A 66-year-old man attended for routine day-case arthroscopic subacromial decompression on an afternoon operating list. He had no significant past medical history and was essentially fit and well. He had undergone two uneventful general anaesthetics for shoulder arthroscopies in the past.
Following induction of anaesthesia, he developed severe hypotension and tachycardia with a SBP of 55 to 60 mmHg. This hypotension was unresponsive to large intravenous fluid boluses and intermittent boluses of metaraminol. He required repeated adrenaline boluses and a noradrenaline infusion. His ECG at the time suggested nonspecific anterolateral ST-segment depression. Surgery was abandoned and the patient was transferred to the high dependency unit for further investigations and resuscitation. His cardiovascular parameters normalised and ECG changes resolved, although the 12-h troponin was modestly elevated at 87 ng l−1. He was discharged home the following day and a cardiology referral was made for further investigations. A subsequent transthoracic echocardiogram was normal and a diagnostic coronary angiogram revealed entirely unobstructed coronary vessels. As a result, the troponin-positive cardiac episode was attributed to profound hypovolaemia and hypotension.
Six months later, the patient was listed for the same operation as an inpatient. He was preloaded with 1000 ml Hartmann's solution and general anaesthesia was induced using midazolam, fentanyl and propofol. He was then transferred on to the operating table and positioned in the left lateral position. Once surgery had commenced, he received intravenous dexamethasone, ondansetron and paracetamol. Ten minutes later, a drop in blood pressure was noted which quickly progressed to cardiac arrest with the ECG showing pulseless electrical activity. The operation was abandoned. He was successfully resuscitated following a brief period of cardiopulmonary resuscitation and was transferred to the intensive care unit.
With a working diagnosis of anaphylaxis, all pertinent investigations were sent after consultation with the immunology team. The immunology results showed significant elevations in serum tryptase levels at 0 h (67.6 mg l−1), 1 h (89.3 mg l−1) and 24 h (8.7 mg l−1) (normal range 2 to 14 mg l−1). Total immunoglobulin E concentration was mildly elevated at 162 kU l−1. Skin prick testing to neat ondansetron (2 mg ml−1 concentration) was negative; however, intradermal testing to ondansetron diluted 1 : 50 in 0.9% isotonic saline showed a 2.5-fold increase in wheal diameter. The result was also positive at 1 : 250 dilution with a two-fold increase in wheal diameter. The above tests to dexamethasone, propofol, midazolam, glycopyrronium, diamorphine, fentanyl, latex and betadine were all negative. Complement and mannose-binding lectin assays were normal.
The patient made a full recovery and was discharged to go home on the third postoperative day.
Six months later, he returned for the same procedure, which was performed using regional anaesthesia in the form of an interscalene block with propofol sedation. He was discharged on the same day. He was pleased with the outcome of the surgery and gave his written consent for the publication of this case report.
The incidence of anaphylaxis in anaesthesia has been estimated at between 1:10 000 and 1:20 000.1,2 It can be a very challenging condition to manage and is fatal in 10% of cases.
There were two serious events which ultimately led to the final diagnosis of anaphylaxis. A multidisciplinary meeting was initiated following the first hypotensive episode. The panel, which comprised several consultant anaesthetists, thought the cause of hypotension was cardiac in nature because of the nonspecific anterolateral ST-segment depression. The cardiology team confirmed that there was no underlying coronary disease or evidence of left ventricular dysfunction and attributed the hypotension to profound hypovolaemia. This was attributed to the prolonged preoperative fasting period (nil by mouth since 18:00 the evening prior to surgery) and the fact that the patient had gone for a morning jog on the day of surgery. A differential diagnosis of anaphylaxis was not considered at that time. The patient was reviewed on two other occasions by anaesthetic consultants. Both were satisfied with the cardiology investigations and did not feel the need to consider other potential causes for the previous hypotensive episode.
The first episode of hypotension observed in our patient was most likely because of his sensitivity to ondansetron, which he had safely tolerated on two previous occasions. Neat ondansetron skin prick test is usually negative in most cases,3,4 as it was in our case, and should not be used to exclude allergy. In our patient, the test was positive down to 1 : 250 dilution. A 1 : 50 dilution is the minimum necessary dilution to consider ondansetron allergy and was positive in none of the 10 healthy controls reported by Fernando and Broadfoot.3
Ondansetron, a selective 5-hydroxytryptamine receptor antagonist, is widely used as an antiemetic because of its efficacy and safety profile. Life-threatening reactions have been uncommonly reported. Fernando and Broadfoot3 reported a reaction that consisted of marked hypotension without skin rash or bronchospasm, as was the case in our patient. Unexplained hypotension during anaesthesia should always be investigated with serial tryptase estimations and ondansetron should be included in the list as a potential agent amongst other drugs if used. It is the responsibility of the anaesthetist to appropriately report, refer and follow-up these patients. A suspected ‘allergy under anaesthesia’ pack containing all relevant paperwork required to make referrals is now available in every theatre in our hospital.
Anaesthetists are probably the clinicians who are most likely to encounter anaphylaxis during their routine practice. As such, they should be wary of overlooking the diagnosis when the patient presents with atypical features, as this is the most common contributor to diagnostic errors.5 In a Scandinavian study,6 42 anaesthetists were tested on a high fidelity simulator and not one made the diagnosis of anaphylaxis within 10 min.
In conclusion, this case reinforces the following statement highlighted by the Association of Anaesthetists of Great Britain and Ireland7:
‘… during anaesthesia when a fall in blood pressure, change in heart rate or difficulty with ventilation are noticed, anaphylaxis is so rarely the cause that inevitably treatment is given for another diagnosis before anaphylaxis is recognised; the response to this treatment has commonly delayed or even prevented recognition of the true cause.’
Acknowledgements relating to this article
Assistance with the article: the authors would like to thank Dr Amolak Bansal, Consultant Immunologist, who helped conduct the immunological tests at St Helier's Hospital.
Financial support and sponsorship: none.
Conflicts of interest: none.
1. Fisher MM, Baldo BA. The incidence and clinical features of anaphylactic reactions during anaesthesia in Australia. Ann Fr Anesth Reanim
2. Laxenaire MC. Epidemiology of anesthetic anaphylactoid reactions. Fourth multicenter survey. Ann Fr Anesth Reanim
3. Fernando SL, Broadfoot AJ. Ondansetron anaphylaxis: a case report and protocol for skin testing. Br J Anaesth
4. Tan J, Mehr S. Anaphylaxis to an ondansetron wafer. J Paediatr Child Health
5. Mark L, Graber MD, Franklin N, Gordon R. Diagnostic errors in internal medicine. Arch Intern Med
6. Jocobsen J, Lindekær AL, Østergaard HT, et al. Management of anaphylactic shock evaluated using a full-scale anaesthesia simulator. Acta Anaesthesiol Scand
7. Association of Anaesthetists of Great Britain and Ireland. Suspected anaphylactic reactions associated with anaesthesia. Anaesthesia