EUVIMEDEUVIMED entdecken
Systematische ÜbersichtMetaanalyseOpen Access · CC BY 4.0

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

Vorhersageleistung des hypotension prediction index und von Machine-Learning-Verfahren bei intraoperativer Hypotonie: systematische Übersicht und Metaanalyse

Ida Mohammadi, Shahryar Rajai Firouzabadi, Melika Hosseinpour, Mohammadhosein Akhlaghpasand iD, Bardia Hajikarimloo, Roozbeh Tavanaei, Amirreza Izadi, Sam Zeraatian-Nejad, Foolad Eghbali

Journal of Translational Medicine · 05.08.2024 · 22 (1)

DOI: 10.1186/s12967-024-05481-4 ↗

Auf einen Blick

43 Studien untersuchten KI-Modelle zur Vorhersage intraoperativer Hypotonie. Der HPI war mit kürzerer Hypotoniedauer und hoher Vorhersageleistung verbunden.

Fragestellung

Wie gut sagen der hypotension prediction index (HPI) und andere KI-Modelle intraoperative Hypotonie voraus?

Methode

Systematische Übersicht und Metaanalyse; Suche in PubMed, Scopus, Web of Science und Embase; Einschluss von KI-Modellen zur Vorhersage intraoperativer Hypotonie bei jeder OP-Art.

Ergebnisse

43 Studien. Unter HPI sanken Dauer der intraoperativen Hypotonie, TWA-MAP < 65 mmHg und AUT-MAP < 65 mmHg. AUROC des HPI 0,89; gepoolte AUROC nicht-HPI-Modelle 0,79 (95 %-KI 0,74–0,83).

Schlussfolgerung der Autor:innen

HPI zeigte eine ausgezeichnete Fähigkeit, hypotensive Episoden vorherzusagen und die Hypotoniedauer zu senken. Andere KI-Modelle, besonders Deep-Learning-Modelle, sagten IOH gut voraus; die Wirkung auf die Dauer bleibt unklar.

Kernaussagen

  • HPI war mit kürzerer Hypotoniedauer verbunden
  • AUROC des HPI: 0,89
  • Nicht-HPI-Modelle: gepoolte AUROC 0,79 (95 %-KI 0,74–0,83)

Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.

Kurz gefasst

Systematische Übersicht in der Zeitschrift Journal of Translational Medicine, online erschienen am 05.08.2024. Verfasst von Ida Mohammadi und 8 weiteren Autor:innen. Beteiligte Einrichtung: Cardiovascular Surgery Research and Development Committee, Iran University of Medical Sciences (IUMS), Tehran, 14665-354, Iran. Hauptthemen laut MEDLINE: Hypotension, Machine Learning. Die Arbeit wurde bisher 58-mal zitiert (Crossref) und nennt 68 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

Auf einen BlickAbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

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.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

HypotensionMachine LearningHumansIntraoperative ComplicationsROC Curve

Stichwörter der Autor:innen

AnesthesiaArtificial intelligenceDeep learningIntraoperative hypotensionMachine learning

Autor:innen und Institutionen

  • Cardiovascular Surgery Research and Development Committee, Iran University of Medical Sciences (IUMS), Tehran, 14665-354, Iran.
  • Cardiovascular Surgery Research and Development Committee, Iran University of Medical Sciences (IUMS), Tehran, 14665-354, Iran. Akhlaghpasandm@yahoo.com.
  • Department of Surgery, Surgery Research Center, School of Medicine, Rasool-E Akram Hospital, Iran University of Medical Sciences, Tehran, Iran. Akhlaghpasandm@yahoo.com.
  • Department of Surgery, Surgery Research Center, School of Medicine, Rasool-E Akram Hospital, Iran University of Medical Sciences, Tehran, Iran.

ORCID-Profile: Mohammadhosein Akhlaghpasand

Zitieren

APA

Mohammadi, I., Firouzabadi, S. R., Hosseinpour, M., Akhlaghpasand, M., Hajikarimloo, B., Tavanaei, R., Izadi, A., Zeraatian-Nejad, S., & Eghbali, F. (2024). Predictive ability of hypotension prediction index and machine learning methods in intraoperative hypotension: a systematic review and meta-analysis. Journal of Translational Medicine, 22(1). https://doi.org/10.1186/s12967-024-05481-4

Vancouver

Mohammadi I, Firouzabadi SR, Hosseinpour M, Akhlaghpasand M, Hajikarimloo B, Tavanaei R, et al. Predictive ability of hypotension prediction index and machine learning methods in intraoperative hypotension: a systematic review and meta-analysis. Journal of Translational Medicine. 2024;22(1). doi:10.1186/s12967-024-05481-4

BibTeX
@article{Mohammadi2024,
  author = {Mohammadi, Ida and Firouzabadi, Shahryar Rajai and Hosseinpour, Melika and Akhlaghpasand, Mohammadhosein and Hajikarimloo, Bardia and Tavanaei, Roozbeh and Izadi, Amirreza and Zeraatian-Nejad, Sam and Eghbali, Foolad},
  title = {Predictive ability of hypotension prediction index and machine learning methods in intraoperative hypotension: a systematic review and meta-analysis},
  journal = {Journal of Translational Medicine},
  year = {2024},
  volume = {22},
  number = {1},
  doi = {10.1186/s12967-024-05481-4}
}
Kostenlos starten →