Intraoperative prediction of postanaesthesia care unit hypotension
Copyright © 2021 British Journal of Anaesthesia. Published by Elsevier Ltd. All rights reserved..
BACKGROUND: Postoperative hypotension is associated with adverse outcomes, but intraoperative prediction of postanaesthesia care unit (PACU) hypotension is not routine in anaesthesiology workflow. Although machine learning models may support clinician prediction of PACU hypotension, clinician acceptance of prediction models is poorly understood.
METHODS: We developed a clinically informed gradient boosting machine learning model using preoperative and intraoperative data from 88 446 surgical patients from 2015 to 2019. Nine anaesthesiologists each made 192 predictions of PACU hypotension using a web-based visualisation tool with and without input from the machine learning model. Questionnaires and interviews were analysed using thematic content analysis for model acceptance by anaesthesiologists.
RESULTS: The model predicted PACU hypotension in 17 029 patients (area under the receiver operating characteristic [AUROC] 0.82 [95% confidence interval {CI}: 0.81-0.83] and average precision 0.40 [95% CI: 0.38-0.42]). On a random representative subset of 192 cases, anaesthesiologist performance improved from AUROC 0.67 (95% CI: 0.60-0.73) to AUROC 0.74 (95% CI: 0.68-0.79) with model predictions and information on risk factors. Anaesthesiologists perceived more value and expressed trust in the prediction model for prospective planning, informing PACU handoffs, and drawing attention to unexpected cases of PACU hypotension, but they doubted the model when predictions and associated features were not aligned with clinical judgement. Anaesthesiologists expressed interest in patient-specific thresholds for defining and treating postoperative hypotension.
CONCLUSIONS: The ability of anaesthesiologists to predict PACU hypotension was improved by exposure to machine learning model predictions. Clinicians acknowledged value and trust in machine learning technology. Increasing familiarity with clinical use of model predictions is needed for effective integration into perioperative workflows.
| Medienart: |
E-Artikel |
|---|
| Erscheinungsjahr: |
2022 |
|---|---|
| Erschienen: |
2022 |
| Enthalten in: |
Zur Gesamtaufnahme - volume:128 |
|---|---|
| Enthalten in: |
British journal of anaesthesia - 128(2022), 4 vom: 01. Apr., Seite 623-635 |
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Palla, Konstantina [VerfasserIn] |
|---|
| Links: |
|---|
| Themen: |
Data science |
|---|
| Anmerkungen: |
Date Completed 04.04.2022 Date Revised 16.09.2023 published: Print-Electronic Citation Status MEDLINE |
|---|
| doi: |
10.1016/j.bja.2021.10.052 |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
NLM334602505 |
|---|
| LEADER | 01000naa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLM334602505 | ||
| 003 | DE-627 | ||
| 005 | 20231225224047.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231225s2022 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1016/j.bja.2021.10.052 |2 doi | |
| 028 | 5 | 2 | |a pubmed24n1115.xml |
| 035 | |a (DE-627)NLM334602505 | ||
| 035 | |a (NLM)34924175 | ||
| 035 | |a (PII)S0007-0912(21)00758-3 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Palla, Konstantina |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Intraoperative prediction of postanaesthesia care unit hypotension |
| 264 | 1 | |c 2022 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a ƒaComputermedien |b c |2 rdamedia | ||
| 338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Date Completed 04.04.2022 | ||
| 500 | |a Date Revised 16.09.2023 | ||
| 500 | |a published: Print-Electronic | ||
| 500 | |a Citation Status MEDLINE | ||
| 520 | |a Copyright © 2021 British Journal of Anaesthesia. Published by Elsevier Ltd. All rights reserved. | ||
| 520 | |a BACKGROUND: Postoperative hypotension is associated with adverse outcomes, but intraoperative prediction of postanaesthesia care unit (PACU) hypotension is not routine in anaesthesiology workflow. Although machine learning models may support clinician prediction of PACU hypotension, clinician acceptance of prediction models is poorly understood | ||
| 520 | |a METHODS: We developed a clinically informed gradient boosting machine learning model using preoperative and intraoperative data from 88 446 surgical patients from 2015 to 2019. Nine anaesthesiologists each made 192 predictions of PACU hypotension using a web-based visualisation tool with and without input from the machine learning model. Questionnaires and interviews were analysed using thematic content analysis for model acceptance by anaesthesiologists | ||
| 520 | |a RESULTS: The model predicted PACU hypotension in 17 029 patients (area under the receiver operating characteristic [AUROC] 0.82 [95% confidence interval {CI}: 0.81-0.83] and average precision 0.40 [95% CI: 0.38-0.42]). On a random representative subset of 192 cases, anaesthesiologist performance improved from AUROC 0.67 (95% CI: 0.60-0.73) to AUROC 0.74 (95% CI: 0.68-0.79) with model predictions and information on risk factors. Anaesthesiologists perceived more value and expressed trust in the prediction model for prospective planning, informing PACU handoffs, and drawing attention to unexpected cases of PACU hypotension, but they doubted the model when predictions and associated features were not aligned with clinical judgement. Anaesthesiologists expressed interest in patient-specific thresholds for defining and treating postoperative hypotension | ||
| 520 | |a CONCLUSIONS: The ability of anaesthesiologists to predict PACU hypotension was improved by exposure to machine learning model predictions. Clinicians acknowledged value and trust in machine learning technology. Increasing familiarity with clinical use of model predictions is needed for effective integration into perioperative workflows | ||
| 650 | 4 | |a Journal Article | |
| 650 | 4 | |a data science | |
| 650 | 4 | |a hypotension | |
| 650 | 4 | |a machine learning | |
| 650 | 4 | |a postanaesthesia care unit | |
| 650 | 4 | |a risk prediction | |
| 700 | 1 | |a Hyland, Stephanie L |e verfasserin |4 aut | |
| 700 | 1 | |a Posner, Karen |e verfasserin |4 aut | |
| 700 | 1 | |a Ghosh, Pratik |e verfasserin |4 aut | |
| 700 | 1 | |a Nair, Bala |e verfasserin |4 aut | |
| 700 | 1 | |a Bristow, Melissa |e verfasserin |4 aut | |
| 700 | 1 | |a Paleva, Yoana |e verfasserin |4 aut | |
| 700 | 1 | |a Williams, Ben |e verfasserin |4 aut | |
| 700 | 1 | |a Fong, Christine |e verfasserin |4 aut | |
| 700 | 1 | |a Van Cleve, Wil |e verfasserin |4 aut | |
| 700 | 1 | |a Long, Dustin R |e verfasserin |4 aut | |
| 700 | 1 | |a Pauldine, Ronald |e verfasserin |4 aut | |
| 700 | 1 | |a O'Hara, Kenton |e verfasserin |4 aut | |
| 700 | 1 | |a Takeda, Kenji |e verfasserin |4 aut | |
| 700 | 1 | |a Vavilala, Monica S |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t British journal of anaesthesia |d 1946 |g 128(2022), 4 vom: 01. Apr., Seite 623-635 |w (DE-627)NLM000000310 |x 1471-6771 |7 nnns |
| 773 | 1 | 8 | |g volume:128 |g year:2022 |g number:4 |g day:01 |g month:04 |g pages:623-635 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.1016/j.bja.2021.10.052 |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a GBV_NLM | ||
| 951 | |a AR | ||
| 952 | |d 128 |j 2022 |e 4 |b 01 |c 04 |h 623-635 | ||