Norepinephrine weaning in septic shock patients by closed loop control based on fuzzy logic

Can Machine Learning Personalize Cardiovascular Therapy in Sepsis?

Large randomized trials in sepsis have generally failed to find effective novel treatments. This is increasingly attributed to patient heterogeneity, including heterogeneous cardiovascular changes in septic shock. We discuss the potential for machine learning systems to personalize cardiovascular resuscitation in sepsis. While the literature is replete with proofs of concept, the technological readiness of current systems is low, with a paucity of clinical trials and proven patient benefit. Systems may be vulnerable to confounding and poor generalization to new patient populations or contemporary patterns of care. Typical electronic health records do not capture rich enough data, at sufficient temporal resolution, to produce systems that make actionable treatment suggestions. To resolve these issues, we recommend a simultaneous focus on technical challenges and removing barriers to translation. This will involve improving data quality, adopting causally grounded models, prioritizing safety assessment and integration into healthcare workflows, conducting randomized clinical trials and aligning with regulatory requirements.

Performance of closed-loop systems for intravenous drug administration: a systematic review and meta-analysis of randomised controlled trials

Closed-loop drug delivery systems are autonomous computers able to administer medication in response to changes in physiological parameters (controlled variables). While limited evidence suggested that closed-loop systems can perform better than manual drug administration in certain settings, this technology remains a research tool with an uncertain risk/benefit profile. Our aim was comparing the performance of closed-loop systems with manual intravenous drug administration in adults. We searched MEDLINE, CENTRAL, and Embase from inception until November 2022, without restriction to language. We assessed for inclusion randomised controlled trials comparing closed-loop and manual administration of intravenous drugs in adults, intraoperatively or in the Intensive Care Unit. We identified 32 studies on closed-loop administration of propofol, noradrenaline, phenylephrine, insulin, neuromuscular blockers, and vasodilators. Most studies were at moderate or high risk of bias. The results showed that closed-loop systems reduced the duration of blood pressure outside prespecified targets during noradrenaline (MD 14.9%, 95% CI 9.6-20.2%, I2 = 66.6%) and vasodilators administration (MD 7.4%, 95% CI 5.2-9.7%, I2 = 62.3%). Closed-loop systems also decreased the duration of recovery after propofol (MD 1.3 min, 95% CI 0.4-2.1 min, I2 = 58.6%) and neuromuscular blockers (MD 9.0 min, 95% CI 7.9-10.0 min, I2 = 0%). The certainty of the evidence was low or very low for most outcomes. Automatic technology may be used to improve the hemodynamic profile during noradrenaline and vasodilators administration and reduce the duration of postanaesthetic recovery.Registration: This systematic review was registered with PROSPERO (CRD42022336950) on the 7th of June 2022.

Nurse management of noradrenaline infusions in intensive care units: An observational study

BACKGROUND

Intensive care nurse management of noradrenaline (norepinephrine) infusions is a common and essential clinical competency for patient haemodynamic support. Nurses titrate and wean noradrenaline infusions to a target blood pressure in a dynamic, high-risk, and unpredictable environment. Titration and weaning are complex interventions, and blood pressure goals are often variable.

OBJECTIVES

The aim was to examine how nurses used blood pressure targets when escalating, weaning, and titrating noradrenaline in intensive care patients admitted for haemodynamic management and explore patient blood pressure responses to changes in noradrenaline doses.

METHODS

In this naturalistic observational study, noradrenaline dose changes were classified as escalation, weaning, and titration changes and analysed to explore nursing practice. The study was undertaken in two adult medical/surgical intensive care units in Melbourne, Australia. Participants included intensive care nurses and patients who received noradrenaline infusions for haemodynamic support.

RESULTS

Observations of 14 nurse-patient dyads provided 25 h of blood pressure and noradrenaline dose data. Patient participants received weight-adjusted maximum noradrenaline doses of between 0.06 mcg/kg/min and 0.87 mcg/kg/minute, with those in the escalation group receiving dose increases of up to 5 mcg to achieve blood pressure goals. During weaning, patients maintained or increased their blood pressure as noradrenaline doses were decreased. Nurses consistently maintained blood pressures at higher than target goals, and despite constant fluctuations, they only documented blood pressure readings hourly.

CONCLUSIONS

Intensive care nurses managed noradrenaline to achieve mean arterial pressure targets that were variable and not evidence based. The disconnection between observed blood pressure fluctuations and nurse documentation of patient blood pressures was reflected in titration practices. Discrepancies between documented and actual blood pressures raised issues about data used by nurses and doctors to inform clinical practice on noradrenaline management.

Identification of module genes and functional pathway analysis in septic shock subtypes by integrated bioinformatics analysis

BACKGROUND

The present study aimed to identify the module genes and key gene functions and biological pathways of septic shock (SS) through integrated bioinformatics analysis.

METHODS

In the study, we performed batch correction and principal component analysis on 282 SS samples and 79 normal control samples in three datasets, GSE26440, GSE95233 and GSE57065, to obtain a combined corrected gene expression matrix containing 21,654 transcripts. Patients with SS were then divided into three molecular subtypes according to sample subtyping analysis.

RESULTS

By analyzing the demographic characteristics of the different subtypes, we found no statistically significant differences in gender ratio and age composition among the three groups. Then, three subtypes of differentially expressed genes (DEGs) and specific upregulated DEGs (SDEGs) were identified by differential gene expression analysis. We found 7361 DEGs in the type I group, 5594 DEGs in the type II group, and 7159 DEGs in the type III group. There were 1698 SDEGs in the type I group, 2443 in the type II group, and 1831 in the type III group. In addition, we analyzed the correlation between the expression data of 5972 SDEGs in the three subtypes and the gender and age of 227 patients, constructed a weighted gene co-expression network, and identified 11 gene modules, among which the module with the highest correlation with gender ratio was MEgrey. The modules with the highest correlation with age composition were MEgrey60 and MElightyellow. Then, by analyzing the differences in module genes among different subgroups of SS, we obtained the differential expression of 11 module genes in four groups: type I, type II, type III and the control group. Finally, we analyzed the Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment of all module DEGs, and the GO function and KEGG pathway enrichment of different module genes were different.

CONCLUSIONS

Our findings aim to identify the specific genes and intrinsic molecular functional pathways of SS subtypes, as well as further explore the genetic and molecular pathophysiological mechanisms of SS.

Closed-Loop Pharmacologic Control of Blood Pressure: A Review of Existing Systems

Blood pressure management is a critical aspect of patient care, particularly in surgical and critical care settings. Closed-loop systems, which utilize real-time data and feedback to adjust treatment interventions, have gained attention for their potential to enhance blood pressure control. This review explores the application of closed-loop systems in blood pressure management. We discuss various closed-loop approaches, including their mechanisms, benefits, and limitations. By harnessing real-time patient data and feedback, closed-loop systems can tailor interventions dynamically, thus enhancing blood pressure regulation. Additionally, we examine the integration of advanced monitoring technologies and artificial intelligence algorithms in closed-loop systems. The review highlights recent studies and their findings, emphasizing the evolving landscape of closed-loop blood pressure management across different clinical scenarios. From the perioperative period to critical care settings, closed-loop systems hold the potential to optimize patient outcomes by precisely adjusting vasopressor administration in response to continuous blood pressure fluctuations. By providing insights into the current state of closed-loop systems for blood pressure control, this review offers a comprehensive overview of their potential contributions to improved patient outcomes and future directions for research and implementation.

A system theory based digital model for predicting the cumulative fluid balance course in intensive care patients

Background: Surgical interventions can cause severe fluid imbalances in patients undergoing cardiac surgery, affecting length of hospital stay and survival. Therefore, appropriate management of daily fluid goals is a key element of postoperative intensive care in these patients. Because fluid balance is influenced by a complex interplay of patient-, surgery- and intensive care unit (ICU)-specific factors, fluid prediction is difficult and often inaccurate. Methods: A novel system theory based digital model for cumulative fluid balance (CFB) prediction is presented using recorded patient fluid data as the sole parameter source by applying the concept of a transfer function. Using a retrospective dataset of n = 618 cardiac intensive care patients, patient-individual models were created and evaluated. RMSE analyses and error calculations were performed for reasonable combinations of model estimation periods and clinically relevant prediction horizons for CFB. Results: Our models have shown that a clinically relevant time horizon for CFB prediction with the combination of 48 h estimation time and 8-16 h prediction time achieves high accuracy. With an 8-h prediction time, nearly 50% of CFB predictions are within ±0.5 L, and 77% are still within the clinically acceptable range of ±1.0 L. Conclusion: Our study has provided a promising proof of principle and may form the basis for further efforts in the development of computational models for fluid prediction that do not require large datasets for training and validation, as is the case with machine learning or AI-based models. The adaptive transfer function approach allows estimation of CFB course on a dynamically changing patient fluid balance system by simulating the response to the current fluid management regime, providing a useful digital tool for clinicians in daily intensive care.

Dynamic Arterial Elastance to Predict Mean Arterial Pressure Decrease after Reduction of Vasopressor in Septic Shock Patients

After fluid status optimization, norepinephrine infusion represents the cornerstone of septic shock treatment. De-escalation of vasopressors should be considered with caution, as hypotension increases the risk of mortality. In this prospective observational study including 42 patients, we assess the role of dynamic elastance (EaDyn), i.e., the ratio between pulse pressure variation and stroke volume variation, which can be measured noninvasively by the MostCare monitoring system, to predict a mean arterial pressure (MAP) drop > 10% 30 min after norepinephrine reduction. Patients were divided into responders (MAP falling > 10%) and non-responders (MAP falling < 10%). The receiver-operating-characteristic curve identified an area under the curve of the EaDyn value to predict a MAP decrease > 10% of 0.84. An EaDyn cut-off of 0.84 predicted a MAP drop > 10% with a sensitivity of 0.71 and a specificity of 0.89. In a multivariate logistic regression, EaDyn was significantly and independently associated with MAP decrease (OR 0.001, 95% confidence interval 0.00001−0.081, p < 0.001). The nomogram model for the probability of MAP decrease > 10% showed a C-index of 0.90. In conclusion, in a septic shock cohort, EaDyn correlates well with the risk of decrease of MAP > 10% after norepinephrine reduction.

Survey of Nurses' Experiences Applying The Joint Commission's Medication Management Titration Standards

BACKGROUND

Critical care nurses titrate continuous infusions of medications to achieve clinical end points. In 2017, The Joint Commission (TJC) placed restrictions on titration practice, decreasing nurses' autonomous decision-making.

OBJECTIVES

To describe the practice and perceptions of nurses regarding the 2017 TJC accreditation/regulatory standards for titration of continuous medication infusions.

METHODS

A survey of nurses' experiences titrating continuous medication infusions was developed, validated, and distributed electronically to members of the American Association of Critical-Care Nurses.

RESULTS

The content validity index for the survey was 1.0 for relevance and 0.95 for clarity. A total of 781 nurses completed the survey; 625 (80%) perceived titration standards to cause delays in patient care, and 726 (93%) experienced moral distress (mean [SD], 4.97 [2.67]; scale, 0-10). Among respondents, 33% could not comply with titration orders, 68% reported suboptimal care resulting from pressure to comply with orders, 70% deviated from orders to meet patient needs, and 84% requested revised orders to ensure compliance. Suboptimal care and delays in care significantly and strongly (regression coefficients ≥0.69) predicted moral distress.

CONCLUSIONS

Critical care nurses perceive TJC medication titration standards to adversely impact patient care and contribute to moral distress. The improved 2020 updates to the standards do not address delays and inability to comply with orders, leading to moral distress. Advocacy is indicated in order to mitigate unintended consequences of TJC medication management titration standards.

The order of vasopressor discontinuation and incidence of hypotension: a retrospective cohort analysis

The optimal order of vasopressor discontinuation during shock resolution remains unclear. We evaluated the incidence of hypotension in patients receiving concomitant vasopressin (VP) and norepinephrine (NE) based on the order of their discontinuation. In this retrospective cohort study, consecutive patients receiving concomitant VP and NE infusions for shock admitted to intensive care units were evaluated. The primary outcome was hypotension incidence following discontinuation of VP or NE (VP1 and NE1 groups, respectively). Secondary outcomes included the incidence of acute kidney injury (AKI) and arrhythmias. Subgroup analysis was conducted by examining outcomes based on the type of shock. Of the 2,035 included patients, 952 (46.8%) were VP1 and 1,083 (53.2%) were NE1. VP1 had a higher incidence of hypotension than NE1 (42.1% vs. 14.2%; P < 0.001), longer time to shock reversal (median: 2.5 vs. 2.2 days; P = .009), higher hospital [29% (278/952) vs. 24% (258/1083); P = .006], and 28-day mortality [37% (348/952) vs. 29% (317/1,083); P < 0.001] when compared with the NE1 group. There were no differences in ICU mortality, ICU and hospital length of stay, new-onset arrhythmia, or AKI incidence between the two groups. In subgroup analyses based on different types of shock, similar outcomes were observed. After adjustments, hypotension in the following 24 h and 28-day mortality were significantly higher in VP1 (Odds ratios (OR) 4.08(3.28, 5.07); p-value < .001 and 1.27(1.04, 1.55); p-value < .001, respectively). Besides, in a multivariable model, the need for renal replacement therapy (OR 1.68 (1.34, 2.12); p-value < .001) was significantly higher in VP1. Among patients with shock who received concomitant VP and NE, the VP1 group was associated with a higher incidence of hypotension in comparison with NE1. Future studies need to validate our findings and their impact on clinical outcomes.

Vasopressor-Sparing Strategies in Patients with Shock: A Scoping-Review and an Evidence-Based Strategy Proposition

Despite the abundant literature on vasopressor therapy, few studies have focused on vasopressor-sparing strategies in patients with shock. We performed a scoping-review of the published studies evaluating vasopressor-sparing strategies by analyzing the results from randomized controlled trials conducted in patients with shock, with a focus on vasopressor doses and/or duration reduction. We analyzed 143 studies, mainly performed in septic shock. Our analysis demonstrated that several pharmacological and non-pharmacological strategies are associated with a decrease in the duration of vasopressor therapy. These strategies are as follows: implementing a weaning strategy, vasopressin use, systemic glucocorticoid administration, beta-blockers, and normothermia. On the contrary, early goal directed therapies, including fluid therapy, oral vasopressors, vitamin C, and renal replacement therapy, are not associated with an increase in vasopressor-free days. Based on these results, we proposed an evidence-based vasopressor management strategy.

Adjusting mean arterial pressure alarms improves the time spent within blood pressure targets in patients with septic shock: A quasi-experimental study

BACKGROUND

Norepinephrine is the first-line vasoactive drug in septic shock. As underdosages or overdosages can be harmful for patients, it seems useful to maintain the mean arterial pressure (MAP) within preset bounds.

OBJECTIVES

We sought to assess whether adjusted MAP alarms could improve MAP control in patients with septic shock.

METHODS

We conducted a quasi-experimental before-and-after study. During two consecutive periods, data on MAP control were obtained from patients with septic shock (n = 50/period) treated with norepinephrine over more than 24 h. The norepinephrine administration protocol, including prescription of the MAP target range (e.g., 65-75 mmHg), was identical during the two periods. During the first period (control group), the preset alarms of the monitor were used (i.e., low and high systolic blood pressure alarms set at 90 and 160 mmHg, respectively). During the second period, adjusted MAP alarms were implemented, with upper and lower bounds corresponding to the prescribed MAP target range (MAP-Alarm group). The primary end point was the percentage of time outside the desired MAP target range during the first 24 h of norepinephrine infusion.

RESULTS

Baseline characteristics were not significantly different. The primary end point was significantly lower in the MAP-Alarm group than in the control group (25 ± 13% versus 51 ± 18%, respectively; p < 0.01). MAP was higher than the target 14 ± 11% of the time in the MAP-Alarm group versus 37 ± 17% in the control group (p < 0.01) and lower than the target 11 ± 9% of the time in the MAP-Alarm versus 21 ± 22% in the control group (p < 0.05). There was no significant difference between the two groups with regard to the dose of norepinephrine, duration of norepinephrine administration, and survival.

CONCLUSIONS

These results suggest that adjusting MAP alarms to the desired MAP target range could dramatically improve the percentage of time spent within MAP targets in patients with septic shock but does not reduce exposure to norepinephrine.

Physiology-Informed Real-Time Mean Arterial Blood Pressure Learning and Prediction for Septic Patients Receiving Norepinephrine

OBJECTIVE

Septic shock is a life-threatening manifestation of infection with a mortality of 20-50% [1]. A catecholamine vasopressor, norepinephrine (NE), is widely used to treat septic shock primarily by increasing blood pressure. For this reason, future blood pressure knowledge is invaluable for properly controlling NE infusion rates in septic patients. However, recent machine learning and data-driven methods often treat the physiological effects of NE as a black box. In this paper, a real-time, physiology-informed human mean arterial blood pressure model for septic shock patients undergoing NE infusion is studied.

METHODS

Our methods combine learning theory, adaptive filter theory, and physiology. We learn least mean square adaptive filters to predict three physiological parameters (heart rate, pulse pressure, and the product of total arterial compliance and arterial resistance) from previous data and previous NE infusion rate. These predictions are combined according to a physiology model to predict future mean arterial blood pressure.

RESULTS

Our model successfully forecasts mean arterial blood pressure on 30 septic patients from two databases. Specifically, we predict mean arterial blood pressure 3.33 minutes to 20 minutes into the future with a root mean square error from 3.56 mmHg to 6.22 mmHg. Additionally, we compare the computational cost of different models and discover a correlation between learned NE response models and a patient's SOFA score.

CONCLUSION

Our approach advances our capability to predict the effects of changing NE infusion rates in septic patients.

SIGNIFICANCE

More accurately predicted MAP can lessen clinicians' workload and reduce error in NE titration.

Automated Blood Pressure Control

Arterial pressure management is a crucial task in the operating room and intensive care unit. In high-risk surgical and in critically ill patients, sustained hypotension is managed with continuous infusion of vasopressor agents, which most commonly have direct α agonist activity like phenylephrine or norepinephrine. The current standard of care to guide vasopressor infusion is manual titration to an arterial pressure target range. This approach may be improved by using automated systems that titrate vasopressor infusions to maintain a target pressure. In this article, we review the evidence behind blood pressure management in the operating room and intensive care unit and discuss current and potential future applications of automated blood pressure control.

Reporting of Organ Support Outcomes in Septic Shock Randomized Controlled Trials: A Methodologic Review-The Sepsis Organ Support Study

OBJECTIVES

Many recent randomized controlled trials in the field of septic shock failed to demonstrate a benefit on mortality. Randomized controlled trials increasingly report organ support duration and organ support-free days as primary or secondary outcomes. We conducted a methodologic systematic review to assess how organ support outcomes were defined and reported in septic shock randomized controlled trials.

DATA SOURCES

MEDLINE via PubMed, Embase, Cochrane Central Register of Controlled Trials, and Web of Science.

STUDY SELECTION

We included randomized controlled trials published between January 2004 and March 2018 that involved septic shock adults and assessed organ support duration and/or organ support-free days for hemodynamic support, respiratory support, or renal replacement therapy.

DATA EXTRACTION

For each randomized controlled trial, we extracted the definitions of organ support duration and organ support-free days. We particularly evaluated how nonsurvivors were accounted for. Study authors were contacted to provide any missing information regarding these definitions.

DATA SYNTHESIS

We included 28 randomized controlled trials. Organ support duration and organ support-free days outcomes were reported in 17 and 15 randomized controlled trials, respectively, for hemodynamic support, 15 and 15 for respiratory support, and five and nine for renal replacement therapy. Nonsurvivors were included in the organ support duration calculation in 13 of 14 randomized controlled trials (93%) for hemodynamic support and nine of 10 (90%) for respiratory support. The organ support-free days definition for hemodynamic support, respiratory support, and renal replacement therapy was reported in six of 15 randomized controlled trials (40%), eight of 15 randomized controlled trials (53%), and six of nine randomized controlled trials (67%) reporting an organ support-free days outcome, respectively. Of these, one half assigned "0" to nonsurvivors, and the other half attributed one point per day alive free of organ support up to a predefined time point.

CONCLUSIONS

This study highlights the heterogeneity and infrequency of organ support duration/organ support-free days outcome reporting in septic shock trials. When reported, the definitions of these outcome measures and methods of calculation are also infrequently reported, in particular how nonsurvivors were accounted for, which may have an important impact on interpretation.

Clinical applications of artificial intelligence in sepsis: A narrative review

Many studies have been published on a variety of clinical applications of artificial intelligence (AI) for sepsis, while there is no overview of the literature. The aim of this review is to give an overview of the literature and thereby identify knowledge gaps and prioritize areas with high priority for further research. A literature search was conducted in PubMed from inception to February 2019. Search terms related to AI were combined with terms regarding sepsis. Articles were included when they reported an area under the receiver operator characteristics curve (AUROC) as outcome measure. Fifteen articles on diagnosis of sepsis with AI models were included. The best performing model reached an AUROC of 0.97. There were also seven articles on prognosis, predicting mortality over time with an AUROC of up to 0.895. Finally, there were three articles on assistance of treatment of sepsis, where the use of AI was associated with the lowest mortality rates. Of the articles, twenty-two were judged to be at high risk of bias or had major concerns regarding applicability. This was mostly because predictor variables in these models, such as blood pressure, were also part of the definition of sepsis, which led to overestimation of the performance. We conclude that AI models have great potential for improving early identification of patients who may benefit from administration of antibiotics. Current AI prediction models to diagnose sepsis are at major risks of bias when the diagnosis criteria are part of the predictor variables in the model. Furthermore, generalizability of these models is poor due to overfitting and a lack of standardized protocols for the construction and validation of the models. Until these problems have been resolved, a large gap remains between the creation of an AI algorithm and its implementation in clinical practice.

Reduced Rank Least Squares for Real-Time Short Term Estimation of Mean Arterial Blood Pressure in Septic Patients Receiving Norepinephrine

Norepinephrine (NE), an endogenous catecholamine, is a mainstay treatment for septic shock, which is a life-threatening manifestation of severe infection. NE counteracts the loss in blood pressure associated with septic shock. However, an NE infusion that is too low fails to counteract the blood pressure drop, and an NE infusion that is too high can cause a hypertensive crisis and heart attack. Ideally, the NE infusion rate should maintain a patient’s mean arterial blood pressure (MAP) above 65 mmHg. There are a few data-driven, quantitative models to predict the MAP, and incorporate NE effects. This paper presents a model, driven by intensive care unit (ICU) measurable data and known NE inputs, to predict the future MAP of an ICU patient. We derive a least square estimation model for MAP based on available ICU data, including heart period, NE infusion rate, and respiration wave. We learn the parameters of our model from initial patient data and then use this information to predict future MAP data. We assess our model with data from 12 septic patients. Our model successfully predicts and tracks MAP when the NE infusion rate changes. Specifically, we predict MAP 3 to 20 min in the future with the mean error of less than 4 to 7 mmHg over 12 patients. Conclusion: this new approach creates the potential to advance methods for predicting NE infusion rate in septic patients. Significance: successfully predicted patients’ MAP could reduce catastrophic human error and lessen clinicians’ workload.

Closed-loop hemodynamic management

As the operating room and intensive care settings become increasingly complex, the required vigilance practitioners must dedicate to a wide array of clinical systems has increased concordantly. The resulting shortage of available attention to these various clinical tasks creates a vacuum for the introduction of systems that can administer well-established goal-directed therapies without significant provider feedback. Recently, there has been an explosion of academic exploration into creating such automated systems, with a strong specific focus on hemodynamic control. Within this field, the largest focus has been on goal-directed fluid therapy as systems automating vasopressor administration have only recently become viable options. Our goal in this review article is to summarize the validity of the relevant goal-directed hemodynamic systems and explore the expanding role of automation within these systems.

Closed-loop vasopressor control: in-silico study of robustness against pharmacodynamic variability

Initial feasibility of a novel closed-loop controller created by our group for closed-loop control of vasopressor infusions has been previously described. In clinical practice, vasopressor potency may be affected by a variety of factors including other pharmacologic agents, organ dysfunction, and vasoplegic states. The purpose of this study was therefore to evaluate the effectiveness of our controller in the face of large variations in drug potency, where 'effective' was defined as convergence on target pressure over time. We hypothesized that the controller would remain effective in the face up to a tenfold variability in drug response. To perform the robustness study, our physiologic simulator was used to create randomized simulated septic patients. 250 simulated patients were managed by the closed-loop in each of 7 norepinephrine responsiveness conditions: 0.1 ×, 0.2 ×, 0.5 ×, 1 ×, 2 ×, 5 ×, and 10 × expected population response to drug dose. Controller performance was evaluated for each level of norepinephrine response using Varvel's criteria as well as time-out-of-target. Median performance error and median absolute performance error were less than 5% in all response levels. Wobble was below 3% and divergence remained negative (i.e. the controller tended to converge towards the target over time) in all norepinephrine response levels, but at the highest response level of 10 × the value approached zero, suggesting the controller may be approaching instability. Response levels of 0.1 × and 0.2 × exhibited significantly higher time-out-of-target in the lower ranges (p < 0.001) compared to the 1 × response level as the controller was slower to correct the initial hypotension. In this simulation study, the closed-loop vasopressor controller remained effective in simulated patients exhibiting 0.1 to 10 × the expected population drug response.

Impact du nombre d’interventions infirmières sur la durée d’administration des catécholamines chez les patients de réanimation : étude INTERVAL

Objectifs : Bien que quotidiennement géré par les infirmier(ière)s diplômé(e)s d’État (IDE) de réanimation, le sevrage des catécholamines a rarement été étudié. L’hypothèse de notre étude était que l’augmentation du nombre d’interventions menées sur la vitesse de perfusion des catécholamines permettait de raccourcir la durée d’administration de celles-ci.

Patients et méthodes : Il s’agissait d’une étude prospective observationnelle dans 21 réanimations. Tous les patients traités par vasopresseurs pendant la période d’étude ont été inclus. Le nombre d’interventions effectuées par l’IDE en charge du patient sur la vitesse d’administration des catécholamines était noté toutes les quatre heures. La posologie de catécholamines en cours ainsi que la pression artérielle moyenne (PAM) du patient étaient également relevées. Les facteurs influençant la rapidité du sevrage des catécholamines étaient également analysés.

Résultats : Nous avons inclus 226 patients dont l’âge moyen était de 65 ± 14 ans. La durée moyenne du traitement par catécholamines était de 71,6 (±81) heures. Le nombre d’interventions IDE par quatre heures était de 1,9 (±2,8) intervention. La durée du traitement par catécholamines était significativement moindre lorsque le nombre d’interventions augmentait (p = 0,04). L’existence d’un protocole spécifique à la gestion des catécholamines raccourcissait leur durée de sevrage. Un IGS2 élevé, une PAM élevée ainsi que la période diurne étaient associés à une augmentation du nombre des interventions IDE.

Conclusion : Le nombre d’interventions par les IDEs menées sur la perfusion de catécholamines était inversement lié à la durée du traitement par catécholamines. Des études complémentaires permettront d’évaluer l’impact de ces interventions sur le pronostic des patients.

A global perspective on vasoactive agents in shock

PURPOSE

We set out to summarize the current knowledge on vasoactive drugs and their use in the management of shock to inform physicians' practices.

METHODS

This is a narrative review by a multidisciplinary, multinational-from six continents-panel of experts including physicians, a pharmacist, trialists, and scientists.

RESULTS AND CONCLUSIONS

Vasoactive drugs are an essential part of shock management. Catecholamines are the most commonly used vasoactive agents in the intensive care unit, and among them norepinephrine is the first-line therapy in most clinical conditions. Inotropes are indicated when myocardial function is depressed and dobutamine remains the first-line therapy. Vasoactive drugs have a narrow therapeutic spectrum and expose the patients to potentially lethal complications. Thus, these agents require precise therapeutic targets, close monitoring with titration to the minimal efficacious dose and should be weaned as promptly as possible. Moreover, the use of vasoactive drugs in shock requires an individualized approach. Vasopressin and possibly angiotensin II may be useful owing to their norepinephrine-sparing effects.

Incidence of hypotension according to the discontinuation order of vasopressors in the management of septic shock: a prospective randomized trial (DOVSS)

BACKGROUND

Vasopressin (AVP) is commonly added to norepinephrine (NE) to reverse shock in patients with sepsis. However, there are no data to support the appropriate strategy of vasopressor tapering in patients on concomitant NE and AVP who are recovering from septic shock. Therefore, the objective of this study was to evaluate the incidence of hypotension while tapering vasopressors in patients on concomitant NE and AVP recovering from septic shock.

METHODS

Patients with septic shock receiving concomitant NE and AVP were randomly assigned to taper NE first (NE group) or AVP first (AVP group). The primary end point was the incidence of hypotension within one hour of tapering of the first vasopressor. We also evaluated the association between serum copeptin levels and the occurrence of hypotension.

RESULTS

The study was stopped early due to a significant difference in the incidence of hypotension after 38 and 40 patients were enrolled in the NE group and the AVP group, respectively. There were 26 patients (68.4%) in the NE group versus 9 patients (22.5%) in the AVP group who developed hypotension after tapering the first vasopressor (p < 0.001). There was a similar finding during the subsequent tapering of the second vasopressor (64.5% in the NE vs 25.0% in the AVP group, p = 0.020). Finally, NE tapering was significantly associated with hypotension during the study period (hazard ratio, 2.221; 95% confidence interval, 1.106-4.460; p = 0.025). The serum copeptin level was lower in patients in whom hypotension developed during tapering of AVP than it was in those without hypotension.

CONCLUSIONS

Tapering NE rather than AVP may be associated with a higher incidence of hypotension in patients recovering from septic shock who are on concomitant NE and AVP. However, further studies with larger sample sizes are required to better determine the appropriate strategy for vasopressor tapering.

TRIAL REGISTRATION

ClinicalTrials.gov, NCT01493102 . Registered on 15 December 2011.

Unexplained mortality differences between septic shock trials: a systematic analysis of population characteristics and control-group mortality rates

Purpose

Although the definition of septic shock has been standardized, some variation in mortality rates among clinical trials is expected. Insights into the sources of heterogeneity may influence the design and interpretation of septic shock studies. We set out to identify inclusion criteria and baseline characteristics associated with between-trial differences in control group mortality rates.

Methods

We conducted a systematic review of RCTs published between 2006 and 2018 that included patients with septic shock. The percentage of variance in control-group mortality attributable to study heterogeneity rather than chance was measured by I². The association between control-group mortality and population characteristics was estimated using linear mixed models and a recursive partitioning algorithm.

Results

Sixty-five septic shock RCTs were included. Overall control-group mortality was 38.6%, with significant heterogeneity (I² = 93%, P < 0.0001) and a 95% prediction interval of 13.5–71.7%. The mean mortality rate did not differ between trials with different definitions of hypotension, infection or vasopressor or mechanical ventilation inclusion criteria. Population characteristics univariately associated with mortality rates were mean Sequential Organ Failure Assessment score (standardized regression coefficient (β) = 0.57, P = 0.007), mean serum creatinine (β = 0.48, P = 0.007), the proportion of patients on mechanical ventilation (β = 0.61, P < 0.001), and the proportion with vasopressors (β = 0.57, P = 0.002). Combinations of population characteristics selected with a linear model and recursive partitioning explained 41 and 42%, respectively, of the heterogeneity in mortality rates.

Conclusions

Among 65 septic shock trials, there was a clinically relevant amount of heterogeneity in control group mortality rates which was explained only partly by differences in inclusion criteria and reported baseline characteristics.

Electronic supplementary material

The online version of this article (10.1007/s00134-018-5134-8) contains supplementary material, which is available to authorized users.

Computer-controlled closed-loop drug infusion system for automated hemodynamic resuscitation in endotoxin-induced shock

Background

Hemodynamic resuscitation in septic shock requires aggressive fluid replacement and appropriate use of vasopressors to optimize arterial pressure (AP) and cardiac output (CO). Because responses to these drugs vary between patients and within patient over time, strict monitoring of patient condition and repetitive adjustment of drug dose are required. This task is time and labor consuming, and is associated with poor adherence to resuscitation guidelines. To overcome this issue, we developed a computer-controlled closed-loop drug infusion system for automated hemodynamic resuscitation in septic shock, and evaluated the performance of the system in a canine model of endotoxin shock.

Methods

Our system monitors AP, CO and central venous pressure, and computes arterial resistance (R), stressed blood volume (V) and Frank-Starling slope of left ventricle (S). The system controls R with noradrenaline (NA), and V with Ringer’s acetate solution (RiA), thereby controlling AP and CO. In 4 dogs, AP and CO were measured invasively. In another 4 dogs, AP and CO were measured less invasively using clinically acceptable modalities, aiming to make the system clinically feasible. In all 8 dogs, endotoxin shock was induced by injecting Escherichia coli lipopolysaccharide, which significantly decreased AP from 95 (91–108) to 43 (39–45) mmHg, and CO from 112 (104–142) to 62 (51–73) ml·min⁻¹·kg⁻¹. The system was then connected to the dogs, and activated. System performance was observed over a period of 4 h.

Results

Our system immediately started infusions of NA and RiA. Within 40 min, RiA increased V to target level, and NA maintained R at target level, while S was concomitantly increased. These resulted in restoration of AP to 70 (69–71) mmHg and CO to 130 (125–138) ml·min⁻¹·kg⁻¹. Median of absolute performance error, an index of precision of control, was small in AP [2.5 (2.1–4.5) %] and CO [2.4 (1.4–5.5) %], which were not increased even when the variables were measured less invasively.

Conclusions

In a canine model of endotoxin shock, our system automatically improved and maintained AP and CO at their target values with small performance error. Our system is potentially an attractive clinical tool for rescuing patients with septic shock.

Electronic supplementary material

The online version of this article (10.1186/s12871-017-0437-9) contains supplementary material, which is available to authorized users.

Physician-Directed Versus Computerized Closed-Loop Control of Blood Pressure Using Phenylephrine in a Swine Model

BACKGROUND

Vasopressors provide a rapid and effective approach to correct hypotension in the perioperative setting. Our group developed a closed-loop control (CLC) system that titrates phenylephrine (PHP) based on the mean arterial pressure (MAP) during general anesthesia. As a means of evaluating system competence, we compared the performance of the automated CLC with physicians. We hypothesized that our CLC algorithm more effectively maintains blood pressure at a specified target with less blood pressure variability and reduces the dose of PHP required.

METHODS

In a crossover study design, 6 swine under general anesthesia were subjected to a normovolemic hypotensive challenge induced by sodium nitroprusside. The physicians (MD) manually changed the PHP infusion rate, and the CLC system performed this task autonomously, adjusted every 3 seconds to achieve a predetermined MAP.

RESULTS

The CLC maintained MAP within 5 mm Hg of the target for (mean ± standard deviation) 93.5% ± 3.9% of the time versus 72.4% ± 26.8% for the MD treatment (P = .054). The mean (standard deviation) percentage of time that the CLC and MD interventions were above target range was 2.1% ± 3.3% and 25.8% ± 27.4% (P = .06), respectively. Control statistics, performance error, median performance error, and median absolute performance error were not different between CLC and MD interventions. PHP infusion rate adjustments by the physician were performed 12 to 80 times in individual studies over a 60-minute period. The total dose of PHP used was not different between the 2 interventions.

CONCLUSIONS

The CLC system performed as well as an anesthesiologist totally focused on MAP control by infusing PHP. Computerized CLC infusion of PHP provided tight blood pressure control under conditions of experimental vasodilation.

The development of a novel knowledge-based weaning algorithm using pulmonary parameters: a simulation study

Weaning is important for patients and clinicians who have to determine correct weaning time so that patients do not become addicted to the ventilator. There are already some predictors developed, such as the rapid shallow breathing index (RSBI), the pressure time index (PTI), and Jabour weaning index. Many important dimensions of weaning are sometimes ignored by these predictors. This is an attempt to develop a knowledge-based weaning process via fuzzy logic that eliminates the disadvantages of the present predictors. Sixteen vital parameters listed in published literature have been used to determine the weaning decisions in the developed system. Since there are considered to be too many individual parameters in it, related parameters were grouped together to determine acid-base balance, adequate oxygenation, adequate pulmonary function, hemodynamic stability, and the psychological status of the patients. To test the performance of the developed algorithm, 20 clinical scenarios were generated using Monte Carlo simulations and the Gaussian distribution method. The developed knowledge-based algorithm and RSBI predictor were applied to the generated scenarios. Finally, a clinician evaluated each clinical scenario independently. The Student's t test was used to show the statistical differences between the developed weaning algorithm, RSBI, and the clinician's evaluation. According to the results obtained, there were no statistical differences between the proposed methods and the clinician evaluations.

Multi-complexity measures of heart rate variability and the effect of vasopressor titration: a prospective cohort study of patients with septic shock

Background

Septic shock is a common and often devastating syndrome marked by severe cardiovascular dysfunction commonly managed with vasopressors. Whether markers of heart rate complexity before vasopressor up-titration could be used to predict success of the up-titration is not known.

Methods

We studied patients with septic shock requiring vasopressor, newly admitted to the intensive care unit. We measured the complexity of heart rate variability (using the ratio of fractal exponents from detrended fluctuation analysis) in the 5 min before all vasopressor up-titrations in the first 24 h of an intensive care unit (ICU) admission. A successful up-titration was defined as one that did not require further up-titration (or decrease in mean arterial pressure) for 60 min.

Results

We studied 95 patients with septic shock, with a median APACHE II of 27 (IQR: 20–37). The median number of up-titrations, normalized to 24 h, was 12.2 (IQR: 8–17) with a maximum of 49. Of the up-titrations, the median proportion of successful interventions was 0.28 (IQR: 0.12–0.42). The median of mean arterial pressure (MAP) at the time of a vasopressor up-titration was 66 mmHg; the average infusion rate of norepinephrine at the time of an up-titration was 0.11 mcg/kg/min. The ratio of fractal exponents was not associated with successful up-titration on univariate or multivariate regression. On exploratory secondary analyses, however, the long-term fractal exponent was associated (p = 0.003) with success of up-titration. Independent of heart rate variability, MAP was associated (p < 0.001) with success of vasopressor up-titration, while neither Sequential Organ Failure Assessment (SOFA) nor Acute Physiology and Chronic Health Evaluation II (APACHE II) score was associated with vasopressor titration.

Conclusions

Only a third of vasopressor up-titrations were successful among patients with septic shock. MAP and the long-term fractal exponent were associated with success of up-titration. These two, complementary variables may be important to the development of rational vasopressor titration protocols.

Electronic supplementary material

The online version of this article (doi:10.1186/s12879-016-1896-1) contains supplementary material, which is available to authorized users.

Clinical management for patients admitted to a critical care unit with severe sepsis or septic shock

BACKGROUND

The Surviving Sepsis Campaign promotes the use of norepinephrine as the first-line inotropic support for patients presenting with severe sepsis or septic shock in cases of persistent hypotension, despite adequate fluid resuscitation. However, there is little published evidence on how much noradrenaline is administered to such patients when admitted to the intensive care unit (ICU). The authors report the clinical management of this group of patients, with a special focus on the total amount and duration of norepinephrine infusion required.

METHODS

A chart review of the admission records of an ICU in Hong Kong was carried out in 2013. A total of 5000 patients were screened by their diagnosis of severe sepsis or septic shock (in the admissions book) between 1 January 2011 and 31 December 2013. A total of 150 of these were identified and 100 included in the study after simultaneous in-depth reviews of their case notes by two of the investigators. The analysis covers those with severe sepsis or septic shock who required ICU admission for further care. Clinical management and outcomes were analysed.

RESULTS

100 patients (median age 61.6; M/F ratio 2:1) met the inclusion criteria. The mean ICU stay was 13.4 days (range=1-371). 14 patients (14%) died in the ICU, with a 28-day mortality rate of 22%. The mean period of mechanical ventilation was 6.1 days (range=0-137). 91.5% (n=43) of patients had been operated on immediately before admission to the ICU, and the majority of these operations had been of the emergency type (97.7%, n=43). The mean total volumes of crystalloid and colloid administered were 3420ml and 478ml, respectively. The mean wean-off period for norepinephrine infusion was 4234minutes (70.5hours). All patients were prescribed norepinephrine for persistent hypotension despite adequate fluid resuscitation, and the mean total amount administered was 87,211mg. Final multiple linear and logistic regression analysis showed different clinical outcomes associated with different covariates, which included: (1) total amount of crystalloid given, positively associated with the total amount and duration of norepinephrine infusion; (2) duration of mechanical ventilation, positively associated with the type of operation the patient had undergone; (3) 28-day mortality rate, positively associated with the INR.

CONCLUSIONS

What this study adds to knowledge about patients suffering from severe sepsis or septic shock: (1) the mean duration of norepinephrine infusion for septic shock patients in an ICU is almost three days; (2) the more crystalloid is required to correct hypoperfusion, the higher the dosage and longer the duration of norepinephrine infusion will be necessary; (3) the longer the patient's INR, the higher the chances of death within 28 days. Since not all patients have their body weight measured on or after admission to the ICU, we suggest further research into indirect estimation of body weight by other means, such as anthropometric measures, to guide the use of drugs and nutritional support in the ICU. In addition, APACHE scores should be included in further studies to compare the severity of the patient's condition in other research. Furthermore, since this study does not cover university hospital ICUs, we suggest that further research concerning such patients should compare and reflect similarities and differences between public and university hospitals in the territory.

Dynamic arterial elastance predicts mean arterial pressure decrease associated with decreasing norepinephrine dosage in septic shock

INTRODUCTION

Gradual reduction of the dosage of norepinephrine (NE) in patients with septic shock is usually left to the physician's discretion. No hemodynamic indicator predictive of the possibility of decreasing the NE dosage is currently available at the bedside. The respiratory pulse pressure variation/respiratory stroke volume variation (dynamic arterial elastance (Eadyn)) ratio has been proposed as an indicator of vascular tone. The purpose of this study was to determine whether Eadyn can be used to predict the decrease in arterial pressure when decreasing the NE dosage in resuscitated sepsis patients.

METHODS

A prospective study was carried out in a university hospital intensive care unit. All consecutive patients with septic shock monitored by PICCO2 for whom the intensive care physician planned to decrease the NE dosage were enrolled. Measurements of hemodynamic and PICCO2 variables were obtained before/after decreasing the NE dosage. Responders were defined by a >15% decrease in mean arterial pressure (MAP).

RESULTS

In total, 35 patients were included. MAP decreased by >15% after decreasing the NE dosage in 37% of patients (n = 13). Clinical characteristics appeared to be similar between responders and nonresponders. Eadyn was lower in responders than in nonresponders (0.75 (0.69 to 0.85) versus 1 (0. 83 to 1.22), P <0.05). Baseline Eadyn was correlated with NE-induced MAP variations (r = 0.47, P = 0.005). An Eadyn less than 0.94 predicted a decrease in arterial pressure, with an area under the receiver-operating characteristic curve of 0.87 (95% confidence interval (95% CI): 0.72 to 0.96; P <0.0001), 100% sensitivity, and 68% specificity.

CONCLUSIONS

In sepsis patients treated with NE, Eadyn may predict the decrease in arterial pressure in response to NE dose reduction. Eadyn may constitute an easy-to-use functional approach to arterial-tone assessment, which may be helpful to identify patients likely to benefit from NE dose reduction.

Principles of pharmacologic hemodynamic management and closed-loop systems

Every day, physicians in critical-care settings are challenged with the hemodynamic management of patients with severe cardiovascular derangements. There is a potential role for closed-loop (automated) systems to assist clinicians in managing these patients and growing interest in the possible applications. In this review, we discuss the basic principles of critical-care hemodynamic management and the closed-loop systems that have been developed to help in this setting.

Vasopressor weaning in patients with septic shock

The purpose of this article is to propose optimal weaning of vasopressors in patients with septic shock. Topics discussed include pathophysiology of sepsis and septic shock, treatment guidelines for sepsis, autoregulation of blood flow, vasopressors used in septic shock, weaning recommendations, monitor alarms in the intensive care unit, and new directions in sepsis research.

Use of heptaminol hydrochloride for catecholamine weaning in septic shock

We analyze in the current study the impact of heptaminol hydrochloride (Heptamyl) administration in patients with septic shock requiring adrenergic support on the duration of vasopressor infusion and on catecholamine delay weaning. In this prospective study were included 49 nonconsecutive patients with septic shock requiring vasopressor infusion and with stable hemodynamic parameters during more than 24 hours. All these patients were included in a random way to receive or not heptaminol hydrochloride. The primary end point was the effect of heptaminol hydrochloride administration on duration of weaning, defined as cessation of vasopressor support. There were 32 males (65%) and 17 females (35%). The mean age (± standard deviation) was 53.9 ± 22.2 years. Norepinephrine was the most commonly used vasopressor agent (73.4%). The comparison between two groups (with and without heptaminol hydrochloride) showed that two groups had the same epidemiologic, clinical, and biologic findings on intensive care unit admission. In our study, we found that the introduction of Heptamyl was associated with a quick decrease of dose of dopamine and norepinephrine in comparison with the Heptamyl-free group. By comparing the two groups, we found that the delay of catecholamine weaning was significantly faster for the dopamine (P = 0.008) and noradrenalin (P = 0.001) in the Heptamyl group. Finally, the intensive care unit mortality rate and the hospital mortality rate were significantly lower in the Heptamyl group. Our study shows a reduction in norepinephrine and dopamine weaning duration in septic patients enrolled in the heptaminol hydrochloride group.

Essential elements of the native glucoregulatory system, which, if appreciated, may help improve the function of glucose controllers in the intensive care unit setting

In 2001, Van den Berghe and colleagues were able to show that tight glucose control decreases morbidity and mortality rates in the intensive care unit (ICU) setting. Several large, prospective, randomized controlled trials have failed to confirm these results. All of these studies attempted tight glucose control using expert-designed algorithms to adjust the rate of intravenous insulin. Unfortunately, these studies each had high rates of hypoglycemia, a high percentage of glucose values outside of the target range, and increased glucose variability. These three measurements have been shown to increase mortality rates in ICU patients. In order to achieve a high rate of success with regards to tight glucose control, a closed-loop system will need to be created. The two main elements of such a system are a continuous glucose sensor and a recursive glucose control algorithm. This review highlights the important elements of the native glucoregulatory system, which, if utilized, may help create a successful glucose control algorithm for a closed-loop system.

Year in review 2008: Critical Care--cardiology

We review key research papers in cardiology and intensive care published during 2008 in Critical Care. We quote studies on the same subject published in other journals if appropriate. Papers have been grouped into three categories: (a) cardiovascular biomarkers in critical illness, (b) haemodynamic management of septic shock, and (c) haemodynamic monitoring.

Can fuzzy logic make things more clear?

Intensive care is a complex environment involving many signals, data and observations. Clinical decision support and artificial intelligence using fuzzy logic and closed loop techniques are methods that might help us to handle this complexity in a safe, effective and efficient way. Merouani and colleagues have performed a study using fuzzy logic and closed loop techniques to more effectively wean patients with sepsis from norepinephrine infusion.