Closed-Loop Control of General Anesthesia: My Clinical Experience

Closed-Loop Control of General Anesthesia: My Clinical Experience Ngai LIU, M.D. n.liu@hopital-foch.org 33(1)-46-25-27-67 Service d Anesthésie Hôpital Foch Suresnes France

JAMA 195, 144 (13) 5 patients, ETHER Major Surgical procedures varying age, both sexes Without untoward effect

Feasibility of Feedback control of Propofol During GA Maintenance Using the BIS monitor -Mortier/Struys: Anaesthesia 1998 Closed-loop controlled administration of propofol using BIS -Morley: Anaesthesia 2 Closed-loop control of anaesthesia: An assessment of the BIS -Struys: Anesthesiology 21 Comparison of closed-loop controlled administration of propofol.. -Leslie/Absalom: Anaesthesia 22 Closed-loop colonoscopy using the BIS -Absalom: Anesthesiology 22 Closed-loop control under combined general and regional anesthesia -Absalom : BJA 23 Closed-loop control of propofol using the BIS minor surgery -Ihmsen/ Naguib: BJA 27 Teletherapeutic drug administration by long distance closed-loop control of propofol -Puri: Anaesth. Intensive Care 27 Closed-loop anaesthesia delivery system using bispectral index -Haddad : IEEE Neural Network 27 Neural Network Adaptive Output Feedback for Intraoperative Anesthesia

Proportional-Differential + TCI propofol (Schnider) + Feed-forward

June 26, pp 47-475 Prospective randomized study 2 patients/group Manual vs Automated induction

We wanted to compare our closed-loop controller to the manual TCI Titration in varying conditions of routine clinical anesthesia. Including minor and major surgery. The titration of remifentanil TCI during the surgical procedure was decided by the physician in charge of the patient in both groups. End point : 4 < BIS < 6 during maintenance.

Percentage BIS 4-6 5 4 5 Manual (n=81) 4 3 3 2 2 21 16 1 11 5 12 6 5 1 2 3 4 46 18 17 1 Closed-loop (n=83) 5 6 7 8 9 1 3 1 2 3 4 5 6 7 8 7 ± 21 vs 89 ± 9, p =.1 Adequate anesthesia was more frequent in the Closed-loop group 9 1

Gray Br Med J 1952 Hypnotic Hypnosis Opioid Surgery? Analgesia Dynamic Balance Paralysis HR, MAP, Pupillometry NMBA

Case report Male 72 years History of coronary disease (CABG and Angioplasty) Beta-blocker Tympanoplasty General Anesthesia Closed-loop Propofol Manual TCI of Remifentanil Bolus of Atracurium

TI draping Arousal Incision Instability! Oscillation! Propofol Target Remi (ng/ml) 6 4 8 12 14 16 2

Movement Your hypothesis?

Anesthesia depth Dynamic Balance between Hypnosis and Analgesia Poor Analgesia BIS Instability (oscillation)!

GA = LOC + Inhibition of noxious stimuli P. Glass Anesthesiology 1998, 88; 5-6 T. Bouillon et al Anesthesiology 24 With a Closed-loop BIS-Propofol system, we obtain a stable level of hypnosis and the BIS can also reflect the degree of inhibition of response to noxious stimuli. If BIS suddenly increases in response to noxious stimuli, this could be a deficit in the analgesic component of anesthesia, and this requires an increase of analgesia. We have implemented a second closed-loop controller between Remifentanil and BIS using a Proportionaldifferential algorithm

GA = LOC + Inhibition of noxious stimuli P. Glass Anesthesiology 1998, 88; 5-6 T. Bouillon et al Anesthesiology 24 With a Closed-loop BIS-Propofol system, we obtain a stable level of hypnosis and the BIS can also reflect learnoffrom failure the degree ofwe inhibition response to noxious stimuli. Not from success. If BIS suddenly increases in response to noxious stimuli, this could be a deficit in the analgesic component of anesthesia, and this requires to increase analgesia. We have implemented a second closed-loop controller between Remifentanil and BIS using a Proportionaldifferential algorithm

Male 27 years, Kidney Transplantation, HTA grefferénale Kidney Tx 12 1 Bis 8 6 4 2 1 31 61 91 121 Time (secondes) 151 181 211 241

Male 27 years, Kidney Transplantation, HTA grefferénale 15 12 13 1 11 8 Bis 9 7 6 5 4 3 2 1-1 1 31 61 91 121 Time (Seconds) 151 181 211 241 propofol

Male 27 years, Kidney Transplantation, HTA Incision Retractor grefferénale 15 12 13 1 11 8 Bis 9 7 6 5 4 3 2 1-1 1 31 61 91 121 Temps (en secondes) 151 181 211 241 propofol

Male, Kidney Transplantation, HTA Incision Retractor grefferénale Titre du graphique 16 12 14 1 12 8 Bis 1 8 6 6 4 4 2 2 1 31 61 91 121 Temps (en secondes) 151 181 211 241 remifentanil propofol

Propofol and Remifentanil guided solely by the BIS : Manual vs Closed-loop Prospective Randomized, Multicenter After informed consent Ethical Committee and AFSSAPS approval Foch, Argenteuil, Angers, Besançon. Manuel vs Dual-loop 1 patients per group End points Percentage 4-6 Movement, HD Behavior of the controller during all types of surgery ASA IV A854 ASA Meeting Chicago 26

Manual Dual-loop ASA Meeting Chicago 26

Percentage of Time 4-6 5 Number of patients 4 3 Dual-loop (n=83) Manual (n=84) 2 1 1 2 3 4 5 6 7 8 9 1 % 71 % ± 19 vs 82 % ± 12 (p <.1)

ASA Meeting San Francisco 27 A 1881 Feedback control of Propofol-SE State Entropy : irregularity (1-32 Hz) : Depth of hypnosis Response Entropy (1-47 Hz) Remifentanil-(RE-SE) (RE-SE) represents (32-47 Hz) : Facial EMG activity or adequacy of anti-nociception

Manual (n=31) SE RE % of patients Dual-loop (n=27) 1 1 8 8 6 6 4 4 2 2 1 1 8 8 6 6 4 4 2 2 1 1 5 5 2 4 6 8 1 12 14 16 18 2 4 6 8 1 12 14 16 18 minutes

Dual-loop in ICU, Prospective randomized study (E. Bourgeois SFAR 27) Manual (n=14) vs Dual-loop sedation (n=15) 2D Graph 12 nombre de modifications de cibles/h 1 % 67 ± 22 vs 38 % ± 21 8 6 5 % 4 2 Propofol 1 Dual-loop 2 Manual 4 < BIS < 6 BIS < 4 BIS > 6

What did the Closed-loop controller teach me?

16 Entropy 14 BIS Female 2 years, 1.7 m, 5 kg, PK Schnider Closed-loop propofol, Remi fixed (4 ng/ml) 12 1 8 6 4 2 12::7 AM 12:5:12 AM 12:1:12 AM 12:15:12 AM 12:2:12 AM 12:25:12 AM Tim e 12:3:12 AM 12:36:7 AM 12:41:7 AM 12:46:7 AM 12:51:7 AM

16 Entropy 14 BIS Female 2 years, 1.7 m, 5 kg, PK Schnider Closed-loop propofol (1%), Remi fixed (4 ng/ml) 12 1 8 6 4 2 12::7 AM 12:5:12 AM 12:1:12 AM 12:15:12 AM 12:2:12 AM 12:25:12 AM Tim e 12:3:12 AM 12:36:7 AM 12:41:7 AM 12:46:7 AM 12:51:7 AM

16 Entropy 14 BIS Female 2 years, 1.7 m, 5Piglet kg, PK Schnider 22 Kg Closed-loop propofol, Remi fixed (4 ng/ml) 3 month! 12 1 8 6 4 2 12::7 AM 12:5:12 AM 12:1:12 AM 12:15:12 AM 12:2:12 AM 12:25:12 AM Tim e 12:3:12 AM 12:36:7 AM 12:41:7 AM 12:46:7 AM 12:51:7 AM

16 Entropy 14 BIS Propofol 2 % 12 1 8 6 4 2 ::7 :5:12 :1:12 :15:12 :2:12 :25:12 :3:12 temps :36:7 :41:7 :46:7 :51:7

Paediatr Anaesth. 27 Sep;17(9):99-1. 9-years-old boy 25 kg Giant lobar emphysema SpO2 = 85 % with 1 l/min O2 Dual-loop Propofol-Remifentanil Induction and Maintenance Intrathecal morphine Right postero-lateral thoracotomy Middle lobectomy Lower lobe atypical resection

Closed-loop administration of propofol and remifentanil guided by the Bispectral Index for transsphenoidal surgery in a patient with extreme gigantism Age : 24 Weight : 125 kg Height : 2.48 m! Blade # 6

Induction Propofol.9 mg/kg Remifentanil 1.9 µg/kg Maintenance 5.2 mg/kg/h.3 µg/kg/min 1 9 8 7 6 5 4 3 2 1 Bis 4-6 : 83 % < 4 : 15 % > 6 : 2 % GS : 23 Extubation : 7 min

Female, 69 years, 1.6 m, 85 kg Aortic Stenosis Aortic valve repair Bis 18 16 14 12 1 Bis 8 6 4 2 :: :5:11 :1:22 :15:33 :2:44 :25:55 :31:6

Female, 69 years, 1.6 m, 85 kg Aortic Stenosis Aortic valve repair 18 16 14 12 1 Bis Sys 8 6 4 2 :: :5:11 :1:22 :15:33 :2:44 :25:55 :31:6

Propofol :.8 mg/kg Remifentanil : 2.6 µg/kg Female, 69 years, 1.6 m, 85 kg Aortic Stenosis Aortic valve repair 18 12 16 1 14 12 8 Bis 1 6 8 propofol remifentanil 6 4 4 2 2 :: Sys :5:11 :1:22 :15:33 :2:44 :25:55 :31:6

Feedback control Continuous Titration to effect compensates for most of PK Errors! Model Concentration Drug Safety +++ The impact of the PK model in a controller is low

Safety Closed-loop controller during cardiac arrest Propofol Concentration (ng/ml) BIS 1 2,5 8 2, 6 1,5 4 1, 2,5 5 1 15 2 25 3 35 4 45 5 Minutes

Safety Closed-loop controller during cardiac arrest Propofol Concentration (ng/ml) BIS 1 2,5 8 2, 6 1,5 4 1, 2,5 5 1 15 2 25 3 35 4 45 5 Minutes

Conclusion Feasibility of Dual-loop Reliable and safe Closed-loop controller could fundamentally change the practice of Anesthesia!!! The clinician will always hold the ultimate responsibility for patient safety O Hara Anesthesiology 92 n.liu@hopital-foch.org Service d Anesthésie Hôpital Foch Suresnes France

Open loop TCI Semi-closed loop TCI Closed-loop What is the best for the physician?

Emergence Manual TCI (n = 81) Closed-loop (n = 83) p 1 ± 7 7±4.17 Extubation Time (min) 1 9 8 7 6 Manual TCI 5 4 Closed-loop group 3 2 1 5 1 15 2 min 25 3 35 4 Greater predictability and better planning of the recovery phase

Prof. Manfred MORARI

Case male 59 y 1 1 1.6 1.5 1.6 Norepinephrine (mg/h) Main events nursing nursing Fibero TEE 1 12 9 %<4 %<45 %>6 %>7 %3 5 %45 6 %4 6 4 23 2 59 75 94 1 8 7 8 6 BIS 5 6 SR propofol remifentanil 4 4 3 2 2 1 1 721 1441 2161 2881 Time (h) 361 4321 541

# Closed-loop Anesthesia decreases #the incidence of Too Deep Anesthesia % 1 Manual TCI Weldon Anesthesiology 22 : A197 8 23 : A33 Lennmarken ASA Monk, ASA 24 : A1361 Closed-loop # 6 Prolonged deep anesthesia was associated with increased 4 risk of one year mortality! 2 4-6 <45 < 4 >6 Figure 3. Efficiency of the system during maintenance of anesthesia. 4 < BIS < 6 = percentage of time in which the BIS value was between 4 and 6 during the maintenance. BIS < 45 = percentage of time in which the Bispectral index (BIS) value was below 45. BIS < 4 = percentage of time in which the BIS value was below a value of 4. BIS > 6 = percentage of time in which the BIS value was greater a value of 6. Manual TCI: Manual TCI group guided by the BIS. Closed-loop: Closed-loop group. # p <.1, ns: not significant.