Background Enhanced recovery after surgery (ERAS) programs are a model of care that aim to improve patient outcomes, reduce complications, and facilitate recovery while reducing healthcare-associated costs and admission length. While such programs have been developed in other surgical subspecialties, there have yet to be guidelines published specifically for laser interstitial thermal therapy (LITT). Here we describe the first multidisciplinary ERAS preliminary protocol for LITT for the treatment of brain tumors. Methods Between the years 2013 and 2021, 184 adult patients consecutively treated with LITT at our single institution were retrospectively analyzed. During this time, a series of pre, intra, and postoperative adjustments were made to the admission course and surgical/anesthesia workflow with the goal of improving recovery and admission length. Results The mean age at surgery was 60.7 years with a median preoperative Karnofsky performance score of 90 ± 13. Lesions were most commonly metastases (50%) and high-grade gliomas (37%). The mean length of stay was 2.4 days, with the average patient being discharged 1.2 days after surgery. There was an overall readmission rate of 8.7% with a LITT-specific readmission rate of 2.2%. Three of 184 patients required repeat intervention in the perioperative period, and there was one perioperative mortality. Conclusions This preliminary study shows the proposed LITT ERAS protocol to be a safe means of discharging patients on postoperative day 1 while preserving outcomes. Although future prospective work is needed to validate this protocol, results show the ERAS approach to be promising for LITT.

Abstract INTRODUCTION The concept of enhanced recovery after surgery (ERAS) due to standardized interventions has been gaining more relevance within neurosurgery. Advances were made both in protocols for spine and cranial surgery. These experiences described many benefits such as less psychological stress, reduction in hospitalization days, and lower hospital costs, without increasing the incidence of complications. However, no experience has described to date the applicability of an ERAS program for laser ablation thermal therapy (LITT). OBJECTIVE To describe our initial experience with the first enhanced recovery program reported for laser ablation for brain tumors. Secondly, to summarize the perioperative clinical outcomes of ERAS applied to LITT. METHODS We performed a retrospective analysis of all adult patients who underwent LITT for oncological lesions from 2013 to 2021. A multidisciplinary program was created by protocolizing interventions carried out along the path of the patient's hospitalization. Each recommendation was individually assessed for its appropriateness for enhancing recovery and for its validity with a focused literature review process. RESULTS A total of 184 patients were included, with a mean age of 60.7 ± 13.5 years, 35% males. 167 tumors were located in the supratentorial compartment, and 17 were infratentorial; the mean tumor diameter was 1.84 ± 1.04 cm. Among the pathologies treated 50.0% were metastasis, and 36.9% were glioblastomas. The mean postoperative day discharge was 1.2 ± 0.8 days. The readmission rate due to surgical complications within 30 days of surgery was 2.7%. These readmission rates fall into what is expected according to published literature without an ERAS program and longer hospital admissions. One death was recorded in the perioperative period. CONCLUSION Clinical interventions that could constitute an ERAS program are feasible in laser ablation of brain tumors. This study could be useful as a preliminary framework for the development of future guidelines.

BACKGROUND:Processed electroencephalogram (EEG) monitors help assess the hypnotic state during general anesthesia or sedation. Maintaining the bispectral index (BIS) or state entropy (SE) between 40 and 60 has been recommended to mitigate anesthesia awareness. Nonetheless, SEs > 70 were frequently observed at end-tidal sevoflurane concentrations unlikely to allow awareness. We sought to determine the prevalence of elevated discordant measurements during BIS and SE monitoring.METHODS:Electronic data collected over 11 months at two academic hospitals were retrospectively reviewed. At the hospital using SE, all cases were included with patients ≥ 18 yr and sevoflurane administered for at least 30 min during surgery. A cohort of cases propensity matched by age and American Society of Anesthesiologist Physical Status were selected from the hospital using BIS. Elevated discordant EEG indices were defined as values > 70 occurring during stable end-tidal sevoflurane concentrations > 1.5%. The odds ratio (OR) based on the probability of a case having at least one elevated discordant SE or BIS lasting ≥ two minutes (primary endpoint) was calculated.RESULTS:At each hospital, 3,690 cases were studied. The mean (95% confidence interval [CI]) incidence of cases with at least one interval of an elevated discordant EEG index lasting at least two minutes was 3.6% (2.8% to 4.4%) for SE compared with 0.24% (0.17% to 0.27%) for BIS (pooled OR, 17.0; 95% CI, 8.3 to 34.7; P < 0.001).CONCLUSIONS:The prevalence of an elevated discordant EEG index is much greater with SE than with BIS. Elevated index values occurring at anesthetic concentrations well above the awareness threshold need to be assessed to determine if they indicate an inadequate depth of anesthesia requiring treatment or if they simply reflect the underlying monitoring algorithm.

In this paper, the development and implementation of L1 adaptive control designs for anesthesia delivery to patients in surgical settings is presented. Our main objectives are the design of model-based feedback controllers ensuring that the patient׳s bispectral index profile tracks a prespecified reference trajectory, and demonstrates robustness to inter-patient variability. Patient models are constructed based on clinical trial data and gray box system identification methods. Controller switching mechanisms and specific safety measures are considered in the design and discussed in the paper. Simulation results are provided, demonstrating the effectiveness of the control methods.

The bispectral (BIS) monitor uses brain electroencephalographic data to measure the depth of sedation and pharmacological response during anaesthetic procedures. In this case, the BIS monitor was used for another purpose, to demonstrate postoperatively to the nursing staff that a patient with history of locked-in syndrome (LIS), who underwent pressure ulcer debridement, had periods of wakefulness and apparent sensation, even with his eyes closed. Furthermore, as patients with LIS can feel pain, despite being unable to move, local block or general anaesthesia should be provided for sharp surgical debridement and other painful procedures. This use of the BIS has shown that as a general rule, the staff should treat the patient as though he might be awake and sensate even if he does not open his eyes or move his limbs. The goal of this study was to continuously monitor pain level and communicate these findings to the entire wound team, i.e. anaesthesiologists, surgeons and nurses.

Previous chapter Next chapter Full AccessProceedings 2013 Proceedings of the Conference on Control and its Applications (CT)Regulation of Anesthesia Delivery via ℒ1-Adaptive ControlEvgeny Kharisov, Carolyn L. Beck, and Marc Bloom, M.DEvgeny Kharisov, Carolyn L. Beck, and Marc Bloom, M.Dpp.191 - 198Chapter DOI:https://doi.org/10.1137/1.9781611973273.26PDFBibTexSections ToolsAdd to favoritesExport CitationTrack CitationsEmail SectionsAboutAbstract In this paper, we present an application of ℒ1-adaptive control design to anesthesia delivery during surgery. Our main objectives are the design of feedback controllers ensuring that the patient's Bispectral Index profile tracks a prespecified reference trajectory, and demonstrating robustness to inter-patient variability. For the design of the controllers we developed patient models based on clinical trial data. Controller switching mechanisms and specific safety measures are considered in the paper. Simulation results are provided demonstrating the effectiveness of the control methods. Previous chapter Next chapter RelatedDetails Published:2013eISBN:978-1-61197-327-3 https://doi.org/10.1137/1.9781611973273Book Series Name:ProceedingsBook Code:PRCT13Book Pages:1-229

INTRODUCTION:The objective of this study was to determine the effects of isoproterenol infusion on level of consciousness during ablation using total intravenous anesthesia.METHODS AND RESULTS:Seven patients undergoing total intravenous anesthesia for atrial fibrillation ablation were monitored for level of consciousness using bispectral EEG levels (BIS). Isoproterenol infusion was performed after the ablation during anesthesia. BIS levels prior to, during, and post-isoproterenol infusion were recorded and correlated to isoproterenol infusion doses. In all patients, BIS levels significantly increased during isoproterenol infusion (median BIS prior to infusion, 46; during infusion, 64 (p < 0.02)). With a subsequent increase in anesthetic medication, BIS levels could again be reduced.CONCLUSION:Isoproterenol infusion alters consciousness level during total intravenous anesthesia for atrial fibrillation ablation. BIS monitoring is a novel way to modulate anesthesia during ablation to potentially optimize patient comfort and ablation success.

BACKGROUND:The perioperative course of patients undergoing intracranial surgery is frequently complicated by hypertensive episodes. Dexmedetomidine (DEX), an alpha-2 adrenoreceptor agonist, is gaining popularity in neuroanesthesia, because its sympatholytic and antinociceptive properties may improve hemodynamic stability at critical moments of surgery. We designed this study to assess the efficacy of DEX in controlling hypertensive responses in patients undergoing intracranial surgery. METHODS:Patients scheduled for elective craniotomy were randomly assigned to receive either sevoflurane-opioid or sevoflurane-opioid-DEX anesthesia. Bispectral index was used to maintain a similar level of hypnosis in both groups (40-50). Opioids, sevoflurane, and vasoactive medications were titrated in a routine manner, at the discretion of the blinded anesthesiologist managing the case, to maintain systolic blood pressure (SBP) targeted within 90-130 mm Hg and heart rate (HR) between 50 and 90 bpm. Hemodynamic variables were continuously recorded and stored on a computer for analysis. Efficacy of the anesthetic technique in controlling SBP or HR is inversely proportional to the area under the curve (AUC) outside the targeted range. Areas under the curves above and below targeted ranges for SBP-time (AUCsbp mm Hg*min/h) and HR-time (bpm*min/h) were compared. Coefficient of variation was used to assess hemodynamic stability. RESULTS:Seventy-two patients were recruited for the study. Computerized records of 56 patients only were analyzed because of technical problems with data collection in 14 cases. AUCsbp for above the targeted range was significantly lower for patients in the DEX group (P=0.044). The coefficient of variation for SBP or HR did not differ between groups. A significantly smaller proportion of patients in the DEX group required treatment with antihypertensive medications (12 of 28, 42% vs 24 of 28, 86%, P=0.0008). The DEX group required fewer opioids in the intraoperative period, but there were no differences in the use of sevoflurane. In the postanesthesia care unit, patients in the DEX group had fewer hypertensive episodes (1.25+/-1.55 vs 2.50+/-2.00, P=0.0114) and were discharged earlier (91+/-17 vs 130+/-27 min, P<0.0001). There were no differences in the requirement for postoperative opioids or antiemetics. CONCLUSIONS:By using indices, which assess a global hemodynamic stability of the anesthetic, we determined that intraoperative DEX infusion was effective for blunting the increases in SBP perioperatively. The use of DEX did not increase the incidence of hypotension or bradycardia, common side effects of the drug.

In this paper, we present initial linear parameter-varying control efforts aimed at closed loop control of anesthesia delivery during surgery. The control designs are completed using a recent system identification based MIMO modeling framework for anesthetic pharmacodynamics.

Engineering and control technology have played and continue to play a major role in medicine over the past half-century, from the invention of the pacemaker in 1950 and ventricular assist devices in the 1980’s, to more recent advances incorporating robotic and image-guided surgery. Modeling and control of drug dosing in clinical pharmacology is one area of medicine in which mathematical modeling is used extensively, and hence is well-suited for applications of control design and analysis techniques. Obvious examples of potential application areas for control in clinical pharmacology include insulin delivery for control of diabetes, chemotherapy dosing and timing for treatment of HIV, and anesthetic dosing during surgery to optimally control sedation level and manage hemodynamic functions. The increasing use of computers in the operating room combined with the recent and ongoing development of non-invasive yet effective means of measuring a number of the goals of anesthesia, such as the bispectral index or BIS measure of sedation, real-time measurements of exhaled gas concentrations by spectroscopic methods, and the use of electromyographic methods to measure lack of movement, promise to make the incorporation of control techniques into the anesthetic delivery process imminent.

El síndrome de Rett (SR) es un trastorno neurológico que afecta principalmente a las mujeres. Se caracteriza por un retraso mental severo después de un periodo de desarrollo aparentemente normal en la infancia temprana. Las principales características incluyen autismo, espasticidad, microcefalia, escoliosis, estereotipias, control anormal de la respiración y convulsiones. Se ha notificado que son muy sensibles a sedantes y tienen una lenta recuperación luego de una anestesia. Nosotros describimos y discutimos el manejo anestésico de una mujer de 26 años con SR y fractura distal del húmero, tratada con reducción y fijación con agujas de Kirschner bajo anestesia regional y sedación; quien tuvo depresión respiratoria luego de administrar 1 mg de midazolam i.v. intraoperatorio. Nuestra experiencia con este caso nos hace concluir y sugerir que no se deben usar benzodiacepinas en SR, y si no existe contraindicación, la anestesia regional es la técnica de elección.Rett Syndrome (RS) is a neurological disorder that almost exclusively affects females. RS is characterized by severe mental retardation following a period of apparent normal development in early childhood. The main clinical features include autism, spasticity, microcephaly, scoliosis, stereotyping, abnormal respiratory control and seizures. Patients with RS have been reported to be extremely sensitive to sedatives and exhibit a slow recovery from anesthesia. We describe and discuss the anesthetic management of a 26-year-old woman with RS and distal fracture of the humerus that underwent reduction and fixation with Kirschner'swires under regional anesthesia and sedation. Intraoperatively, the patient experienced respiratory depression following the administration of 1 mg of I.V. midazolam. Our experience in this case led us to conclude and suggest that benzodiazepines should not be used in RS and in the absence of contraindications the technique of choice should be regional anesthesia.

s: 33rd Annual Meeting of the Society of Neurosurgical Anesthesia and Critical Care: MONITORING/ELECTROPHYSIOLOGY

Anesthesiologists are often asked to provide sedation for ERCP and their choice of medications for monitored anesthesia care (MAC) is largely operator dependent. When propofol is used, patients often cross into deep sedation, which has been shown to correspond to an average Bispectral Index monitoring (BIS) level of 70 (Bower AL et al. Gastrointest Endosc. 2000 Aug;52(2):192-6). The aim of our study was to compare BIS rates with vital signs and clinical assessment during ERCP in patients receiving MAC. Methods: After institutional IRB approval, 23 adults, ASA 1 to 3 were enrolled in the study. The BIS monitor was recorded throughout, but hidden from the anesthesiologist. Vital signs, patient's clinical signs of consciousness, and the timing and doses of all drugs administered were recorded. Results: The mean heart rate, blood pressure and O2 saturation at the end of the procedure were not significantly different from the start values. However, the mean BIS score of 63 at the end of the procedure was significantly different from the mean of 95 at the start (p < 0.0001), representing a deep level of sedation at the time of withdrawal of the endoscope. 16 patients received extra sedation during the procedure. In 3 of 7 patients (43%) in whom additional doses of fentanyl were given, the BIS score was below 60 at the time. Similar low BIS scores were present in 2 out of 6 patients (33%) in whom the propofol drip rate was increased and in 3 of 11 patients (27%) who received additional boluses of propofol. Eight of the 23 patients in the study (35%) received additional medication despite being already deeply sedated (BIS < 60). None of these 8 patients had clinical signs of oversedation or a drop in O2 saturation to < 90%, and only one of them had a decrease in blood pressure of more than 30% from the initial value. No cardiorespiratory complications occurred in this study. There was great heterogeneity to the drugs and doses and means of administration (bolus vs. drip) used for sedation. Conclusion: Vital signs and clinical signs do not correlate well with the patient's level of sedation. Patients remain deeply sedated upon completion of ERCP making continuous clinical observation essential to assure patients' safety. Knowledge about the BIS score would have been useful to provide additional information about the level of consciousness and presumably would have prevented additional amounts of sedatives been given to patients already at a deep level of sedation.

The standard modeling paradigm used to describe the relationship between input anesthetic agents and output patient endpoint variables are single-input single-output pharmacokinetic-pharmacodynamic (PK-PD) compartment models. In this paper, we propose the use of multivariable piecewise-linear models to describe the relations between inputs that include anesthesia, surgical stimuli and disturbances to a variety of patient output variables. Subspace identification methods are applied to clinical data to construct the models. A comparison of predicted and measured responses is completed, which includes predictions from PK-PD models, and piecewise-linear time-invariant models.