Monitored Anesthesia Care for the Acute Ischemic Stroke Patient with End-stage Pulmonary Disease.

The majority of patients who suffer acute ischemic stroke (AIS) from large vessel occlusion are at a significant risk for disability or death. Because patients on veno-arterial extracorporeal membrane oxygenation (VA ECMO) are therapeutically anticoagulated, intravenous recombinant tissue plasminogen activator is contraindicated. For AIS management, these patients must undergo emergent intra-arterial therapy. Presented is a patient on VA ECMO who subsequently suffered a large vessel embolic stroke requiring emergent surgical intervention. The decision by our anesthetic team to perform the procedure under monitored anesthesia care is discussed.

INTRODUCTION

Over a third of acute ischemic strokes (AISs) originate from large vessel occlusion (LVO) and consequently have significant risk for severe disability or death. Due to the increased morbidity and mortality associated with delaying intervention, attempted recanalization should commence as quickly as is reasonably possible. Emergent therapeutic options for ischemic stroke include either intravenous (i.v.) thrombolysis or intra-arterial therapy. First-line management of ischemic strokes should seek to restore cerebral perfusion through i.v. recombinant tissue plasminogen activator (rt-PA). In eligible patients, intravascular rt-PA is the sole medical intervention shown to improve outcomes in AIS when administered during the first 3–4.5 h of symptomatology. Unfortunately, only 3%–8.5% of all stroke patients are ever treated with rt-PA,[1] and intra-arterial therapy should be considered whenever i.v. rt-PA is contraindicated because of delayed presentation, recent surgery, or coagulopathy. Recently, the paradigm of care for patients presenting with strokes from LVO has shifted to favor endovascular surgery over medical management.[2] Multiple trials have demonstrated the benefits of early and aggressive treatment given the poor natural history of LVO strokes and general resilience of LVO strokes to intravascular rt-PA.[3]

For the anesthesia team, the decision between general anesthesia and monitored anesthesia care (MAC) should be made weighing the risks and benefits for the individual patient. The axiom “time is brain” applies to all aspects of AIS care, and the choice of anesthetic technique in AIS-LVO patients should facilitate swift treatment in the safest manner possible. General anesthesia offers the benefits of patient immobility, pain control, and airway protection. In AIS-LVO patients, endovascular therapies for recanalization performed under general anesthesia have significantly poorer outcomes than those performed under conscious sedation.[456] The reasons for the poorer outcomes are a matter of debate[78] but have been attributed to increased hemodynamic instability exacerbating cerebral ischemia, delays in treatment time, and neurotoxicity from the general anesthetic agents themselves.[49] Conscious sedation offers the benefits of improved hemodynamic stability and reduced time to induction, but sedated patients are prone to continued moment, which in turn may lengthen procedural time and precipitate iatrogenic complications. In addition, the sedated airway is unprotected and may be vulnerable to apneic episodes, which may worsen ischemia. We present the case of a 60-year-old female with left middle cerebral artery (MCA)-M1 occlusion undergoing emergent endovascular thrombectomy under MAC. The decision to pursue MAC will be discussed.

CASE REPORT

A 60-year-old African-American female (77 kg, 167 cm) with no known drug allergies presented to the operating room for emergent thrombectomy. Her symptoms included minimally responsive behavior and acute motor aphasia with right upper and lower extremity weakness. Her medical history was significant for depression, hyperlipidemia, scoliosis, end-stage lung disease secondary to fibrotic nonspecific interstitial pneumonia (NSIP), and World Health Organization Group 3 pulmonary hypertension [Figure 1]. Before her most recent admission for increasing dyspnea, she was a New York Heart Association classes III and IV on home oxygen (8–15 L/min oxygen requirement) and pirfenidone for her NSIP while awaiting lung transplant. In the months prior, she had multiple admissions and was treated with diuresis, steroids, and i.v. antibiotic therapy for similar complaints of worsening dyspnea thought to be a progression of her lung disease in the setting of a pulmonary hypertensive crisis and volume overload.

Figure 1

World Health Organization classification of pulmonary hypertensive diseases

Three months prior, the patient was again admitted for dyspnea at rest. Her initial blood, respiratory, and urine cultures were all negative with a decreasing white blood cell count. A transthoracic echocardiography showed borderline left ventricular hypertrophy (60% ejection fraction), stable right ventricular size increases with severely decreased right ventricular systolic function, mild tricuspid regurgitation, and pulmonary artery systolic pressures of 60 mmHg. Computed tomography (CT) angiogram of the chest failed to demonstrate pulmonary embolus or consolidation.

The patient was diuresed for 3 days, and her brain natriuretic peptide decreased; however, her oxygen requirement and dyspnea failed to improve. She was started on high dose i.v. steroids for presumed flare of her NSIP and prophylactic i.v. antibiotic coverage against healthcare-associated pneumonia. Eventually, she was transferred to the Medical Intensive Care Unit (MICU) where inhaled iloprost was started for symptomatic pulmonary hypertension, and venoarterial extracorporeal membrane oxygenation (VA-ECMO) diverted from the right internal jugular vein to the right brachiocephalic artery with a therapeutic heparin drip was initiated as a bridge to lung transplant.

One week following transfer, these measures were complicated by a decreasing hemoglobin and new onset hemothorax requiring partial thromboplastin time goal reduction. The patient subsequently became subtherapeutic on her heparin drip and developed new onset right upper and lower extremity weakness with a progressing motor aphasia. CT angiogram of the head and neck revealed distal left MCA-M1 occlusion. She was taken for emergent thrombectomy [Figure 2].

Figure 2

Frontal view of left internal carotid angiograms. (a) Pretreatment angiogram shows total occlusion of the proximal M1 segment of the left middle cerebral artery (white arrow). (b) Immediate posttreatment intraoperative angiogram shows complete patency of the M1 segment

The patient was brought to the neurological intervention suite, and standard American Society of Anesthesiologists monitors were attached, a left brachial arterial line was placed for invasive blood pressure monitoring, her high flow nasal cannula was replaced with bilevel positive airway pressure (BiPAP) with inspiratory positive airway pressure/expiratory positive airway pressure set at 20 cm H2O/5 cm H2O, respectively, on 100% FiO2, with proper end-tidal capnography monitoring, VA-ECMO, and the patient's heparin drip were continued throughout, and both of her peripheral i.v. catheters were maintained. Baseline vitals were sinus tachycardia at 102 b.p.m, blood pressure of 155/111 mmHg, respiratory rate of 24, and SpO2 of 91%. A baseline arterial blood gas was drawn demonstrating pH of 7.40, PaCO2 of 65 mmHg, PaO2 of 70 mmHg, and HCO3− of 40.8. Cefazolin 2 g was administered for antimicrobial prophylaxis, and 1 g of levetiracetam was administered for antiepileptic prophylaxis. A propofol drip was started at 30 ug.kg-1.min-1, and a phenylephrine infusion was used to maintain mean arterial pressures (MAPs) between 80 and 100 mmHg.

Before infiltration of the right groin by surgical team, a 10 mg bolus of propofol was administered. The patient remained sedated and hemodynamically stable with spontaneous respirations assisted with BiPAP throughout the entire case. During the case, propofol boluses of 10 mg were used to supplement the continuous i.v. infusion.

Once the surgical team had mechanically evacuated the patient's MCA clot, the phenylephrine infusion was halted, and a nicardipine drip was titrated to maintain MAPs between 65 and 85 mmHg. Propofol sedation was stopped upon closure and hemostasis of the right groin entry site. The patient was placed back on high flow nasal cannula and allowed to emerge from sedation, where an immediate neurological examination demonstrated improvement in speech as well as right upper and lower extremity strength. The patient was transferred back to the MICU on VA-ECMO, heparin drip, and a nicardipine drip. The patient subsequently was accepted for and underwent a double lung transplant 3 days after her MCA thrombectomy.

DISCUSSION

There has been no gold standard for the anesthetic management of AISs.[10] The decision to pursue MAC versus general anesthesia has been based on clinical and individual patient characteristics coupled with surgical preferences. Some patients with acute strokes present with severely altered mental statuses, inability to reliably protect their airway from aspiration, involuntary sporadic movements, and intolerance to any surgical stimulation without general anesthesia. To facilitate surgical clot lysis and removal, the anesthesia provider undergoes the induction, maintenance, and emergence from general anesthesia understanding the morbidity and mortality with the hemodynamic changes involved. Our patient's severe cardiopulmonary disease coupled with her active listing on the lung transplant waiting list meant that a general anesthetic technique would not only drastically increase her morbidity and mortality but also risk losing her active status on the lung transplant list given the likelihood of a prolonged mechanical ventilation.

Monitoring cerebral oximetry with near-infrared spectroscopy (NIRS) may noninvasively measure regional cerebral perfusion, recanalization, and desaturations from sedation hypoventilation.[11] Cerebral oximetry was not directly measured in our patient, and future studies should investigate the prognostic value of NIRS in stroke-specific setups. We assessed systemic oxygenation status from the left radial arterial line because our patient was supported with a central VA-ECMO diverted from the internal jugular vein to the right brachiocephalic artery.

The pharmacological strategy for this case was to primarily maintain spontaneous ventilation with noninvasive positive pressure support through BiPAP to avoid severe hypercarbia and pulmonary hypertensive crisis. In addition, our secondary goal was to provide the patient with enough sedation to tolerate the surgical procedure and prevent acute changes in hemodynamic responses to noxious stimuli. Hypotension caused by propofol sedation was counterbalanced with an i.v. infusion of phenylephrine (20–60 mcg/min). Fentanyl and other i.v. opioids were avoided during the case given their intrinsic and synergistic potential to depress our patient's respiratory efforts when used with propofol sedation. Midazolam was avoided given our patient's baseline alteration in mental status.

Multiple other medications could have been used for this case. Dexmedetomidine could have been used for its ability to produce sedation, anxiolysis, and analgesia without significantly depressing respirations although it has not been shown to be superior to propofol alone.[12] Inhaled nitric oxide or prostaglandin therapy could have been used to offload the right ventricle. Milrinone, epinephrine, and dobutamine could have been used to further increase inotropy. Norepinephrine and vasopressin could have been used for hemodynamic support. Some of these medications would have required a central venous catheter given their potential to cause phlebitis. Placement of a central venous catheter for closer hemodynamic monitoring and administration of medications was taken into consideration preoperatively. Given that our patient was hemodynamically stable and responsive to minimal doses of phenylephrine, it was decided that the morbidity associated with the delay in surgical intervention outweighed any benefits of having a central line in place.

CONCLUSION

MAC was chosen in the setting of this acute stroke because our patient was listed on the lung transplant waiting list and prolonged intubation would risk losing her active status, there was a perceived increase in morbidity associated with general anesthesia, the patient was not obtunded, and the airway was felt to be sufficiently protected.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.