Automated systems for perioperative goal-directed hemodynamic therapy

Closed-loop anesthesia: foundations and applications in contemporary perioperative medicine

A closed-loop automatically controls a variable using the principle of feedback. Automation within anesthesia typically aims to improve the stability of a controlled variable and reduce workload associated with simple repetitive tasks. This approach attempts to limit errors due to distractions or fatigue while simultaneously increasing compliance to evidence based perioperative protocols. The ultimate goal is to use these advantages over manual care to improve patient outcome. For more than twenty years, clinical studies in anesthesia have demonstrated the superiority of closed-loop systems compared to manual control for stabilizing a single variable, reducing practitioner workload, and safely administering therapies. This research has focused on various closed-loops that coupled inputs and outputs such as the processed electroencephalogram with propofol, blood pressure with vasopressors, and dynamic predictors of fluid responsiveness with fluid therapy. Recently, multiple simultaneous independent closed-loop systems have been tested in practice and one study has demonstrated a clinical benefit on postoperative cognitive dysfunction. Despite their advantages, these tools still require that a well-trained practitioner maintains situation awareness, understands how closed-loop systems react to each variable, and is ready to retake control if the closed-loop systems fail. In the future, multiple input multiple output closed-loop systems will control anesthetic, fluid and vasopressor titration and may perhaps integrate other key systems, such as the anesthesia machine. Human supervision will nonetheless always be indispensable as situation awareness, communication, and prediction of events remain irreplaceable human factors.

Perioperative Fluid and Vasopressor Therapy in 2050: From Experimental Medicine to Personalization Through Automation

Intravenous (IV) fluids and vasopressor agents are key components of hemodynamic management. Since their introduction, their use in the perioperative setting has continued to evolve, and we are now on the brink of automated administration. IV fluid therapy was first described in Scotland during the 1832 cholera epidemic, when pioneers in medicine saved critically ill patients dying from hypovolemic shock. However, widespread use of IV fluids only began in the 20th century. Epinephrine was discovered and purified in the United States at the end of the 19th century, but its short half-life limited its implementation into patient care. Advances in venous access, including the introduction of the central venous catheter, and the ability to administer continuous infusions of fluids and vasopressors rather than just boluses, facilitated the use of fluids and adrenergic agents. With the advent of advanced hemodynamic monitoring, most notably the pulmonary artery catheter, the role of fluids and vasopressors in the maintenance of tissue oxygenation through adequate cardiac output and perfusion pressure became more clearly established, and hemodynamic goals could be established to better titrate fluid and vasopressor therapy. Less invasive hemodynamic monitoring techniques, using echography, pulse contour analysis, and heart-lung interactions, have facilitated hemodynamic monitoring at the bedside. Most recently, advances have been made in closed-loop fluid and vasopressor therapy, which apply computer assistance to interpret hemodynamic variables and therapy. Development and increased use of artificial intelligence will likely represent a major step toward fully automated hemodynamic management in the perioperative environment in the near future. In this narrative review, we discuss the key events in experimental medicine that have led to the current status of fluid and vasopressor therapies and describe the potential benefits that future automation has to offer.

Closing the loop: automation in anesthesiology is coming

Anesthesiology and intensive care medicine provide fertile ground for innovation in automation, but to date we have only achieved preliminary studies in closed-loop intravenous drug administration. Anesthesiologists have yet to implement these tools on a large scale despite clear evidence that they outperform manual titration. Closed-loops continuously assess a predefined variable as input into a controller and then attempt to establish equilibrium by administering a treatment as output. The aim is to decrease the error between the closed-loop controller's input and output. In this editorial we consider the available intravenous anesthesia closed-loop systems, try to clarify why they have not yet been implemented on a large scale, see what they offer, and propose the future steps towards automation in anesthesia.

Associations between clinical interventions and transcutaneous blood gas values in postoperative patients

PURPOSE

Postoperative monitoring of circulation and respiration is pivotal to guide intervention strategies and ensure patient outcomes. Transcutaneous blood gas monitoring (TCM) may allow for noninvasive assessment of changes in cardiopulmonary function after surgery, including a more direct assessment of local micro-perfusion and metabolism. To form the basis for studies assessing the clinical impact of TCM complication detection and goal-directed-therapy, we examined the association between clinical interventions in the postoperative period and changes in transcutaneous blood gasses.

METHODS

Two-hundred adult patients who have had major surgery were enrolled prospectively and monitored with transcutaneous blood gas measurements (oxygen (TcPO2) and carbon dioxide (TcPCO2)) for 2 h in the post anaesthesia care unit, with recording of all clinical interventions. The primary outcome was changes in TcPO2, secondarily TcPCO2, from 5 min before a clinical intervention versus 5 min after, analysed with paired t-test.

RESULTS

Data from 190 patients with 686 interventions were analysed. During clinical interventions, a mean change in TcPO2 of 0.99 mmHg (95% CI-1.79-0.2, p = 0.015) and TcPCO2 of-0.67 mmHg (95% CI 0.36-0.98, p < 0.001) was detected.

CONCLUSION

Clinical interventions resulted in significant changes in transcutaneous oxygen and carbon dioxide. These findings suggest future studies to assess the clinical value of changes in transcutaneous PO2 and PCO2 in a postoperative setting.

TRIAL REGISTRY

Clinical trial number: NCT04735380.

CLINICAL TRIAL REGISTRY

https://clinicaltrials.gov/ct2/show/NCT04735380.

Physiological Modeling of Hemodynamic Responses to Sodium Nitroprusside

BACKGROUND

Computational modeling of physiology has become a routine element in the development, evaluation, and safety testing of many types of medical devices. Members of the Food and Drug Administration have recently published a manuscript detailing the development, validation, and sensitivity testing of a computational model for blood volume, cardiac stroke volume, and blood pressure, noting that such a model might be useful in the development of closed-loop fluid administration systems. In the present study, we have expanded on this model to include the pharmacologic effect of sodium nitroprusside and calibrated the model against our previous experimental animal model data.

METHODS

Beginning with the model elements in the original publication, we added six new parameters to control the effect of sodium nitroprusside: two for the onset time and clearance rates, two for the stroke volume effect (which includes venodilation as a "hidden" element), and two for the direct effect on arterial blood pressure. Using this new model, we then calibrated the predictive performance against previously collected animal study data using nitroprusside infusions to simulate shock with the primary emphasis on MAP. Root-mean-squared error (RMSE) was calculated, and the performance was compared to the performance of the model in the original study.

RESULTS

RMSE of model-predicted MAP to actual MAP was lower than that reported in the original model, but higher for SV and CO. The individually fit models showed lower RMSE than using the population average values for parameters, suggesting the fitting process was effective in identifying improved parameters. Use of partially fit models after removal of the lowest variance population parameters showed a very minor decrement in improvement over the fully fit models.

CONCLUSION

The new model added the clinical effects of SNP and was successfully calibrated against experimental data with an RMSE of <10% for mean arterial pressure. Model-predicted MAP showed an error similar to that seen in the original base model when using fluid shifts, heart rate, and drug dose as model inputs.

Predicting personalised remifentanil effect site concentration for surgical incision using the nociception level index: A prospective calibration and validation study

BACKGROUND

Inadequate antinociception can cause haemodynamic instability. The nociception level (NOL) index measures response to noxious stimuli, but its capacity to predict optimal antinociception is unknown.

OBJECTIVE

To determine if NOL index change to a tetanic stimulus in cardiac and noncardiac surgery patients could predict the required remifentanil concentration for haemodynamic stability at skin incision.

DESIGN

A prospective two-phase cohort study.

SETTING

University hospital.

PATIENTS

Patients undergoing remifentanil-propofol target controlled infusion (TCI) anaesthesia.

INTERVENTIONS

During the calibration phase, investigators evaluated the tetanic stimulus induced NOL index change under standardised TCI remifentanil-propofol anaesthesia during a no-touch period [bispectral index (BIS) between 40 and 60, NOL index under 15]. If the NOL index change was 20 or greater following tetanic stimulation, investigators repeated the tetanus at higher remifentanil concentrations until the response was blunted. Surgeons incised the skin at this remifentanil concentration. The investigators derived a prediction model and in the validation phase calculated, using the NOL response to a single tetanus, the required incision remifentanil concentration for the start of surgery.

MAIN OUTCOME

Haemodynamic stability at incision [i.e. maximum heart rate (HR) < 20% increase from baseline, minimum HR (40 bpm) and mean arterial pressure (MAP) ± <20% of baseline].

RESULTS

During the calibration phase, no patient had hypertension. Two patients had a HR increase slightly greater than 20% (25.4 and 26.7%) within the first 2 min of surgery, but neither of these two patients had a HR above 76 bpm. Two patients were slightly hypotensive after incision (MAP 64 and 73 mmHg). During the validation phase, neither tachycardia nor hypotension occurred, but MAP increased to 21.5% above baseline for one patient.

CONCLUSION

During a no-touch period in patients under steady-state general anaesthesia [propofol effect site concentration (Ce) required for BIS between 40 and 60], the NOL index response to a tetanic stimulus under remifentanil antinociception can be used to personalise remifentanil Ce for the start of surgery and ensure stable haemodynamics.

TRIAL REGISTRATION

ClinicalTrials.gov: NCT03324269.

Increased risk of hip fracture mortality associated with intraoperative hypotension in elderly hip fracture patients is related to under resuscitation

BACKGROUND

As the US and world population ages, hip fractures are increasingly more common. The mortality associated with these fractures remains high both in the immediate postoperative period and at one year. Perioperative resuscitation in this population is of key interest to prevent organ injury and mortality. Our objectives were to evaluate the effect of fluid resuscitation and hemodynamic status in the form of mean arterial pressure (MAP) on inpatient mortality of hip fracture patients.

METHODS

An institutional database was queried to compare elderly hip fracture patients that sustained in-hospital mortality to a matched control cohort. Pre-, intra-, and post-operative intravenous fluid (IVF) administration and MAP were extracted from the electronic medical record. Time from hospital presentation to the OR was also recorded.

RESULTS

1,114 total hip fractures were identified during the two-year study period, 16 of which suffered inpatient mortalities. The mortality cohort was then matched with a control of 394 hip fracture patients for the same period based on age, sex, and Charlson Comorbidity Index (CCI). Conditional logistical regression analysis found odds ratios (OR) indicating that longer time between presentation and surgery (OR per additional hour: 1.05; 95% CI: 1.01-1.08) and lower intraoperative minimum MAP (OR per 5 mmHg decrease: 0.77; 95% CI: 0.61-0.97) were associated with significantly increased odds of mortality. There was also a marginal relationship between greater intraoperative IVF administration and reduced odds of mortality (OR per 500 cc additional fluid: 0.61; 95% CI: 0.37-1.00).

CONCLUSION

Extended time from presentation to surgery and intraoperative hypotension were associated with increased likelihood of inpatient mortality in an elderly hip fracture cohort, with a possible additional effect of under-resuscitation. Further investigation into a safe intraoperative minimum MAP should be pursued.

LEVEL OF EVIDENCE

Level III.

Collective Variational Inference for Personalized and Generative Physiological Modeling: A Case Study on Hemorrhage Resuscitation

OBJECTIVE

Individual physiological experiments typically provide useful but incomplete information about a studied physiological process. As a result, inferring the unknown parameters of a physiological model from experimental data is often challenging. The objective of this paper is to propose and illustrate the efficacy of a collective variational inference (C-VI) method, intended to reconcile low-information and heterogeneous data from a collection of experiments to produce robust personalized and generative physiological models.

METHODS

To derive the C-VI method, we utilize a probabilistic graphical model to impose structure on the available physiological data, and algorithmically characterize the graphical model using variational Bayesian inference techniques. To illustrate the efficacy of the C-VI method, we apply it to a case study on the mathematical modeling of hemorrhage resuscitation.

RESULTS

In the context of hemorrhage resuscitation modeling, the C-VI method could reconcile heterogeneous combinations of hematocrit, cardiac output, and blood pressure data across multiple experiments to obtain (i) robust personalized models along with associated measures of uncertainty and signal quality, and (ii) a generative model capable of reproducing the physiological behavior of the population.

CONCLUSION

The C-VI method facilitates the personalized and generative modeling of physiological processes in the presence of low-information and heterogeneous data.

SIGNIFICANCE

The resulting models provide a solid basis for the development and testing of interpretable physiological monitoring, decision-support, and closed-loop control algorithms.

Hemodynamic Changes via the Lung Recruitment Maneuver Can Predict Fluid Responsiveness in Stroke Volume and Arterial Pressure During One-Lung Ventilation

BACKGROUND

We aimed to evaluate the ability of lung recruitment maneuver-induced hemodynamic changes to predict fluid responsiveness in patients undergoing lung-protective ventilation during one-lung ventilation (OLV).

METHODS

Thirty patients undergoing thoracic surgery with OLV (tidal volume: 6 mL/kg of ideal body weight and positive end-expiratory pressure: 5 cm H2O) were enrolled. The study protocol began 30 minutes after starting OLV. Simultaneous recordings were performed for hemodynamic variables of heart rate, mean arterial pressure (MAP), stroke volume (SV), pulse pressure variation (PPV), and stroke volume variation (SVV) were recorded at 4 time points: before recruitment maneuver (continuous airway pressure: 30 cm H2O for 30 seconds), at the end of recruitment maneuver, and before and after volume loading (250 mL over 10 minutes). Patients were recognized as fluid responders if the increase in SV or MAP was >10%. Receiver operating characteristic curves for percent decrease in SV and MAP by recruitment maneuver (ΔSVRM and ΔMAPRM, respectively) were generated to evaluate the ability to discriminate fluid responders from nonresponders. The gray-zone approach was applied for ΔSVRM and ΔMAPRM.

RESULTS

Of 30 patients, there were 17 SV-responders (57%) and 12 blood pressure (BP)-responders (40%). Area under the curve (AUC) for ΔSVRM to discriminate SV-responders from nonresponders was 0.84 (95% confidence interval [CI], 0.67-0.95; P < .001). The best threshold for ΔSVRM to discriminate the SV-responders was -23.7% (95% CI, -41.2 to -17.8; sensitivity, 76.5% [95% CI, 50.1-93.2]; specificity, 84.6% [95% CI, 54.6-98.1]). For BP-responders, AUC for ΔMAPRM was 0.80 (95% CI, 0.61-0.92, P < .001). The best threshold for ΔMAPRM was -17.3% (95% CI, -23.9 to -5.1; sensitivity, 75.0% [95% CI, 42.8-94.5]; specificity, 77.8% [95% CI, 52.4-93.6]). With the gray-zone approach, the inconclusive range of ΔSVRM for SV-responders was -40.1% to -13.8% including 13 (43%) patients, and that of ΔMAPRM was -23.9% to -5.1%, which included 16 (53%) patients.

CONCLUSIONS

ΔSVRM and ΔMAPRM could predict hemodynamic responses after volume expansion during OLV.

Effect of dexmedetomidine on Nociception Level Index-guided remifentanil antinociception: A randomised controlled trial

BACKGROUND

The effect of dexmedetomidine on Nociception Level Index-guided (Medasense, Israel) antinociception to reduce intra-operative opioid requirements has not been previously investigated.

OBJECTIVE

We aimed to determine if low-dose dexmedetomidine would reduce remifentanil requirements during Nociception Level Index-guided antinociception without increasing complications associated with dexmedetomidine.

DESIGN

Double-blind randomised controlled trial.

SETTING

Two university teaching hospitals in Brussels, Belgium.

PATIENTS

American Society of Anesthesiologists 1 and 2 patients (n = 58) undergoing maxillofacial or cervicofacial surgery under propofol--remifentanil target-controlled infusion anaesthesia.

INTERVENTIONS

A 30 min infusion of dexmedetomidine, or equal volume of 0.9% NaCl, was infused at 1.2 μg kg-1 h-1 immediately preceding induction and then decreased to 0.6 μg kg-1 h-1 until 30 min before ending surgery. Nociception Level Index and frontal electroencephalogram guided the remifentanil and propofol infusions, respectively.

MAIN OUTCOMES

The primary outcome was the remifentanil requirement. Other outcomes included the propofol requirement, cardiovascular status and postoperative outcome.

RESULTS

Mean ± SD remifentanil (3.96 ± 1.95 vs. 4.42 ± 2.04 ng ml-1; P = 0.0024) and propofol (2.78 ± 1.36 vs. 3.06 ± 1.29 μg ml-1; P = 0.0046) TCI effect site concentrations were lower in the dexmedetomidine group at 30 min postincision and remained lower throughout surgery. When remifentanil (0.133 ± 0.085 vs. 0.198 ± 0.086 μg kg-1 min-1; P = 0.0074) and propofol (5.7 ± 2.72 vs. 7.4 ± 2.80 mg kg-1 h-1; P = 0.0228) requirements are represented as infusion rates, this effect became statistically significant at 2 h postincision.

CONCLUSION

In ASA 1 and 2 patients receiving Nociception Level Index-guided antinociception, dexmedetomidine decreases intra-operative remifentanil requirements. Combined frontal electroencephalogram and Nociception Level Index monitoring can measure dexmedetomidine's hypnotic and opioid-sparing effects during remifentanil-propofol target-controlled infusion anaesthesia.

TRIAL REGISTRATIONS

Clinicaltrials.gov: NCT03912740, EudraCT: 2018-004512-22.

Intraoperative hypotension during liver transplant surgery is associated with postoperative acute kidney injury: a historical cohort study

Background

Acute kidney injury (AKI) occurs frequently after liver transplant surgery and is associated with significant morbidity and mortality. While the impact of intraoperative hypotension (IOH) on postoperative AKI has been well demonstrated in patients undergoing a wide variety of non-cardiac surgeries, it remains poorly studied in liver transplant surgery. We tested the hypothesis that IOH is associated with AKI following liver transplant surgery.

Methods

This historical cohort study included all patients who underwent liver transplant surgery between 2014 and 2019 except those with a preoperative creatinine > 1.5 mg/dl and/or who had combined transplantation surgery. IOH was defined as any mean arterial pressure (MAP) < 65 mmHg and was classified according to the percentage of case time during which the MAP was < 65 mmHg into three groups, based on the interquartile range of the study cohort: “short” (Quartile 1, < 8.6% of case time), “intermediate” (Quartiles 2–3, 8.6–39.5%) and “long” (Quartile 4, > 39.5%) duration. AKI stages were classified according to a “modified” “Kidney Disease: Improving Global Outcomes” (KDIGO) criteria. Logistic regression modelling was conducted to assess the association between IOH and postoperative AKI. The model was run both as a univariate and with multiple perioperative covariates to test for robustness to confounders.

Results

Of the 205 patients who met our inclusion criteria, 117 (57.1%) developed AKI. Fifty-two (25%), 102 (50%) and 51 (25%) patients had short, intermediate and long duration of IOH respectively. In multivariate analysis, IOH was independently associated with an increased risk of AKI (adjusted odds ratio [OR] 1.05; 95%CI 1.02–1.09; P < 0.001). Compared to “short duration” of IOH, “intermediate duration” was associated with a 10-fold increased risk of developing AKI (OR 9.7; 95%CI 4.1–22.7; P < 0.001). “Long duration” was associated with an even greater risk of AKI compared to “short duration” (OR 34.6; 95%CI 11.5-108.6; P < 0.001).

Conclusions

Intraoperative hypotension is independently associated with the development of AKI after liver transplant surgery. The longer the MAP is < 65 mmHg, the higher the risk the patient will develop AKI in the immediate postoperative period, and the greater the likely severity. Anesthesiologists and surgeons must therefore make every effort to avoid IOH during surgery.

Continuous noninvasive monitoring of arterial pressure using the vascular unloading technique in comparison to the invasive gold standard in elderly comorbid patients: A prospective observational study

BACKGROUND AND AIMS

Elderly patients aged ≥65 years represent a growing population in the perioperative field, particularly orthopedic and vascular surgery. The higher degree of age-related or comorbid-dependent vascular alterations renders these patients at risk for hemodynamic complications and likely denote a possible limitation for modern, non-invasive arterial pressure monitoring devices. The aim was to compare vascular unloading technique-derived to invasive measurements of systolic (SAP), diastolic (DAP), and mean arterial pressure (MAP) in elderly perioperative patients.

METHODS

This prospective observational study included patients aged ≥65 years scheduled for orthopedic and patients ≥50 years with peripheral artery disease Fontaine stage ≥ II scheduled for vascular surgery, respectively. Invasive radial artery and non-invasive finger-cuff (Nexfin system) arterial pressures were recorded before and after induction of general anesthesia and during surgery. Correlation, Bland-Altman, and concordance analyses were performed. Measurements of arterial pressure were also compared during intraoperative hypotension (MAP <70 mm Hg) and hypertension (MAP >105 mm Hg).

RESULTS

Sixty patients with orthopedic (N = 25, mean (SD) age 77 (5) years) and vascular surgery (N = 35, age 69 [10] years) were enrolled. Seven hundred data pairs of all patients were analysed and pooled bias and percentage error were: SAP: 14.43 mm Hg, 43.79%; DAP: -2.40 mm Hg, 53.78% and MAP: 1.73 mm Hg, 45.05%. Concordance rates were 84.01% for SAP, 77.87% for DAP, and 86.47% for MAP. Predefined criteria for interchangeability of absolute and trending values could neither be reached in the overall nor in the subgroup analyses orthopedic vs vascular surgery. During hypertension, percentage error was found to be lowest for all pressure values, still not reaching predefined criteria.

CONCLUSION

Arterial pressure monitoring with the vascular unloading technique did not reach criteria of interchangeability for absolute and trending values. Nevertheless, the putatively beneficial use of noninvasive arterial pressure measurements should be further evaluated in the elderly perioperative patient.

[Guiding opioid-free intravenous antinociception with the Analgesia Nociception Index: a case report]

BACKGROUND

Opioid-free anesthesia decreases the incidence of opioid adverse events, but its optimal antinociceptive depth has not been clearly defined. Personalizing intraoperative opioid-free infusions with a nociception monitor may be the solution.

CASE REPORT

We describe the feasibility and potential limitations of titrating opioid-free antinociception during major abdominal surgery using the Analgesia Nociception Index (Mdoloris, Lille, France) in an obese patient. After stabilizing the patient's nociception-antinociception balance intraoperatively we quickly reversed anesthesia and the patient did not require postoperative opioids.

CONCLUSION

Personalizing opioid-free antinociception with a nociception monitor is feasible. It may optimize intraoperative antinociception and improve postoperative comfort.

Guiding opioid-free intravenous antinociception with the Analgesia Nociception Index: a case report

Background

Opioid-free anesthesia decreases the incidence of opioid adverse events, but its optimal antinociceptive depth has not been clearly defined. Personalizing intraoperative opioid-free infusions with a nociception monitor may be the solution.

Case report

We describe the feasibility and potential limitations of titrating opioid-free antinociception during major abdominal surgery using the Analgesia Nociception Index (Mdoloris, Lille, France) in an obese patient. After stabilizing the patient’s nociception-antinociception balance intraoperatively we quickly reversed anesthesia and the patient did not require postoperative opioids.

Conclusion

Personalizing opioid-free antinociception with a nociception monitor is feasible. It may optimize intraoperative antinociception and improve postoperative comfort.

Using Dynamic Variables to Guide Perioperative Fluid Management

Dynamic variables that quantify the variations in the arterial pressure and plethysmographic waveforms during mechanical ventilation reflect fluid responsiveness. These variables may be helpful in identifying occult hypovolemia and in preventing unnecessary fluid administration.