Anesth Analg 2008; 106:520-522
© 2008 International Anesthesia Research Society
doi: 10.1213/ane.0b013e3181606ae5
PATIENT SAFETY
Unintentional Intracerebroventricular Administration of Etomidate and Rocuronium
Stephen Howell, MD, and
Richard P. Driver, Jr, MD
From the Department of Anesthesiology, West Virginia University School of Medicine, 3618 Robert C. Byrd Health Science Center, West Virginia.
Address correspondence and reprint requests to Stephen Howell, MD, Department of Anesthesiology, West Virginia University School of Medicine, 3618 Robert C. Byrd Health Science Center, P.O. Box 9134, Morgantown, WV 26506-9134. Address e-mail to howells2{at}rcbhsc.wvu.edu.
Abstract
We report the unintentional intracerebroventricular administration of etomidate and rocuronium through a ventriculostomy catheter in the intensive care unit during a rapid sequence induction for endotracheal intubation. Rapid loss of consciousness, apnea, and vocal cord abduction occurred, creating excellent intubating conditions. The patient remained hemodynamically stable throughout induction and intubation. Inexperience of nursing personnel and systems errors were factors contributing to the drug error. The management, clinical course, predisposing factors, and strategies for prevention will be discussed.
Medication errors continue to place hospitalized patients at risk for preventable injury. This case report details a serious and preventable medication error, which, to our knowledge, has not been described. Systems-based errors undoubtedly increased the risk of this adverse event.
CASE REPORT
A 61-yr-old, 100 kg, man presented to the Emergency Department with headache, confusion, and a Glasgow comma scale (GCS) score of 14. A computed tomography (CT) scan demonstrated subarachnoid hemorrhage. Cerebral angiography revealed a ruptured right anterior communicating A1–A2 aneurysm with broad-based anatomy, unfavorable for endovascular coiling. The patient was admitted to the Intensive Care Unit and a right lateral ventriculostomy catheter was placed. On postbleed day 2, a right frontotemporal craniotomy was performed with uneventful microsurgical clipping of the aneurysm. The intraoperative course was unremarkable and the patient was tracheally extubated postoperatively, with a GCS score of 14. On postbleed day 3, the patient had a decline in mental status secondary to cerebral vasospasm and required endotracheal intubation for airway protection. Hypervolemia, hypertension, and hemodilution ("triple-H") therapy was maintained. The following day the patient self-extubated, remained stable, and was off mechanical ventilation for a period of 24 h. His GCS score at this time was 11.
Reintubation was eventually required due to increasing oxygen requirements and waxing and waning mental status. The patient was confused and combative, though intermittently following simple commands. The management plan was a rapid sequence induction. The patient was administered 100% oxygen by mask and nursing personnel acquired the requested induction drugs. A verbal order was given for etomidate (Amidate®) 20 mg (10 mL, 2 mg/mL) and rocuronium (Zemuron®) 100 mg (10 mL, 10 mg/mL) IV. The drugs were unintentionally injected into the ventriculostomy catheter, which was mistaken by nursing personnel for an IV line. The drugs were injected at a rate typical for rapid sequence induction, approximately 25–50 mL/min. Approximately 1 min after intracerebroventricular (ICV) administration of etomidate and rocuronium, the patient lost consciousness and became apneic. The nurse recognized the error and informed the anesthesia provider. Endotracheal intubation was immediately performed under direct laryngoscopy with excellent intubating conditions: grade 1 laryngoscopy, abducted vocal cords, and absence of coughing and movement. Hemodynamic stability was maintained with no significant change in heart rate or arterial blood pressure observed. Immediately after the intubation, 20 mL of cerebrospinal fluid was aspirated from the ventriculostomy catheter and the catheter was left open to drain. Over the course of 10 min, the cerebrospinal fluid became progressively more blood tinged. The neurosurgical team was contacted and further management was discussed. A repeat CT scan demonstrated some effacement of the right lateral ventricle, a new finding of uncertain significance.
The nature and severity of the medication error was discussed in detail with the patient’s family. After several hours, the patient began to briskly withdraw to pain, but could not follow simple commands, GCS score 8. Subsequent cerebral angiography revealed persistent vasospasm in the A2 and A3 distribution.
The patient’s neurological examination never improved and his hospital course was complicated by renal failure, atrial fibrillation, and toxic epidermal necrolysis (presumably due to phenytoin therapy for seizure prophylaxis). The patient continued on standard therapy for cerebral vasospasm until transcranial Doppler examinations were consistent with resolution. The patient went on to develop infarctions in the right temporoccipital lobe and bilateral periventricular white matter as demonstrated on CT scan. On post-bleed day 30, without hope of meaningful neurological recovery, his family decided to withdraw supportive care. An autopsy was not performed.
DISCUSSION
Medication administration errors continue to be a source of undue morbidity and mortality. Identifying factors that contribute to these errors is crucial to their prevention. Inexperience and a lack of vigilance on the part of the intensive care unit nursing personnel certainly contributed to this error. It is critical to trace the course of any line before injection of medication, and this simple step would have prevented this tragic error. However, a systems error identified was the similarity in appearance between the pressure transducer tubing of the ventriculostomy catheter and tubing attached to the central venous catheter. In addition, comparison of the two systems showed that the position of the stop-cocks and injection ports was nearly identical. The ventriculostomy catheter should be attached to tubing that has a unique appearance, either by color coding or labeling. Additionally, the ventriculostomy catheter should be devoid of injection ports that could facilitate ICV injection. The preventative strategy adopted by our institution was to place red "dead-end" caps on all ventriculostomy access ports. A superior solution may be possible at the manufacturing level, in that external ventricular drain systems could be redesigned to prevent the attachment of standard-sized syringes and stop-cocks.
Other medications have been inadvertently injected into ventriculostomy catheters.1,2 ICV-administered medications can exert central nervous system effects through a variety of mechanisms. Non–receptor-mediated mechanisms of action of ICV-injected substances include direct tissue damage and changes in cerebrospinal fluid pH, osmolality, and intracranial pressure. Drugs given by the ICV route may not readily diffuse into brain parenchyma.3 Diffusion into neuronal tissue from the cerebrospinal fluid is likely a function of molecular weight, lipid solubility, and time. In the rabbit model, 1 h after ICV administration of radiolabeled methotrexate and inulin, "40% of brain sections contained appreciable quantities of tracer," with the highest concentrations occurring in the "gray matter adjacent to the cerebrospinal fluid, including the hippocampus, thalamus, caudate nucleus, and periaquaductal gray."4 Thus, ICV-administered drugs may exert receptor-mediated effects at or near the brain– cerebrospinal fluid interface. Alternatively, drugs in the cerebrospinal fluid can be redistributed into the peripheral circulation with subsequent transport across the blood–brain barrier. Thus, an ICV infusion may, in some instances, mimic a slow IV infusion.3
The ICV injection of etomidate in 35% propylene glycol has been reported in rats.5 According to Zhang et al. the ICV infusion of etomidate, "significantly decreased isoflurane MAC(tail-clamp)" in the rat model, and was four times more potent than IV infused etomidate.5 In addition, the neurological examination was grossly normal on discontinuation of the ICV etomidate infusion.5 Another  -amino butyric acid-mediated anesthetic, pentobarbital, has demonstrated anesthetic (suppressed tail-pinch response) properties when given by ICV infusion in the rat model.6 To our knowledge, the ICV etomidate injection in humans has not been described. Based on this case report, ICV etomidate may produce anesthetic effects similar to those described in the rat model, as evidenced by rapid loss of consciousness and absence of response to endotracheal intubation. It is possible that an acute change in intracranial pressure could have contributed to the rapid loss of consciousness seen after the bolus injection of 20 mL of fluid into the right cerebral ventricle. Any change in intracranial pressure would have been sustained for only about 1 min until an equal volume of fluid was withdrawn from the ventriculostomy catheter. Moreover, the patient was hemodynamically stable during the induction and intubation, which is not consistent with the bradycardia and hypertension that usually accompanies unconsciousness mediated by acute intracranial hypertension. Redistribution of etomidate into the peripheral circulation with subsequent redistribution into the central nervous system is possible but is incompatible with the timing of the onset of unconsciousness seen in the case report.
Since rocuronium (Zemuron) is prepared in an acetic acid solution (pH 4), it is possibly caustic to central nervous system tissues when delivered into the cerebrospinal fluid. Based on the physical properties of rocuronium (octanol:water partition coefficient 0.5 at 20°C), it would tend to remain in the cerebrospinal fluid after injection, with rapid penetration into neuronal tissue unlikely. Nondepolarizing neuromuscular blocking drugs have been shown to act as competitive antagonists at the neuronal acetylcholine receptors in vitro.7 Thus, after ICV injection, rocuronium could theoretically exert a receptor-mediated effect at neuronal acetylcholine receptors. It is also possible that, after ICV injection, rocuronium could redistribute to the peripheral circulation and result in neuromuscular blockade. This would undoubtedly take more than 1 min to occur; thus, immobility during tracheal intubation one minute after ICV rocuronium is unlikely due to peripheral neuromuscular blockade.
In light of the patient’s complex underlying pathology and lack of autopsy, it is impossible to know what long-term effect the ICV injection of these drugs had on this patient.
Footnotes
Accepted for publication October 18, 2007.
REFERENCES
- Kristof RA, Clusmann H, Koehler W, Fink KB, Schramm J. Treatment of accidental high dose intraventricular mezlocillin application by cerebrospinal fluid exchange. J Neurol Neurosurg Psychiatr 1998;64:379–81[Abstract/Free Full Text]
- Lau G. Accidental intraventricular vincristine administration: an avoidable iatrogenic death. Med Sci Law 1996;36:263–5[Web of Science][Medline]
- Pardridge, WM. Drug delivery to the brain. J Cereb Blood Flow Metab 1997;17:713–31[Web of Science][Medline]
- Grossman SA, Reinhard CS, Loats HL. The intracerebral penetration of intraventricularly administered methotrexate: a quantitative autoradiographic study. J Neurooncol 1989;7:319–28[Medline]
- Zhang Y, Stabernack CR, Dutton R, Sonner J, Trudell JR, Mihic SJ, Yamakura T, Harris RA, Gong D, Eger EI II. Luciferase as a model for the site of inhaled anesthetic action. Anesth Analgesia 2001;93:1246–52[Abstract/Free Full Text]
- Jugovac I, Imas O, Hudetz AG. Behavioral and electroencephalographic effects of intracerebroventricularly infused pentobarbital, propofol, fentanyl, and midazolam. Anesthesiology 2006;105: 764–78[Web of Science][Medline]
- Jonsson M, Gurley D, Dabrowski M, Larsson O, Johnson EC, Eriksson LI. Distinct pharmacologic properties of neuromuscular blocking agents on human neuronal nicotinic acetylcholine receptors: a possible explanation for the train-of-four fade. Anesthesiology 2006;105:521–33[Web of Science][Medline]
This article has been cited by other articles:
|
|
|
|
 
G. A. Mashour
Etomidate, Nitric Oxide, and Cerebral Vasospasm
Anesth. Analg.,
July 1, 2008;
107(1):
343 - 343.
[Full Text]
[PDF]
|
|
|
|
|
|
|
S. M. Howell and R. P. Driver
Etomidate, Nitric Oxide, and Cerebral Vasospasm
Anesth. Analg.,
July 1, 2008;
107(1):
343 - 344.
[Full Text]
[PDF]
|
|
|
|
