Real-world outcomes of the hypotension prediction index in the management of intraoperative hypotension during non-cardiac surgery: a retrospective clinical study

Evaluation of hypotension prediction index software in patients undergoing orthotopic liver transplantation: retrospective observational study

BACKGROUND

Extreme hemodynamic changes, especially intraoperative hypotension (IOH), are common and often prolonged during Liver Transplant (LT) surgery and during initial hours of recovery. Hypotension Prediction Index (HPI) software is one of the tools which can help in proactive hemodynamic management. The accuracy of the advanced hemodynamic parameters such as Cardiac Output (CO) and Systemic Vascular Resistance (SVR) obtained from HPI software and prediction performance of the HPI in LT surgery remains unknown.

METHODS

This was a retrospective observational study conducted in a tertiary academic center with a large liver transplant program. We enrolled 23 adult LT patients who received both Pulmonary Artery Catheter (PAC) and HPI software monitoring. Primarily, we evaluated agreement between PAC and HPI software measured CO and SVR. A priori, we defined a relative difference of less than 20% between measurements as an adequate agreement for a pair of measurements and estimated the Lin's Concordance Correlation Coefficient and Bland-Altman Limits of Agreement (LOA). Clinically acceptable LOA was defined as ± 1 L.min-1 for CO and ± 200 dynes s.cm-5 for SVR. Secondary outcome was the ability of the HPI to predict future hypotension, defined as Mean Arterial Pressure (MAP) less than 65 mmHg lasting at least one minute. We estimated sensitivity, positive predictive value, and time from alert to hypotensive events for HPI software.

RESULTS

Overall, 125 pairs of CO and 122 pairs of SVR records were obtained from 23 patients. Based on our predefined criteria, only 42% (95% CI 30%, 55%) of CO records and 53% (95% CI 28%, 72%) of SVR records from HPI software were considered to agree with those from PAC. Across all patients, there were a total of 1860 HPI alerts (HPI ≥ 85) and 642 hypotensive events (MAP < 65 mmHg). Out of the 642 hypotensive events, 618 events were predicted by HPI alert with sensitivity of 0.96 (95% CI: 0.95). Many times, the HPI value remained above alert level and was followed by multiple hypotensive events. Thus, to evaluate PPV and time to hypotension metric, we considered only the first HPI alert followed by a hypotensive event ("true alerts"). The "true alert" was the first alert when there were several alerts before a hypotension. There were 614 "true alerts" and the PPV for HPI was 0.33 (95% CI 0.31, 0.35). The median time from HPI alert to hypotension was 3.3 [Q1, Q3: 1, 9.3] mins.

CONCLUSION

There was poor agreement between the pulmonary artery catheter and HPI software calculated advanced hemodynamic parameters (CO and SVR), in the patients undergoing LT surgery. HPI software had high sensitivity but poor specificity for hypotension prediction, resulting in a high burden of false alarms.

The hypotension prediction index in major abdominal surgery - A prospective randomised clinical trial protocol

Background

Patients undergoing major abdominal surgery are at increased risk of developing perioperative hypotension, which is associated with increased mortality and morbidity. Despite using advanced technologies such as evaluating arterial pressure derived cardiac output, anaesthetic management to maintain hemodynamic stability is still reactive when the clinical decision is made after hypotension has developed. Previous perioperative goal-directed studies have not proven the benefits of this approach with high certainty. A new, approved technology called the Hypotension Prediction Index (HPI) aims to prevent hypotension occurrence by allowing the precise hemodynamic monitoring of patients under general anaesthesia, significantly reducing intraoperative hypotension events. This prospective randomised clinical trial aims to compare the rate of perioperative hypotension in patients undergoing major abdominal surgery according to their type of hemodynamic monitoring.

Methods

and Analysis: Patients meeting the inclusion criteria will be randomly assigned to receive hemodynamic assessment with arterial pressure cardiac output (APCO) monitoring (group A) or hemodynamic monitoring with the HPI software (group B). The primary outcome is a time-weighted average (TWA) mean arterial pressure (MAP) of <65 mmHg: TWA MAP = (depth of hypotension [in mmHg] below a MAP of 65 mmHg × time [in minutes] spent below a MAP of 65 mmHg)/total duration of the operation (in minutes). Its secondary outcomes include perioperative hemodynamic management and the rate of postoperative complications.

Ethics and dissemination

This trial was approved by the Ethics Committee of the Poznan University of Medical Sciences (KB-559/220; date: 01/07/2022). Its results will be submitted for publication in a peer-reviewed journal.

Trial registration number

NCT06247384.

Beyond the debut: unpacking six years of Hypotension Prediction Index software in intraoperative hypotension prevention - a systematic review and meta-analysis

PURPOSE

Intraoperative hypotension (IOH) during general anesthesia is associated with higher morbidity and mortality, although randomized trials have not established a causal relation. Historically, our approach to IOH has been reactive. The Hypotension Prediction Index (HPI) is a machine learning software that predicts hypotension minutes in advance. This systematic review and meta-analysis explores whether using HPI alongside a personalized treatment protocol decreases intraoperative hypotension.

METHODS

A systematic search was performed in Pubmed and Scopus to retrieve articles published from January 2018 to February 2024 regarding the impact of the HPI software on reducing IOH in adult patients undergoing non-cardio/thoracic surgery. Excluded were case series, case reports, meta-analyses, systematic reviews, and studies using non-invasive arterial waveform analysis. The risk of bias was assessed by the Cochrane risk-of-bias tool (RoB 2) and the Risk Of Bias In Non-randomised Studies (ROBINS-I). A meta-analysis was undertaken solely for outcomes where sufficient data were available from the included studies.

RESULTS

9 RCTs and 5 cohort studies were retrieved. The overall median differences between the HPI-guided and the control groups were - 0.21 (95% CI:-0.33, -0.09) - p < 0.001 for the Time-Weighted Average (TWA) of Mean Arterial Pressure (MAP) < 65mmHg, -3.71 (95% CI= -6.67, -0.74)-p = 0.014 for the incidence of hypotensive episodes per patient, and - 10.11 (95% CI= -15.82, -4.40)-p = 0.001 for the duration of hypotension. Notably a large amount of heterogeneity was detected among the studies.

CONCLUSIONS

While the combination of HPI software with personalized treatment protocols may prevent intraoperative hypotension (IOH), the large heterogeneity among the studies and the lack of reliable data on its clinical significance necessitate further investigation.

Comparison of Differences in Cohort (Forward) and Case Control (Backward) Methodologic Approaches for Validation of the Hypotension Prediction Index

BACKGROUND

The Hypotension Prediction Index (the index) software is a machine learning algorithm that detects physiologic changes that may lead to hypotension. The original validation used a case control (backward) analysis that has been suggested to be biased. This study therefore conducted a cohort (forward) analysis and compared this to the original validation technique.

METHODS

A retrospective analysis of data from previously reported studies was conducted. All data were analyzed identically with two different methodologies, and receiver operating characteristic curves were constructed. Both backward and forward analyses were performed to examine differences in area under the receiver operating characteristic curves for the Hypotension Prediction Index and other hemodynamic variables to predict a mean arterial pressure (MAP) less than 65 mmHg for at least 1 min 5, 10, and 15 min in advance.

RESULTS

The analysis included 2,022 patients, yielding 4,152,124 measurements taken at 20-s intervals. The area under the curve for the index predicting hypotension analyzed by backward and forward methodologies respectively was 0.957 (95% CI, 0.947 to 0.964) versus 0.923 (95% CI, 0.912 to 0.933) 5 min in advance, 0.933 (95% CI, 0.924 to 0.942) versus 0.923 (95% CI, 0.911 to 0.933) 10 min in advance, and 0.929 (95% CI, 0.918 to 0.938) versus 0.926 (95% CI, 0.914 to 0.937) 15 min in advance. No variable other than MAP had an area under the curve greater than 0.7. The areas under the curve using forward analysis for MAP predicting hypotension 5, 10, and 15 min in advance were 0.932 (95% CI, 0.920 to 0.940), 0.929 (95% CI, 0.918 to 0.938), and 0.932 (95% CI, 0.921 to 0.940), respectively. The R2 for the variation in the index due to MAP was 0.77.

CONCLUSIONS

Using an updated methodology, the study found that the utility of the Hypotension Prediction Index to predict future hypotensive events is high, with an area under the receiver operating characteristics curve similar to that of the original validation method.

EDITOR’S PERSPECTIVE

Predictive ability of hypotension prediction index and machine learning methods in intraoperative hypotension: a systematic review and meta-analysis

INTRODUCTION

Intraoperative Hypotension (IOH) poses a substantial risk during surgical procedures. The integration of Artificial Intelligence (AI) in predicting IOH holds promise for enhancing detection capabilities, providing an opportunity to improve patient outcomes. This systematic review and meta analysis explores the intersection of AI and IOH prediction, addressing the crucial need for effective monitoring in surgical settings.

METHOD

A search of Pubmed, Scopus, Web of Science, and Embase was conducted. Screening involved two-phase assessments by independent reviewers, ensuring adherence to predefined PICOS criteria. Included studies focused on AI models predicting IOH in any type of surgery. Due to the high number of studies evaluating the hypotension prediction index (HPI), we conducted two sets of meta-analyses: one involving the HPI studies and one including non-HPI studies. In the HPI studies the following outcomes were analyzed: cumulative duration of IOH per patient, time weighted average of mean arterial pressure < 65 (TWA-MAP < 65), area under the threshold of mean arterial pressure (AUT-MAP), and area under the receiver operating characteristics curve (AUROC). In the non-HPI studies, we examined the pooled AUROC of all AI models other than HPI.

RESULTS

43 studies were included in this review. Studies showed significant reduction in IOH duration, TWA-MAP < 65 mmHg, and AUT-MAP < 65 mmHg in groups where HPI was used. AUROC for HPI algorithms demonstrated strong predictive performance (AUROC = 0.89, 95CI). Non-HPI models had a pooled AUROC of 0.79 (95CI: 0.74, 0.83).

CONCLUSION

HPI demonstrated excellent ability to predict hypotensive episodes and hence reduce the duration of hypotension. Other AI models, particularly those based on deep learning methods, also indicated a great ability to predict IOH, while their capacity to reduce IOH-related indices such as duration remains unclear.

Impact of clinicians' behavior, an educational intervention with mandated blood pressure and the hypotension prediction index software on intraoperative hypotension: a mixed methods study

PURPOSE

Intraoperative hypotension (IOH) is associated with adverse outcomes. We therefore explored beliefs regarding IOH and barriers to its treatment. Secondarily, we assessed if an educational intervention and mandated mean arterial pressure (MAP), or the implementation of the Hypotension Prediction Index-software (HPI) were associated with a reduction in IOH.

METHODS

Structured interviews (n = 27) and questionnaires (n = 84) were conducted to explore clinicians' beliefs and barriers to IOH treatment, in addition to usefulness of HPI questionnaires (n = 14). 150 elective major surgical patients who required invasive blood pressure monitoring were included in three cohorts to assess incidence and time-weighted average (TWA) of hypotension (MAP < 65 mmHg). Cohort one received standard care (baseline), the clinicians of cohort two had a training on hypotension and a mandated MAP > 65 mmHg, and patients of the third cohort received protocolized care using the HPI.

RESULTS

Clinicians felt challenged to manage IOH in some patients, yet they reported sufficient knowledge and skills. HPI-software was considered useful and beneficial. No difference was found in incidence of IOH between cohorts. TWA was comparable between baseline and education cohort (0.15 mmHg [0.05-0.41] vs. 0.11 mmHg [0.02-0.37]), but was significantly lower in the HPI cohort (0.04 mmHg [0.00 to 0.11], p < 0.05 compared to both).

CONCLUSIONS

Clinicians believed they had sufficient knowledge and skills, which could explain why no difference was found after the educational intervention. In the HPI cohort, IOH was significantly reduced compared to baseline, therefore HPI-software may help prevent IOH.

TRIAL REGISTRATION

ISRCTN 17,085,700 on May 9th, 2019.

AcumenTM hypotension prediction index guidance for prevention and treatment of hypotension in noncardiac surgery: a prospective, single-arm, multicenter trial

BACKGROUND

Intraoperative hypotension is common during noncardiac surgery and is associated with postoperative myocardial infarction, acute kidney injury, stroke, and severe infection. The Hypotension Prediction Index software is an algorithm based on arterial waveform analysis that alerts clinicians of the patient's likelihood of experiencing a future hypotensive event, defined as mean arterial pressure < 65 mmHg for at least 1 min.

METHODS

Two analyses included (1) a prospective, single-arm trial, with continuous blood pressure measurements from study monitors, compared to a historical comparison cohort. (2) A post hoc analysis of a subset of trial participants versus a propensity score-weighted contemporaneous comparison group, using external data from the Multicenter Perioperative Outcomes Group (MPOG). The trial included 485 subjects in 11 sites; 406 were in the final effectiveness analysis. The post hoc analysis included 457 trial participants and 15,796 comparison patients. Patients were eligible if aged 18 years or older, American Society of Anesthesiologists (ASA) physical status 3 or 4, and scheduled for moderate- to high-risk noncardiac surgery expected to last at least 3 h.

MEASUREMENTS

minutes of mean arterial pressure (MAP) below 65 mmHg and area under MAP < 65 mmHg.

RESULTS

Analysis 1: Trial subjects (n = 406) experienced a mean of 9 ± 13 min of MAP below 65 mmHg, compared with the MPOG historical control mean of 25 ± 41 min, a 65% reduction (p < 0.001). Subjects with at least one episode of hypotension (n = 293) had a mean of 12 ± 14 min of MAP below 65 mmHg compared with the MPOG historical control mean of 28 ± 43 min, a 58% reduction (p< 0.001). Analysis 2: In the post hoc inverse probability treatment weighting model, patients in the trial demonstrated a 35% reduction in minutes of hypotension compared to a contemporaneous comparison group [exponentiated coefficient: - 0.35 (95%CI - 0.43, - 0.27); p < 0.001].

CONCLUSIONS

The use of prediction software for blood pressure management was associated with a clinically meaningful reduction in the duration of intraoperative hypotension. Further studies must investigate whether predictive algorithms to prevent hypotension can reduce adverse outcomes.

TRIAL REGISTRATION

Clinical trial number: NCT03805217. Registry URL: https://clinicaltrials.gov/ct2/show/NCT03805217 . Principal investigator: Xiaodong Bao, MD, PhD. Date of registration: January 15, 2019.

The Incidence of Perioperative Hypotension in Patients Undergoing Major Abdominal Surgery with the Use of Arterial Waveform Analysis and the Hypotension Prediction Index Hemodynamic Monitoring-A Retrospective Analysis

Intraoperative hypotension (IH) is common in patients receiving general anesthesia and can lead to serious complications such as kidney failure, myocardial injury and increased mortality. The Hypotension Prediction Index (HPI) algorithm is a machine learning system that analyzes the arterial pressure waveform and alerts the clinician of an impending hypotension event. The purpose of the study was to compare the frequency of perioperative hypotension in patients undergoing major abdominal surgery with different types of hemodynamic monitoring. The study included 61 patients who were monitored with the arterial pressure-based cardiac output (APCO) technology (FloTrac group) and 62 patients with the Hypotension Prediction Index algorithm (HPI group). Our primary outcome was the time-weighted average (TWA) of hypotension below < 65 mmHg. The median TWA of hypotension in the FloTrac group was 0.31 mmHg versus 0.09 mmHg in the HPI group (p = 0.000009). In the FloTrac group, the average time of hypotension was 27.9 min vs. 8.1 min in the HPI group (p = 0.000023). By applying the HPI algorithm in addition to an arterial waveform analysis alone, we were able to significantly decrease the frequency and duration of perioperative hypotension events in patients who underwent major abdominal surgery.

Personalized Predictive Hemodynamic Management for Gynecologic Oncologic Surgery: Feasibility of Cost-Benefit Derivatives of Digital Medical Devices

BACKGROUND

Intraoperative hypotension is associated with increased perioperative complications, hospital length of stay (LOS) and healthcare expenditure in gynecologic surgery. We tested the hypothesis that the adoption of a machine learning-based warning algorithm (hypotension prediction index-HPI) might yield an economic advantage, with a reduction in adverse outcomes that outweighs the costs for its implementation as a medical device.

METHODS

A retrospective-matched cohort cost-benefit Italian study in gynecologic surgery was conducted. Sixty-six female patients treated with standard goal-directed therapy (GDT) were matched in a 2:1 ratio with thirty-three patients treated with HPI based on ASA status, diagnosis, procedure, surgical duration and age.

RESULTS

The most relevant contributor to medical costs was operating room occupation (46%), followed by hospital stay (30%) and medical devices (15%). Patients in the HPI group had EURO 300 greater outlay for medical devices without major differences in total costs (GDT 5425 (3505, 8127), HPI 5227 (4201, 7023) p = 0.697). A pre-specified subgroup analysis of 50% of patients undergoing laparotomic surgery showed similar medical device costs and total costs, with a non-significant saving of EUR 1000 in the HPI group (GDT 8005 (5961, 9679), HPI 7023 (5227, 11,438), p = 0.945). The hospital LOS and intensive care unit stay were similar in the cohorts and subgroups.

CONCLUSIONS

Implementation of HPI is associated with a scenario of cost neutrality, with possible economic advantage in high-risk settings.

Artificial intelligence in anesthesiology

Artificial intelligence (AI) is rapidly evolving and gaining attention in the medical world. Our aim is to provide readers with insights into this quickly changing medical landscape and the role of clinicians in the middle of this popular technology. In this review, our aim is to explain some of the increasingly frequently used AI terminology explicitly for physicians. Next, we give a summation, an overview of currently existing applications, future possibilities for AI in the medical field of anesthesiology and thoroughly highlight possible problems that could arise from implementing this technology in daily practice. Therefore, we conducted a literature search, including all types of articles published between the first of January 2010 and the 1st of May 2023, written in English, and having a free full text available. We searched Pubmed, Medline, and Embase using “artificial intelligence”, “machine learning”, “deep learning”, “neural networks” and “anesthesiology” as MESH terms. To structure these findings, we divided the results into five categories: preoperatively, perioperatively, postoperatively, AI in the intensive care unit and finally, AI used for teaching purposes. In the first category, we found AI applications for airway assessment, risk prediction, and logistic support. Secondly, we made a summation of AI applications used during the operation. AI can predict hypotensive events, delivering automated anesthesia, reducing false alarms, and aiding in the analysis of ultrasound anatomy in locoregional anesthesia and echocardiography. Thirdly, namely postoperatively, AI can be applied in predicting acute kidney injury, pulmonary complications, postoperative cognitive dysfunction and can help to diagnose postoperative pain in children. At the intensive care unit, AI tools discriminate acute respiratory distress syndrome (ARDS) from pulmonary oedema in pleural ultrasound, predict mortality and sepsis more accurately, and predict survival rates in severe Coronavirus-19 (COVID-19). Finally, AI has been described in training residents in spinal ultrasound, simulation, and plexus block anatomy. Several concerns must be addressed regarding the use of AI. Firstly, this software does not explain its decision process (i.e., the ‘black box problem’). Secondly, to develop AI models and decision support systems, we need big and accurate datasets, unfortunately with potential unknown bias. Thirdly, we need an ethical and legal framework before implementing this technology. At the end of this paper, we discuss whether this technology will be able to replace the clinician one day. This paper adds value to already existing literature because it not only offers a summation of existing literature on AI applications in anesthesiology but also gives clear definitions of AI itself and critically assesses implementation of this technology.

Hypotension Prediction Index and Incidence of Perioperative Hypotension: A Single-Center Propensity-Score-Matched Analysis

(1) Background: Intraoperative hypotension is common and is associated with increased morbidity and mortality. The Hypotension Prediction Index (HPI) is an advancement of arterial waveform analysis and allows preventive treatments. We used a propensity-score-matched study design to test whether application of the HPI reduces hypotensive events in non-cardiac surgery patients; (2) Methods: 769 patients were selected for propensity score matching. After matching, both HPI and non-HPI groups together comprised n = 136 patients. A goal-directed treatment protocol was applied in both groups. The primary endpoint was the incidence and duration of hypotensive events defined as MAP < 65 mmHg, evaluated by the time-weighted average (TWA) of hypotension. (3) Results: The median TWA of hypotension below 65 mmHg in the matched cohort was 0.180 mmHg (IQR 0.060, 0.410) in the non-HPI group vs. 0.070 mmHg (IQR 0.020, 0.240) in the HPI group (p < 0.001). TWA was higher in patients with ASA classification III/IV (0.170 mmHg; IQR 0.035, 0.365) than in patients with ASA status II (0.100; IQR 0.020, 0.250; p = 0.02). Stratification by intervention group showed no differences in the HPI group while TWA values in the non-HPI group were more than twice as high in patients with ASA status III/IV (p = 0.01); (4) Conclusions: HPI reduces intraoperative hypotension in a matched cohort seen for TWA below 65 mmHg and relative time in hypotension. In addition, non-HPI patients with ASA status III/IV showed a higher TWA compared with HPI-patients, indicating an advantageous effect of using HPI in patients at higher risk.

Intraoperative Hypotension Prediction—A Proactive Perioperative Hemodynamic Management—A Literature Review

Intraoperative hypotension (IH) is a frequent phenomenon affecting a substantial number of patients undergoing general anesthesia. The occurrence of IH is related to significant perioperative complications, including kidney failure, myocardial injury, and even increased mortality. Despite advanced hemodynamic monitoring and protocols utilizing goal directed therapy, our management is still reactive; we intervene when the episode of hypotension has already occurred. This literature review evaluated the Hypotension Prediction Index (HPI), which is designed to predict and reduce the incidence of IH. The HPI algorithm is based on a machine learning algorithm that analyzes the arterial pressure waveform as an input and the occurrence of hypotension with MAP <65 mmHg for at least 1 min as an output. There are several studies, both retrospective and prospective, showing a significant reduction in IH episodes with the use of the HPI algorithm. However, the level of evidence on the use of HPI remains very low, and further studies are needed to show the benefits of this algorithm on perioperative outcomes.

Effect of hypotension prediction index in the prevention of intraoperative hypotension during noncardiac surgery: A systematic review

Intraoperative hypotension (IOH) is common in noncardiac surgery and is associated with serious postoperative complications. Hypotension Prediction Index (HPI) has shown high sensitivity and specificity for predicting hypotension and may reduce IOH in noncardiac surgery. We conducted a systematic review of randomized controlled trials (RCTs) to evaluate the applications and effects of HPI in reducing hypotension during noncardiac surgery. We comprehensively searched the PubMed, Embase, Cochrane Library, Google Scholar, and http://ClinicalTrials.gov databases to identify RCTs conducted before May 2022. The primary outcome measures were the time-weighted average (TWA) of hypotension and the area under the hypotensive threshold (65 mmHg). Secondary outcomes were the incidence and duration of hypotension and the percentage of hypotensive time during surgery. The Cochrane Risk of Bias (RoB) tool was used to assess the quality of selected studies. We conducted data synthesis for median differences and assessed the certainty of evidence using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) approach. We included five studies with a total of 461 patients. Limited evidence suggested that HPI-guided intraoperative hemodynamics management leads to lower a) TWA of hypotension (median of difference of medians [MDM], -0.27 mmHg; 95% confidence interval [CI], -0.38, -0.01), b) area under the hypotensive threshold (MDM, -60.28 mmHg*min; 95% CI, -74.00, -1.30), c) incidence of hypotension (MDM, -4.50; 95% CI, -5.00, -4.00), d) total duration of hypotension (MDM, -12.80 min; 95% CI, -16.11, -3.39), and e) percentage of hypotension (MDM, -5.80; 95% CI, -6.65, -4.82) than routine hemodynamic management during noncardiac surgery. However, only very low- to low-quality evidence on the benefit of intraoperative HPI-based hemodynamic management is available. Our review revealed that HPI has the potential to reduce the occurrence, duration, and severity of IOH during noncardiac surgery compared to standard intraoperative care with proper adherence to the protocol. Systematic review registration PROSPERO CRD42022333834.

Prediction and Prevention of Intraoperative Hypotension with the Hypotension Prediction Index: A Narrative Review

Intraoperative hypotension is common and has been associated with adverse events. Although association does not imply causation, predicting and preventing hypotension may improve postoperative outcomes. This review summarizes current evidence on the development and validation of an artificial intelligence predictive algorithm, the Hypotension Prediction (HPI) (formerly known as the Hypotension Probability Indicator). This machine learning model can arguably predict hypotension up to 15 min before its occurrence. Several validation studies, retrospective cohorts, as well as a few prospective randomized trials, have been published in the last years, reporting promising results. Larger trials are needed to definitively assess the usefulness of this algorithm in optimizing postoperative outcomes.