposures, the remimazolam formulation was considered the most clinically plausible causative agent.
Case 2
A 50-year-old woman (161.0 cm, 64.0 kg) was scheduled to undergo a partial mastectomy for ductal carcinoma in situ. She had no significant comorbidities. Six months earlier, she had undergone general anesthesia at another hospital without peri-anesthetic complications. Her medical history was notable for allergies to peach and pollen. Preoperative laboratory tests, chest radiography, and electrocardiography were unremarkable. She was classified as American Society of Anesthesiologists physical status I.
Total intravenous anesthesia using remimazolam besylate (Byfavo®, Hana Pharm Co., Ltd.) and remifentanil was planned. Baseline vital signs were blood pressure 128/73 mmHg, heart rate 83 beats/min, and oxygen saturation 97%. Following preoxygenation with 100% oxygen, glycopyrrolate 0.2 mg was administered intravenously. Continuous infusions of remifentanil (0.13 μg/kg/min) and remimazolam (6 mg/kg/h) were then initiated. Within 1–2 min of starting remimazolam, after a cumulative dose of 10 mg, the patient lost consciousness and developed facial flushing. An immediate hypersensitivity reaction to the remimazolam formulation was suspected, and both infusions were discontinued. Additional assistance was requested. Rocuronium 50 mg was administered intravenously to facilitate endotracheal intubation, and the trachea was successfully intubated. Epinephrine 100 μg was administered intravenously, and rapid fluid resuscitation was initiated. Dexamethasone 5 mg and chlorpheniramine 4 mg were also administered intravenously as adjunctive therapy. Several minutes later, end-tidal CO2 decreased to 23 mmHg, whereas oxygen saturation remained at 100%. The lowest recorded noninvasive blood pressure was 45/28 mmHg, accompanied by tachycardia (113–118 beats/min). Anesthesia was converted to sevoflurane in 100% oxygen. No prominent respiratory symptoms were observed. Because severe hypotension persisted, a continuous epinephrine infusion was initiated.
An arterial catheter was placed for invasive hemodynamic monitoring. Arterial blood gas analysis and serum tryptase measurements were obtained 15 min after symptom onset. Arterial blood gas analysis showed pH 7.346, PaCO2 41.6 mmHg, PaO2 419 mmHg, HCO3− 22.7 mmol/L, base excess −3.0 mmol/L, and lactate 1.3 mmol/L. Thirty minutes after symptom onset, hydrocortisone 100 mg was administered intravenously. Surgery commenced 30 min after symptom onset under ongoing vasopressor support. During the procedure, arterial blood pressure was maintained between 95–132 mmHg systolic and 50–70 mmHg diastolic. A total of 2,000 ml of intravenous fluid was administered intraoperatively. A repeat serum tryptase measurement was obtained 90 min after symptom onset. At the end of surgery, the trachea was extubated, and the patient was transferred to the post-anesthesia care unit while receiving vasopressor support. She remained hemodynamically stable and was subsequently transferred to a monitored intermediate care unit, where she underwent continuous monitoring and close observation for 48 h. Vasopressor support was discontinued 22 h after the event. A convalescent baseline serum tryptase level was not obtained. The patient was discharged 7 days later without sequelae.
Serum tryptase levels measured 15 min and 90 min after symptom onset, and reported 1 day later, were 5.3 μg/L and 5.1 μg/L, respectively (institutional reference range < 11.0 μg/L). One month later, skin testing was performed for all administered agents, including remifentanil, glycopyrrolate, remimazolam, and rocuronium. Skin test results for remifentanil, glycopyrrolate, and rocuronium were negative. In contrast, remimazolam produced positive results: the skin prick test at a concentration of 1:1 (1 mg/ml) was weakly positive, and the intradermal test at 1:10 dilution (0.1 mg/ml) was positive. Based on the clinical course and delayed skin test positivity, remimazolam formulation-associated perioperative anaphylaxis was considered highly likely, although sensitization to dextran 40 could not be definitively excluded.
DISCUSSION
Both patients developed severe cardiovascular instability within minutes of exposure to the remimazolam formulation. However, post-event evaluation, including serum tryptase measurement and delayed skin testing, yielded contrasting findings. Case 1 presented predominantly with profound hypotension and a marked elevation in serum tryptase, whereas delayed skin testing was negative. In contrast, Case 2 exhibited early facial flushing followed by severe hypotension, persistently normal serum tryptase levels, and positive delayed skin test results to remimazolam formulation. Taken together, these cases demonstrate that remimazolam formulation-associated peri-induction anaphylaxis may present with heterogeneous clinical and laboratory findings, as summarized in Table 1. They further illustrate that standard post-event evaluation may remain inconclusive despite severe peri-induction reactions. Consequently, interpretation requires integration of the peri-induction chronology, exposure history, and complementary test results. These observations support a multifaceted clinical approach to suspected remimazolam formulation-associated hypersensitivity, consistent with contemporary recommendations for the evaluation of perioperative allergic reactions and anaphylaxis [7-9]. A key message of these cases is that a normal acute tryptase level does not exclude clinically significant anaphylaxis. Acute serum tryptase reflects the magnitude and timing of systemic mast cell activation and may remain within the reference range despite clinically significant anaphylaxis [7,10]. Contemporary recommendations emphasize that serum tryptase should be interpreted within the clinical context and, whenever possible, compared with a baseline value obtained after full recovery [7,11]. Within this framework, an event-related increase exceeding 1.2 × baseline + 2 μg/L is generally considered supportive of mast cell activation [7]. In Case 2, however, a baseline tryptase value was unavailable, precluding formal application of this criterion. This represents an important limitation, as a tryptase concentration within the laboratory reference range may nevertheless reflect a clinically meaningful rise from baseline. Delayed skin testing also has important limitations. Negative results do not fully exclude perioperative hypersensitivity, particularly for drugs or agents with poorly established immunoglobulin E (IgE)-mediated mechanisms and for which skin testing has not been validated [10]. Positive results, in turn, must be interpreted in the context of test timing, methodology, non–IgE-mediated mechanisms, alternative effector-cell activation pathways, and the possibility of excipient-related reactions [8,11,12]. Accordingly, these contrasting findings observed in these cases are mechanistically plausible and should not be regarded as mutually exclusive.
Clinical Comparison of Two Cases of Suspected Remimazolam Formulation-Associated Peri-induction Anaphylaxis
Attribution of the culprit agent in suspected remimazolam reactions is further complicated by formulation components. The commercially available remimazolam besylate preparation contains dextran 40, which has been implicated as a potential contributor to hypersensitivity reactions. Immediate dextran-associated reactions have historically been linked to IgG-mediated immune-complex formation and complement activation rather than a purely classical IgE-mediated mechanism, and this possibility has been specifically highlighted in the remimazolam anaphylaxis literature [13-15]. Because additional testing for dextran was not feasible in our patients, we could not determine whether the reactions were attributable to remimazolam itself, dextran 40, or the formulation as a whole. Therefore, our findings are most appropriately interpreted as evidence of remimazolam formulation-associated hypersensitivity rather than definitive proof of sensitization to the active drug alone. This distinction is not merely semantic; it has practical implications for allergy evaluation, interpretation of skin test results, and planning of future anesthetic management.
These cases also offer several practical lessons. First, abrupt hypotension immediately after remimazolam administration should prompt consideration of perioperative anaphylaxis even when cutaneous manifestations, bronchospasm, or airway edema are absent or subtle [7,11]. Second, with regard to acute management, current perioperative anaphylaxis recommendations emphasize early epinephrine administration and rapid intravenous crystalloid resuscitation as the priority interventions [9,16]. If profound cardiovascular collapse persists despite initial epinephrine administration and intravenous fluid resuscitation, cardiopulmonary resuscitation should be initiated without delay, particularly when systolic blood pressure remains below 50 mmHg [16]. Antihistamines and corticosteroids should not be considered first-line treatments for life-threatening anaphylaxis and must not delay epinephrine administration or fluid resuscitation. When used, these agents should be regarded as adjunctive therapies after initial stabilization [16]. In both cases, epinephrine-based hemodynamic support and intravenous fluid resuscitation were initiated promptly after severe hemodynamic instability was recognized. Accordingly, dexamethasone and chlorpheniramine, which were administered as adjunctive medications in Case 2, should not be interpreted as first-line emergency therapies. Third, post-anaphylaxis monitoring also warrants careful consideration. Biphasic or protracted reactions may occur after apparent recovery, with reported estimates of biphasic reactions ranging from 4.6% to as high as 20%. Consequently, close observation is required after anaphylaxis [17]. However, the appropriate duration and intensity of monitoring should be individualized according to severity, clinical course, and response to treatment, as recommendations vary [17]. In the perioperative setting, severe reactions often necessitate ICU admission, and monitoring in an appropriate environment for at least 24 h has been recommended [9]. In the present cases, Case 1 was closely monitored in the ICU and Case 2 in a monitored intermediate-care unit for more than 24 h, given refractory hypotension requiring vasoactive support in Case 1 and the continued need for vasopressor support until 22 h after the event in Case 2. No biphasic reaction occurred in either patient. Fourth, the post-event evaluation should be structured and multidisciplinary, incorporating acute and baseline serum tryptase measurements, formal allergy referral, and appropriately timed skin testing when feasible [7,8,11]. The management of cases where skin tests fail to identify a culprit agent must be addressed. Because several drugs are commonly administered in rapid succession during induction, definitive culprit identification can be challenging. Although drug provocation testing (DPT) may provide confirmatory evidence in the evaluation of drug hypersensitivity, its application to perioperative medications is limited by the potent pharmacological effects of anesthetic agents and the risk of re-inducing severe reactions [8]. Furthermore, in Case 1, the clinical risk of re-inducing anaphylaxis through DPT was considered unacceptably high because of the patient’s underlying comorbidities. Accordingly, planning subsequent anesthetic management often requires a pragmatic risk-reduction approach rather than definitive confirmation of the causative agent.
In Case 1, remimazolam was the only newly introduced agent compared with the patient’s previous uneventful anesthetic exposures. Cefazedone had been tolerated during a prior surgery, and delayed skin testing was negative, making it a less likely culprit. However, because cefazedone was being administered at the time of the index reaction and was not re-administered during the rescheduled surgery, its contribution could not be definitively excluded. Accordingly, the uneventful subsequent anesthetic course after substitution of remimazolam with propofol should therefore be interpreted as supporting a pragmatic risk-minimization strategy rather than as definitive evidence that remimazolam was the sole culprit. More broadly, this case suggests that careful comparison with previous uneventful anesthetic exposures may help guide targeted modifications to subsequent anesthetic plans, while recognizing the inherent uncertainty when multiple drugs are administered in rapid succession.
During the 3-year study period, remimazolam was administered to 1,592 patients at our institution, and two severe hypersensitivity reactions were identified (corresponding to a crude frequency of 0.13%). Although this frequency is numerically higher than the incidence estimates generally cited for perioperative anaphylaxis in large epidemiologic studies and exceeds that reported in recent prospective data from Japan, it should be interpreted with considerable caution [18,19]. This observation does not constitute a formal incidence estimate, as it is based on only two events from a single center and is therefore subject to substantial imprecision, case-ascertainment bias, local practice patterns, and denominator effects. Nonetheless, as remimazolam use expands in routine anesthesia practice, even a small number of severe hypersensitivity reactions underscores the need for continued vigilance and systematic post-event evaluation.
Notes
FUNDING
None.
CONFLICTS OF INTEREST
No potential conflict of interest relevant to this article was reported.
DATA AVAILABILITY STATEMENT
Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.
AUTHOR CONTRIBUTIONS
Conceptualization: Hyun-Seong Lee, Daeseok Oh. Data curation: Hyun-Seong Lee, Daeseok Oh. Formal analysis: Hyun-Seong Lee, Daeseok Oh. Methodology: Hyun-Seong Lee, Daeseok Oh. Project administration: Hyun-Seong Lee, Daeseok Oh. Visualization: Hyun-Seong Lee, Daeseok Oh. Writing - original draft: Hyun-Seong Lee, Daeseok Oh. Writing - review & editing: Hyun-Seong Lee, Min Woo Lee, Daeseok Oh. Investigation: Hyun-Seong Lee, Daeseok Oh. Resources: Hyun-Seong Lee, Min Woo Lee, Daeseok Oh. Software: Hyun-Seong Lee, Daeseok Oh. Supervision: Hyun-Seong Lee, Daeseok Oh. Validation: Hyun-Seong Lee, Daeseok Oh.