Magnetic resonance imaging is a non-invasive test with a low complication rate even when using contrast agents. Nevertheless, even routine examinations can cause serious complications, as this case report illustrates.
A woman in her fifties was to undergo an outpatient magnetic resonance imaging (MRI) scan with contrast as part of the follow-up of a chronic neurological disorder. She had never been exposed to MRI contrast agents before, but had no known drug allergies.
Approximately one minute after the administration of a gadolinium-based contrast agent, gadoteric acid (Dotarem®), the patient suffered a generalised seizure and became unresponsive. She was immediately removed from the MRI machine and wheeled out of the room on a gurney.
A 'code blue' alert (cardiac arrest) was triggered and a specialty registrar in anaesthesiology arrived one minute later. The patient was by then lying in the recovery position with spontaneous agonal respiration. She was unconscious, flushed over her entire body, and had a weak radial pulse bilaterally. An anaphylactic reaction was immediately suspected, and she received 0.5 mg adrenaline intramuscularly followed by 0.05 mg adrenaline intravenously. Her first measured blood pressure was 144/90 mm Hg and she had sinus tachycardia of 110–130 beats per minute. She was transferred to the intensive care unit with ongoing assisted ventilation. She did not have bronchospasm.
She was intubated immediately upon arrival owing to a Glasgow Coma Scale (GCS) score of 3 and partial airway obstruction. Upon intubation, she received 50 mg esketamine (Ketanest), 0.1 mg fentanyl and 75 mg suxamethonium (Curacit), all intravenously. The placement of the tube was confirmed with end-tidal CO2 capnography, and the initial reading was 4.5 kPa, indicating adequate circulation. One litre of Ringer's acetate was infused. During further attempts to establish invasive arterial pressure monitoring, pulseless electrical activity (PEA) was detected. Ultrasound revealed no detectable brachial, radial or femoral pulse, despite a stable sinus rhythm of 70 beats per minute. Advanced cardiopulmonary resuscitation was started 27 minutes after the code blue alert.
Echocardiography during resuscitation revealed cardiac standstill, and 1 mg adrenaline was then administered intravenously in each loop. Pulse monitoring revealed persistent pulseless electrical activity. The LUCAS mechanical chest compression system was used for ten minutes before the patient regained a pulse-generating rhythm, as revealed by pulse pressure measured at the femoral artery. She received advanced cardiopulmonary resuscitation for 17 minutes in total. She remained hypotensive with systolic blood pressure below 80 mm Hg, and infusion of adrenaline 0.2 µg/kg/min was started. In addition, she received intravenous infusion of 6 000 ml of Ringer's acetate over the first two hours, followed by infusion at 300 ml/h.
During the first 24 hours, the patient received a total of 12 litres of intravenous fluids. Haemodynamic monitoring with a pulse-induced continuous cardiac output (PiCCO) catheter revealed significant fluid requirements, low cardiac output and high peripheral resistance. PiCCO measurements were used to guide further fluid therapy.
Following the return of spontaneous circulation, glucocorticoids (Solu-Cortef) 250 mg and dexchlorpheniramine 10 mg were administered intravenously. A central venous catheter and dialysis catheter were inserted. The patient began to show signs of responsiveness with the need for analgosedation, and therefore therapeutic hypothermia was not induced.
Dialysis for elimination of the contrast agent was performed for 1 hour and 20 minutes, but was hindered by cardiovascular instability and had to be discontinued owing to catheter dysfunction. Adrenaline infusion was replaced with dobutamine infusion 5 µg/kg/min to reduce peripheral resistance. The dobutamine infusion was phased out over 48 hours and the patient was extubated 46 hours after the code blue alert.
The patient had intact cerebral function, and respiratory and haemodynamic stability. Brain MRI without contrast seven days after the event showed no signs of ischaemic brain damage.
With the exception of anisocoria, the patient had no obvious sequelae. She was discharged home after 18 days.
The patient has consented to the publication of the article. The article has been peer-reviewed.