Rate and Prognosis of Patients under Conscious Sedation Requiring Emergent Intubation during Neuroendovascular Procedures

Discussion

Due to the lack of standardization in large device trials,^4,5 many neurointerventionalists are reluctant to treat patients under conscious sedation, especially due to the potential risk of airway damage and aspiration from an emergent intubation.^3,6 This reason was rated by 43% of interventionalists as the most concerning complication in a survey involving 49 physicians nationwide.² Our study demonstrates that the rate of emergent conversion from conscious sedation to general anesthesia was very low (1.7%). Emergent conversion to general anesthesia was not associated with worse outcomes than those observed in planned general anesthesia.

Previous smaller studies assessing the rates of conversion to general anesthesia during acute ischemic stroke treatment, aneurysm embolization, and carotid artery stent placement have reported rates varying from 1.3% to 3%.^7⇓–9 According to a study by Jumaa et al,⁸ 2 patients (2.7%) were emergently intubated because of conscious sedation failure among 73 patients with acute ischemic stroke treated endovascularly with conscious sedation. In elective carotid stent placements, Chamczuk et al⁹ found that only 2 of 63 (3%) patients had to be converted to general anesthesia. Qureshi et al⁷ demonstrated successful embolization of ruptured and unruptured aneurysms with patients under conscious sedation with low complication rates. Two conversions occurred in 150 aneurysm cases that underwent embolization with Guglielmi detachable coils (Boston Scientific, Natick, Massachusetts). Both ICU and hospital mean length of stay were significantly higher in the general anesthesia cohort compared with conscious sedation in both ruptured (11 versus 17 and 20 versus 30 days) and unruptured (3 versus 11 and 6 versus 20 days) groups. Morbidity and mortality rates for ruptured (1.6% and 3%) and unruptured (2% and 0%) aneurysms were lower than or similar to those in previous studies in which general anesthesia was used (3%–17% and 0%–8%, respectively), though direct comparison is precluded by differences in patient characteristics.

Although some case series have reported the safety of specific individual procedures with the patient under conscious sedation,^{1,7⇓⇓⇓⇓–12} large case series have only reported upon the endovascular treatment of acute ischemic stroke.^1,8 General anesthesia was associated with a longer ICU stay, higher rates of pneumonia, larger infarcts, and worse clinical outcome in studies conducted in patients undergoing acute ischemic stroke treatment.⁸ Avoidance of general anesthesia was also associated with reduced odds of cardiovascular complications 30 days after carotid endarterectomy.^13,14 Presumably, the higher rate is related to occurrence of anesthetic-related deaths and severe cardiovascular adverse events.^15,16 Common adverse events associated with general anesthesia reported by Forrest et al^15,16 included tachycardia (41%), hypotension (31%), hypertension (27%), bradycardia (19%), and ventricular arrhythmias (6%). Detection of clinical events related to iatrogenic thromboembolism is prevented by prolonged recovery from anesthesia; therefore, the possibility of early revascularization is lost.¹⁷ Abou-Chebl et al¹⁰reported that in 37 patients who underwent intracranial angioplasty and stent placement under conscious sedation, operators modified the interventional technique in 61% of patients on the basis of real-time patient reporting of intraprocedural symptoms.

Previous studies have demonstrated significant hospital cost reduction by using conscious sedation instead of general anesthesia. In addition to the procedural costs of anesthetic agents, anesthesia personnel, and mechanical ventilation, considerable expenses can be avoided with reduction of hospital and ICU stays. At some institutions, general anesthesia for a 3-hour embolization procedure may cost $1350⁷; the avoidance this fee reduced mean total hospital cost at 1 institution by 41%.¹⁸

Even so, a technically successful procedure (ie, revascularization after ischemic stroke or technically successful aneurysm embolization) does not necessarily correlate with good outcome. Advocates for conscious sedation argue that the advantage of real-time input from the patient and the clinical examination outweigh the risk of complications from patient mobility.^19,20 Other potential disadvantages of performing these procedures with the patient under general anesthesia include induction-related hypotension, delay in procedure initiation, the need for additional personnel and equipment in the angiography suite, and inherent delay in obtaining a postoperative neurologic examination.

Patient selection is important for successful completion of neuroendovascular procedures with the patient under conscious sedation. Procedural type, anticipated duration, and patient characteristics are important in identifying appropriate patients for conscious sedation. Certain procedures such as extracranial carotid artery stent placement infrequently require general anesthesia because of the limited impact of patient motion, head immobility, and hemodynamic control on procedural success.²¹ However, a complex aneurysm that requires real-time assessment in multiple angiographic projections or a very small aneurysm is best treated with the patient under general anesthesia because even small-magnitude patient movement can interfere with the success of the procedure. Similarly, anticipated prolonged duration in a case of complex dural arteriovenous fistula (several hours of embolization time) may be better performed with the patient under general anesthesia. Patient-related factors such as a poor level of consciousness, communicative disorders, or difficulty in remaining immobile due to a chronic pain disorder need to be evaluated during preprocedural clinical assessment or during a preceding diagnostic cerebral angiogram with the patient under conscious sedation.

Some limitations should be considered before interpretation of our results. Due to a small number of patients in each intervention group, the precision of the estimation regarding conversion rates about individual subgroups defined by procedure type is not optimal. There was also no long-term follow-up in many patients, thereby limiting our ability to determine the long-term consequences of the choice of anesthetic technique. Conversely, the retrospective nature of the study prevented bias in choosing a sedation technique because the practitioners were unaware that this issue would be studied. Thus, decisions were made in real-time on the basis of the clinical data available to the neurointerventionalists. A randomized controlled trial with a standardized inclusion/exclusion protocol is needed to determine the comparative benefits of procedures performed under general anesthesia or conscious sedation.