The effect of intraoperative hypotension on postoperative delirium: a meta-analysis and systematic review

Background: Postoperative delirium (POD) is a common complication linked to poor outcomes, yet its relationship with intraoperative hypotension (IOH) remains unclear. The objective of this study was to clarify the association between intraoperative IOH and POD.

Method: We searched four databases (PubMed, EMBASE, Cochrane Library, and Web of Science) from their inception to June 14, 2025. Randomized controlled trials (RCTs) and observational studies were included when IOH was incorporated as a predictive variable for POD in adult patients undergoing elective surgery under general anesthesia. Risk ratio (RR) and odds ratio (OR) were calculated using a random-effect model separately in RCTs and observational studies.

Results: In total, thirty out of 1,261 studies were included for the systematic review, of which 18 studies were eligible for quantitative meta-analysis. The remaining 12 studies were excluded due to incompatible data formats: eight used continuous metrics, and four lacked extractable effect estimates. IOH was significantly associated with an elevated risk of POD in both RCTs (RR: 1.89, 95%CI: 1.31–2.74) and observational studies (OR: 2.48, 95%CI: 1.64–3.75). Subgroup analysis of observational studies revealed that IOH defined by absolute threshold (OR: 4.11, 95%CI: 2.05–8.24) and mean arterial pressure (MAP) (OR: 2.90, 95%CI: 1.56–5.39) was related to a higher risk of POD. This heterogeneity was further explored by meta-regression, which identified the threshold nature of the IOH definition as a significant source of heterogeneity and a key effect modifier (p = 0.048).

Conclusion: Our meta-analysis demonstrates a statistically significant association between IOH and increased risk of POD. However, substantial methodological heterogeneity across the included studies limits the robustness of these findings. The current evidence should therefore be interpreted as exploratory, highlighting the need for more standardized investigations in this field.

Systematic review registration: Identifier CRD42023424166.

1 Introduction

Postoperative delirium (POD) is defined as an acute disorder of attention and cognition that occurs in the hospital up to 1 week post-procedure or until discharge (1, 2). The incidence of POD varies from 12 to 51% with different surgery types (3). POD is associated with an increase in adverse postoperative complications, delayed rehabilitation, and a higher mortality rate and imposes a heavy burden on the healthcare system (4, 5). A wide range of risk factors for POD has been suggested, including advanced age, pre-existing cognitive impairment, sleep disturbance, psychiatric disorders, and coexisting medical conditions (6–8).

Intraoperative hypotension (IOH) is commonly described as a decrease in the mean arterial pressure (MAP) or the systolic blood pressure (SBP) below a predefined threshold during surgery. However, there is no consensus definition, leading to substantial heterogeneity in its reporting across studies. This heterogeneity stems from variations in multiple definitional components: the blood pressure parameter (MAP vs. SBP), the threshold nature (absolute vs. relative), and often additional criteria such as minimum duration or number of episodes. Furthermore, studies quantify hypotension exposure using fundamentally different formats, primarily as dichotomous outcomes (IOH present/absent) or as continuous metrics (e.g., area under the curve). This lack of standardization directly impacts the reported incidence of IOH, which fluctuates widely between 6 and 44% (9–11) In several studies, IOH was thought to be associated with POD, one of the mechanisms of which was realized by reducing the cerebral blood infusion and oxygen supply (12, 13).

However, the association between IOH and POD remains inconclusive, as evidenced by conflicting conclusions from previous meta-analyses. These discrepant conclusions likely stem from limitations such as restricted search strategies, insufficient sample sizes, and crucially, a lack of systematic handling of the pronounced heterogeneity in IOH definitions across primary studies (14, 15). Therefore, we conducted a systematic review and meta-analysis and aimed to clarify whether IOH contributes to the development of POD.

2 Materials and methods

This meta-analysis was registered on PROSPERO (registration number: CRD42023424166) and performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines (16).

2.1 Search strategy

Two authors (XW-Y and HL-Z) independently searched four databases (PubMed, EMBASE, Cochrane Library, and Web of Science) from database establishment to June 14, 2025. Search terms contained both Medical Subject Headings (MeSH) terms and free text to define exposure (IOH) and outcome (POD). The full search strategy is available in Supplementary Data.

2.2 Study selection and eligibility

Two authors (XW-Y and HL-Z) independently reviewed the identified studies. The full-text of potentially relevant articles was retrieved after screening titles and abstracts for eligibility. After the removal of duplicates, we screened the titles and abstracts for relevance and retrieved the accessible full text to identify the eligibility of studies for inclusion. Reviews, conference abstracts, letters, case series, and studies of pediatric surgery were all excluded. Disagreements were resolved by discussion with another author (Q-L).

2.3 Inclusion and exclusion

Inclusion criteria: (a) Design: Randomized controlled trials (RCTs) and observational studies; (b) Population: patients undergoing elective surgery under general anesthesia; (c) Intervention or exposure: different blood pressure groups or IOH with a clear definition; (d) Primary outcomes: POD evaluated with any proven and effective measuring tool. Exclusion criteria: (a) non-availability of full texts; (b) language other than English or Chinese.

2.4 Data extraction

Two authors (XW-Y and HL-Z) extracted data from each study, and conflicts were resolved by consensus with another author (Q-L) to ensure consistency and accuracy. The following information was extracted from each included study: study design, patient characteristics, type of surgery, definition of IOH, screening material of POD, timing of POD screening, definition of POD, and POD incidence.

To systematically manage the heterogeneity in the definition of IOH, we deconstructed its core component. The core components of any definition of IOH included the Blood Pressure Parameter (MAP or SBP) and the Threshold Nature (absolute threshold or relative threshold). A proportion of definitions intended to generate a dichotomous outcome incorporated additional criteria for dichotomization, such as stipulating a minimum duration or a minimum number of episodes below the threshold, as a requisite for confirming an IOH event.

Based on the format of the data presented for the IOH variable, studies were classified into two distinct groups. The first group, termed “Dichotomous Definitions,” utilized the core components to classify patients into “IOH present” or “IOH absent.” The second group, termed “Continuous Metrics,” quantified hypotension exposure as a continuous variable using distinct calculation methods, such as variably derived Area Under the Curve (AUC) or other composite measures. Owing to fundamental inconsistencies in the algorithms for these continuous metrics across studies—which introduced insurmountable methodological heterogeneity—studies in this category were included in the qualitative synthesis but excluded from the meta-analysis.

Finally, to ensure clarity and avoid double-counting in our synthesis, when a study reported multiple IOH definitions, a single ‘primary’ definition was selected for synthesis. The selection was based on a frequency analysis across all included studies: we first tabulated all unique definitions from studies that reported only one definition. For studies reporting multiple definitions, the definition that matched the most frequently occurring one in the aforementioned tabulation was selected as the primary definition. Similarly, multiple comparable groups within a single study were combined into a single group for analysis.

2.5 Risk of bias

Two authors (XW-Y and HL-Z) independently assessed the quality of included studies using the Newcastle-Ottawa Scale for observational studies (17) and the Cochrane risk-of-bias tool for RCTs (18).

2.6 Data synthesis

Systematic review and meta-analysis were performed using Review Manager 5.4.1. The analytical approach was determined by the classification of the IOH definition. For studies employing “Dichotomous Definitions,” data were extracted into 2 × 2 contingency tables to calculate the effect size. Conversely, studies utilizing “Continuous Metrics” for IOH were not included in the meta-analysis due to insurmountable methodological heterogeneity in pooling these measures. However, the findings from these studies were presented descriptively.

We maintained segregation between RCTs and observational studies in our meta-analysis due to fundamental methodological disparities and potential clinical heterogeneity across study designs. Risk ratio (RR) and odds ratio (OR) were calculated separately for RCTs and observational studies. A random-effect meta-analysis model was established to test the difference in incidence of POD between patients with or without IOH (19). The inverse variance method was used for weighing the studies. Between-study variance was estimated using the DerSimonian and Laird method. The I² value was calculated to evaluate heterogeneity. An I² value > 50% indicates substantial heterogeneity. Due to the limited number of available RCTs, both the subgroup analyses and meta-regression were restricted to observational studies to ensure robust and interpretable results. Subgroup analyses were performed to descriptively compare effect sizes across categories of IOH definition, screening material of POD, and region. To explore sources of heterogeneity, univariable meta-regression analyses were then applied using the restricted maximum-likelihood method to examine the relationship between these pre-specified covariates and the effect size; a multivariable meta-regression was not performed due to the limited number of studies available for each model, which would have resulted in an underpowered and statistically unstable analysis. Furthermore, sensitivity analysis was performed to assess the influence of individual studies on the pooled results by sequentially excluding each study. A funnel plot was used to assess the potential publication bias. Given the limited number of RCTs, which precludes reliable testing, Egger’s linear regression test was applied specifically to the pooled analysis of observational studies to statistically evaluate funnel plot asymmetry.

3 Results

3.1 Search results and study characteristics

The study selection process is presented in Figure 1. The initial search produced 1,261 articles. After 145 duplicate references were removed, a total of 1,116 articles were screened for eligibility. Sixty-three articles remained after excluding records according to title or abstract. After screening the full texts, 33 articles were excluded. As a result, a total of 30 articles with 372,158 patients were included in the systematic review, of which 18 articles with 322,879 patients were fit for the meta-analysis (9–11, 14, 15, 20–44).

Figure 1

Figure 1. PRISMA study flow diagram.

The detailed characteristics of the 30 articles included are presented in Table 1. Overall, this review included five RCTs, 15 cohort studies, and 10 case–controlled studies with publication dates ranging from 1998 to 2025. The mean age ranged from 48.34 ± 12.87 years to 81.5 ± 5.5 years. Langer et al. (30) designed an RCT; however, since the outcome of interest was studied in a cohort design, it was classified as a cohort study.

Table 1

Table 1. Study characteristics of included studies.

The definitions of IOH and POD are classified and summarized in Table 2. Following our pre-specified classification criteria, the definitions of IOH from the included studies were systematically categorized. Core components of the definitions, namely the blood pressure parameter and threshold nature, were analyzed across all studies. Among the 30 included studies, 25 used an absolute threshold, with the most frequently used definition (employed by nine studies) being a MAP less than 60 mmHg. Ten studies utilized a relative threshold, the most common of which (in five studies) was a MAP decline to less than 70% of the preoperative baseline. Based on how these components were applied for data presentation, studies were grouped accordingly. In five RCTs and seventeen observational studies, IOH was defined using “Dichotomous Definitions.” Furthermore, four of these studies incorporated additional criteria for dichotomization, requiring that blood pressure remain below the threshold for a defined duration or occur in a specified number of episodes to confirm an IOH event. The remaining eight observational studies quantified hypotension exposure using “Continuous Metrics,” the majority of which employed variably calculated AUC. The screening tools for POD varied between studies. The Confusion Assessment Method (CAM) and its intensive care unit variant (CAM-ICU) were used to screen for POD in 14 studies and seven studies, respectively. The Mini-Mental State Examination (MMSE) or International Classification of Diseases (ICD) codes were applied in the other seven studies. Time points used to evaluate cognitive function also varied between studies. Only eight studies performed POD screening lasting up to 7 days after surgery.

Table 2

Table 2. Definition of intraoperative hypotension and postoperative delirium in the included studies.

Fifteen out of 30 studies found no statistically significant differences in the incidence of POD between hypotensive and normotensive patients, and these 15 studies comprised three RCTs and twelve observational studies. Seven studies investigated the association between IOH and POD in cardiac surgeries, and three of these found a significant difference. However, a formal statistical comparison between cardiac and non-cardiac surgeries was precluded by both methodological and design constraints. Among the seven cardiac surgery studies, the majority (four using Continuous Metrics and one cohort with unextractable data) were inherently excluded from the planned observational study-based subgroup analysis. This left only one eligible observational study (a case–control study) for the cardiac surgery group. A comparison based on a single study is statistically untenable, thus preventing a valid subgroup analysis. The POD incidence was reported in 18 studies out of the 22 studies that described IOH using “Dichotomous Definitions,” ranging from 0.9 to 58% in the hypotensive groups and 0 to 33% in the normotensive groups. Of the eight studies that quantified hypotension exposure using “Continuous Metrics,” only Maheshwari et al. found that IOH was associated with POD; however, in the other seven studies, IOH was not related to POD.

3.2 Meta-analysis

As shown in Figures 2, 3, IOH was significantly associated with an elevated risk of POD in both RCTs (RR: 1.89, 95%CI: 1.31–2.74) and observational studies (OR: 2.48, 95%CI: 1.64–3.75). Heterogeneity between observational studies was high (I² = 89%), whereas it was low (I² = 22%) among RCTs.

Figure 2

Figure 2. Primary outcome (RCTs)—forest plot showing the risk ratios for the association of intraoperative hypotension with postoperative delirium.

Figure 3

Figure 3. Primary outcome (observational studies)—forest plot showing the odds ratios for the association of intraoperative hypotension with postoperative delirium.

Figure 4 shows the findings of the subgroup meta-analysis of observational studies based on various factors. In the subgroup analysis stratified by the Threshold Nature of the IOH definition, a significant association with a higher risk of POD was observed for absolute thresholds (OR: 4.11, 95% CI: 2.05–8.24). Conversely, no significant association was identified for relative thresholds (OR: 1.36, 95% CI: 0.93–1.99). When stratified by the Blood Pressure Parameter used in the definition, a significant association was observed for definitions based on MAP (OR: 2.90, 95% CI: 1.56–5.39), whereas the association for definitions based on SBP did not reach statistical significance (OR: 2.51, 95% CI: 0.96–6.57). Additionally, IOH was associated with a higher risk of POD in the subgroup meta-analysis using the POD screening tool. In the subgroup meta-analysis based on region, IOH was related to an increased risk of POD in both Asia (OR: 3.04, 95%CI: 1.49–6.21) and Europe (OR: 5.54, 95%CI: 2.22–13.79) studies, but IOH showed no significant association with POD (OR: 1.20, 95%CI: 0.70–2.05) in studies of America.

Figure 4

Figure 4. Subgroup analysis (observational studies)—forest plot of odds ratios for the association between intraoperative hypotension and postoperative delirium. Results are stratified by the definition of intraoperative hypotension (Threshold Nature, Blood Pressure Parameter), the Screening tool of delirium, and Region. CAM, Confusion Assessment Method; DSM-5, Diagnostic and statistical manual of mental disorders, 5th edition; IOH, Intraoperative hypotension; MAP, Mean arterial pressure; SBP, Systolic blood pressure.

To investigate potential sources of high heterogeneity, we performed a univariate meta-regression using the same variables as in the subgroup analyses. As shown in Supplementary Table S1, the association between IOH and POD was notably affected by the moderator Threshold Nature, substantially reducing the tau value (from 0.9526 to 0.8069) and the I² statistic (from 94.14 to 88.94%), suggesting that this moderator explains a portion of the observed heterogeneity. Furthermore, as shown in Supplementary Table S2, this variable significantly modified the adjusted overall estimate (OR: 2.86, p = 0.048), indicating it is an important effect modifier. Conversely, other moderators, including Blood Pressure Parameter, region, study type, and delirium assessment tool, did not significantly explain heterogeneity or alter the effect size, with p-values for the moderator tests all greater than 0.05.

3.3 Sensitivity analysis and publication bias

Sensitivity analysis was performed to explore the impact of individual studies on the pooled results. The sequential exclusion of any of the included studies did not change the stability and reliability of the results. Visual inspection of the funnel plots in Figure 5 showed a generally symmetrical distribution across both RCTs and observational studies, with larger and smaller studies reporting both negative and positive results. Given the limited number of RCTs, formal statistical testing for publication bias was not performed. For observational studies, publication bias was further assessed using Egger’s linear regression test, which indicated no statistically significant asymmetry (p = 0.37). Taken together, these assessments suggest that the overall findings are unlikely to be substantially influenced by publication bias.

Figure 5

Figure 5. Funnel plot for RCTs and observational studies. (A) Funnel plot for RCTs. (B) Funnel plot for observational studies.

3.4 Quality assessment

The quality of the included studies is shown in Figure 6 and Tables 3, 4. A qualitative summary of the main concerns is as follows: For RCTs, the overall risk of bias was low. The primary limitation was the inherent risk of performance bias, given the impossibility of blinding clinicians to intraoperative blood pressure management. For cohort studies, common methodological concerns included insufficient follow-up duration for the outcome and questions regarding the representativeness of the exposed cohort in some studies. The selection and comparability of controls in case–control studies were generally well addressed.

Figure 6

Figure 6. Risk of bias summary: authors’ judgement on risk of bias for included RCTs. Assessed using the Cochrane Collaboration’s Risk of Bias Tool. Green: low risk/Yellow: unclear risk/Red: high risk.

Table 3

Table 3. Risk of bias summary: authors’ judgement on risk of bias for included cohort studies.^a

Table 4

Table 4. Risk of bias summary: authors’ judgement on risk of bias for included case–control studies.^a

4 Discussion

The quantitative analysis performed on observational studies and RCTs showed that IOH was associated with an increased risk of POD. A critical finding emerged from further investigation: the Threshold Nature of the IOH definition was identified as a pivotal factor. Subgroup analysis revealed a strong and significant association for absolute threshold, but not for relative threshold. This striking discrepancy was confirmed by meta-regression, which established the Threshold Nature as a significant source of heterogeneity and a key effect modifier.

Previous studies have found that IOH is strongly associated with myocardial injury, myocardial infarction, renal injury, and death (12). Organ hypoperfusion, including the heart, kidneys, and brain, might be the potential cause of these postoperative complications. However, so far, some evidence has revealed that postoperative neurological complications might not be associated with brain hypoperfusion. The study conducted by Feng et al. (45) suggested no significant correlation between IOH and the incidence of POD based on only two RCTs. Furthermore, Wijnberge et al. (46) found that IOH was not associated with delirium in a subgroup meta-analysis comprising three cohort studies. However, in a recent meta-analysis by Cai et al. (12), found that IOH was associated with POD in non-cardiac surgery based on the results synthesized by two RCTs and 15 cohort studies. We conducted a comprehensive search, including the latest literature. The conclusions of the aforementioned studies were limited by a combination of factors, including small sample size, inadequate search strategy, and a predominance of observational studies. Thus, we designed this meta-analysis and found that IOH was associated with the increased risk of POD in both RCTs and observational studies.

Studies that defined IOH by absolute threshold found a stronger association than studies that defined IOH by relative threshold. Besides, the comparison of studies defining IOH by MAP to studies defining IOH by SBP was consistent with the above. This pattern can be explained mechanistically by the inherent difference in these definitional criteria. For a substantial subset of patients, particularly those with hypertension, a relative threshold corresponds to a higher absolute pressure than an absolute threshold. Similarly, the SBP threshold is physiologically higher than the MAP threshold. This is empirically supported by Wickham et al., who reported a higher incidence of IOH when using relative or SBP-based definitions, indicating that for many individuals, these criteria identify a less severe state of hypotension (47). Consequently, definitions using absolute or MAP thresholds serve as more specific markers of severe hypotension, whereas those using relative or SBP thresholds capture a broader, heterogeneous group with a diluted association to POD. Given the stronger and more consistent association observed with absolute MAP thresholds, our findings support the use of standardized definitions centered on an absolute MAP value (e.g., < 60 mmHg) in future studies of IOH and POD. This approach is recommended to reduce definitional heterogeneity and facilitate comparability across studies. Ultimately, to definitively determine the optimal threshold for clinical guidance, a definitive large-scale RCT is required to directly compare the effects of multiple candidate MAP thresholds on POD incidence.

The screening material used in our included studies for POD is quite different. Diagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM-5) criteria, the gold standard for diagnosing delirium, were used in only two of the quantitative studies because they are difficult for non-psychiatrists to apply in clinical practice. Both of these studies supported the finding that IOH is associated with POD. Eight studies included in our analysis used CAM as a POD identification instrument, and three of these eight studies found that IOH was not associated with POD. Different POD screening tools identify delirium somewhat differently and assess different domains. Thus, different POD screening tools cause the reported incidence of POD to fluctuate across included studies, which is one reason why a consensus on the relationship between IOH and POD has not been reached.

Cerebral hypotension may be the most plausible mechanism by which IOH could result in POD. A reduction in cerebral blood flow can increase the probability of POD (48). The cerebral autoregulation system maintains stable cerebral blood flow over a relatively wide range as cerebral perfusion pressure varies. A sufficient decrease in arterial blood pressure could reach the lower boundary of cerebral autoregulation and result in a low cerebral blood flow (13). Different individuals may have different upper and lower limits of autoregulation. Besides, the limits of autoregulation are dynamic because of the multiple factors that influence it (49–51). These factors might also be a possible reason for the different results on the relationship between IOH and POD. Chen et al. found cerebral oximetry index-guided blood pressure management during CPB was associated with a reduced incidence and severity of POD following acute type A aortic dissection surgery (52). Their study suggested that monitoring the cerebral oximetry index might help to identify the patient-specific optimal blood pressure ranges that maintain adequate cerebral perfusion. Yamanoi et al. (53) found that a prolonged double-low period, defined as a bispectral index < 45 and a MAP <75 mmHg during general anesthesia, was independently associated with an increased incidence of POD in surgical ICU patients. Therefore, combining multimodal biomarkers might improve the individualized determination of cerebral autoregulation limits.

While cerebral hypoperfusion due to IOH represents one plausible mechanism, as discussed in the preceding paragraph, the recently proposed ‘Protective Hemodynamics’ strategy offers a complementary, hypothesis-generating perspective on the potential mechanisms underlying our results. This alternative view posits that hypotension may serve primarily as a marker of patients who will develop adverse events, rather than the direct cause. According to this paradigm, vasopressors administered to correct hypotension might induce excessive vasoconstriction in the splanchnic and cerebral vasculature, potentially leading to impaired organ perfusion and injury (54). Although RCTs included in our analysis support the association between IOH and POD, the limited number of available RCTs precludes definitive conclusions regarding causality. We cannot rule out the possibility that the aforementioned mechanism underlies our findings. Consequently, the primary clinical implication of our study, and a central question for future research, is to determine which therapeutic strategy is paramount for improving patient outcomes: rigorously avoiding hypotension per se, or minimizing exposure to vasopressors to reduce their potential detrimental effects. It is crucial to recognize that these two goals are not inherently mutually exclusive. Addressing this dilemma requires innovative clinical trial designs. Frameworks such as the C.L.E.A.R. approach, which operationalizes the principles of protective hemodynamics to simultaneously prevent profound hypotension and restrict vasopressor use, provide a conceptual basis for such future investigations (54).

This study has several limitations that should be considered when interpreting the findings. First, substantial heterogeneity was observed, particularly among the observational studies, which may limit the generalizability of the pooled results. This heterogeneity primarily stems from several methodological variations, primarily in the definitions of key variables. Specifically regarding IOH, the definitions varied in their core components, including the Blood Pressure Parameter and the Threshold Nature. Furthermore, among studies employing “Dichotomous Definitions,” there was inconsistency in the application of additional criteria for dichotomization. This methodological heterogeneity may reduce the persuasiveness of the pooled results, as differing definitions could influence the observed strength and direction of the association. Regarding POD assessment, it was not uniform across studies, employing different screening tools and assessment timings, which together could lead to outcome misclassification. Although a shorter screening window is pragmatically easier and may capture most episodes, it might not fully represent the entire postoperative course. In addition to the methodological heterogeneity discussed above, other unmeasured factors, such as variations in surgical populations, specific anesthesia protocols, and patient comorbidities, may have also contributed to the observed heterogeneity but could not be quantitatively assessed due to inconsistent reporting across the included studies. Finally, the current evidence base has constraints. The number of RCTs was limited, and most data were derived from observational studies that are susceptible to residual confounding. Furthermore, among the studies that quantified hypotension exposure using “Continuous Metrics,” the specific calculation methods were inconsistent across studies. Therefore, it is not currently possible to definitively establish a precise exposure-response relationship between the duration or depth of hypotension and the risk of POD based on the available evidence.

5 Conclusion

Our systematic review and meta-analysis found that IOH was statistically significantly associated with a higher risk of POD. However, the meta-analysis was limited by the numerous methodological differences among the included studies. Future research conducted in a standardized manner is needed to increase generalizability and facilitate easier interpretation of study results.

Data availability statement

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding authors.

Author contributions

XY: Data curation, Investigation, Project administration, Writing – original draft, Writing – review & editing. HZ: Data curation, Methodology, Writing – original draft, Writing – review & editing. QL (3rd author): Investigation, Methodology, Writing – review & editing. QL (4th author): Methodology, Writing – review & editing. HY: Methodology, Writing – review & editing. JL: Data curation, Supervision, Writing – review & editing.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This work was supported by the Science and Technology Project of Sichuan Provincial Health Commission (Clinical Research Project), grant number 23LCYJ024, and Sichuan Science and Technology Program grant number 23ZYKJCX0004.

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that Generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fmed.2025.1690490/full#supplementary-material

Abbreviations

POD, Postoperative delirium; IOH, Intraoperative hypotension; MAP, Mean arterial pressure; SBP, Systolic blood pressure; PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analysis; MeSH, Medical Subject Headings; RCTs, Randomized controlled trials; AUC, Area under the curve; RR, Risk ratio; OR, Odds ratio; CAM, Confusion Assessment Method; MMSE, Mini-Mental State Examination (MMSE); ICD, International Classification of Diseases; DSM-5, Diagnostic and statistical manual of mental disorders, 5th edition.

References

1. Evered, L, Silbert, B, Knopman, DS, Scott, DA, DeKosky, ST, Rasmussen, LS, et al. Recommendations for the nomenclature of cognitive change associated with anaesthesia and surgery-2018. Br J Anaesth. (2018) 121:1005–12. doi: 10.1016/j.bja.2017.11.087,

PubMed Abstract | Crossref Full Text | Google Scholar

2. Aldecoa, C, Bettelli, G, Bilotta, F, Sanders, RD, Audisio, R, Borozdina, A, et al. European Society of Anaesthesiology evidence-based and consensus-based guideline on postoperative delirium. Eur J Anaesthesiol. (2017) 34:192–214. doi: 10.1097/EJA.0000000000000594,

PubMed Abstract | Crossref Full Text | Google Scholar

3. Vasilevskis, EE, Han, JH, Hughes, CG, and Ely, EW. Epidemiology and risk factors for delirium across hospital settings. Best Pract Res Clin Anaesthesiol. (2012) 26:277–87. doi: 10.1016/j.bpa.2012.07.003,

PubMed Abstract | Crossref Full Text | Google Scholar

4. Abelha, FJ, Luís, C, Veiga, D, Parente, D, Fernandes, V, Santos, P, et al. Outcome and quality of life in patients with postoperative delirium during an ICU stay following major surgery. Crit Care. (2013) 17:R257. doi: 10.1186/cc13084,

PubMed Abstract | Crossref Full Text | Google Scholar

5. Zenilman, ME. Delirium: an important postoperative complication. JAMA. (2017) 317:77–8. doi: 10.1001/jama.2016.18174,

PubMed Abstract | Crossref Full Text | Google Scholar

6. Bramley, P, McArthur, K, Blayney, A, and McCullagh, I. Risk factors for postoperative delirium: an umbrella review of systematic reviews. Int J Surg. (2021) 93:106063. doi: 10.1016/j.ijsu.2021.106063,

PubMed Abstract | Crossref Full Text | Google Scholar

7. Mevorach, L, Forookhi, A, Farcomeni, A, Romagnoli, S, and Bilotta, F. Perioperative risk factors associated with increased incidence of postoperative delirium: systematic review, meta-analysis, and grading of recommendations assessment, development, and evaluation system report of clinical literature. Br J Anaesth. (2023) 130:e254–62. doi: 10.1016/j.bja.2022.05.032,

PubMed Abstract | Crossref Full Text | Google Scholar

8. Tang, X, Li, J, Yang, B, Lei, C, and Dong, H. Efficacy of sleep interventions on postoperative delirium: a systematic review and meta-analysis of randomized controlled trials. Anesthesiol Perioper Sci. (2023) 1:29. doi: 10.1007/s44254-023-00027-1

Crossref Full Text | Google Scholar

9. Hirsch, J, DePalma, G, Tsai, TT, Sands, LP, and Leung, JM. Impact of intraoperative hypotension and blood pressure fluctuations on early postoperative delirium after non-cardiac surgery. Br J Anaesth. (2015) 115:418–26. doi: 10.1093/bja/aeu458,

PubMed Abstract | Crossref Full Text | Google Scholar

10. Wachtendorf, LJ, Azimaraghi, O, Santer, P, Linhardt, FC, Blank, M, Suleiman, A, et al. Association between intraoperative arterial hypotension and postoperative delirium after noncardiac surgery: a retrospective multicenter cohort study. Anesth Analg. (2022) 134:822–33. doi: 10.1213/ane.0000000000005739,

PubMed Abstract | Crossref Full Text | Google Scholar

11. Yang, L, Sun, D-f, Han, J, Liu, R, Wang, L-j, and Zhang, Z-z. Effects of intraoperative hemodynamics on incidence of postoperative delirium in elderly patients: a retrospective study. Med Sci Monit. (2016) 22:1093–100. doi: 10.12659/MSM.895520,

PubMed Abstract | Crossref Full Text | Google Scholar

12. Cai, J, Tang, M, Wu, H, Yuan, J, Liang, H, Wu, X, et al. Association of intraoperative hypotension and severe postoperative complications during non-cardiac surgery in adult patients: a systematic review and meta-analysis. Heliyon. (2023) 9:e15997. doi: 10.1016/j.heliyon.2023.e15997,

PubMed Abstract | Crossref Full Text | Google Scholar

13. Lassen, NA. Cerebral blood flow and oxygen consumption in man. Physiol Rev. (1959) 39:183–238. doi: 10.1152/physrev.1959.39.2.183,

PubMed Abstract | Crossref Full Text | Google Scholar

14. Wu, ZS, Huang, R, Liang, LP, Zhu, YT, Zheng, QM, and Jiang, LP. Characteristics and influencing factors of postoperative delirium among elderly patients in intensive care unit. J Shanghai Jiaotong Univ. (2021) 41:1580–6. doi: 10.3969/j.issn.1674-8115.2021.12.006

Crossref Full Text | Google Scholar

15. Nakatani, S, Ida, M, Wang, X, Naito, Y, and Kawaguchi, M. Incidence and factors associated with postoperative delirium in patients undergoing transurethral resection of bladder tumor. JA Clin Rep. (2022) 8:6. doi: 10.1186/s40981-022-00497-5,

PubMed Abstract | Crossref Full Text | Google Scholar

16. Page, MJ, McKenzie, JE, Bossuyt, PM, Boutron, I, Hoffmann, TC, Mulrow, CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. (2021) 372:n71. doi: 10.1136/bmj.n71

Crossref Full Text | Google Scholar

17. Wells, GA, Wells, G, Shea, B, Shea, B, O'Connell, D, Peterson, J, et al. The Newcastle-Ottawa scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses: Ottawa Hospital Research Institute; 2014 Available online at: http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp

Google Scholar

18. Higgins, JPT, Altman, DG, Gøtzsche, PC, Jüni, P, Moher, D, Oxman, AD, et al. The Cochrane collaboration's tool for assessing risk of bias in randomised trials. BMJ. (2011) 343:d5928. doi: 10.1136/bmj.d5928,

PubMed Abstract | Crossref Full Text | Google Scholar

19. Borenstein, M, Hedges, LV, Higgins, JPT, and Rothstein, HR. A basic introduction to fixed-effect and random-effects models for meta-analysis. Res Synth Methods. (2010) 1:97–111. doi: 10.1002/jrsm.12,

PubMed Abstract | Crossref Full Text | Google Scholar

20. Marcantonio, ER, Goldman, L, Orav, EJ, Cook, EF, and Lee, TE. The association of intraoperative factors with the development of postoperative delirium. Am J Med. (1998) 105:380–4. doi: 10.1016/S0002-9343(98)00292-7,

PubMed Abstract | Crossref Full Text | Google Scholar

21. Patti, R, Saitta, M, Cusumano, G, Termine, G, and Di Vita, G. Risk factors for postoperative delirium after colorectal surgery for carcinoma. Eur J Oncol Nurs. (2011) 15:519–23. doi: 10.1016/j.ejon.2011.01.004,

PubMed Abstract | Crossref Full Text | Google Scholar

22. Tognoni, P, Simonato, A, Robutti, N, Pisani, M, Cataldi, A, Monacelli, F, et al. Preoperative risk factors for postoperative delirium (POD) after urological surgery in the elderly. Arch Gerontol Geriatr. (2011) 52:e166–9. doi: 10.1016/j.archger.2010.10.021,

PubMed Abstract | Crossref Full Text | Google Scholar

23. Guo, Z, Liu, J, Li, J, Wang, X, Guo, H, Ma, P, et al. Postoperative delirium in severely burned patients undergoing early escharotomy: incidence, risk factors, and outcomes. J Burn Care Res. (2017) 38:e370–6. doi: 10.1097/BCR.0000000000000397,

PubMed Abstract | Crossref Full Text | Google Scholar

24. Jiang, X, Chen, D, Lou, Y, and Li, Z. Risk factors for postoperative delirium after spine surgery in middle- and old-aged patients. Aging Clin Exp Res. (2017) 29:1039–44. doi: 10.1007/s40520-016-0640-4,

PubMed Abstract | Crossref Full Text | Google Scholar

25. Kim, S-w, Kim, N, Kim, J, Kim, K-M, and Lee, S. Risk factors for postoperative delirium following total knee arthroplasty in elderly patients. Anesth Pain Med. (2018) 13:143–8. doi: 10.17085/apm.2018.13.2.143

Crossref Full Text | Google Scholar

26. Lyu, W, Wang, Z, Yang, C, Wang, X, Feng, H, Jiao, Y, et al. Delirium following emergency surgery in patients with gastrointestinal obstruction or perforation: a retrospective cohort study for 5 years. J Gastrointest Surg. (2025):102083. doi: 10.1016/j.gassur.2025.102083

Crossref Full Text | Google Scholar

27. Wesselink, EM, Kappen, TH, van Klei, WA, Dieleman, JM, van Dijk, D, and Slooter, AJ. Intraoperative hypotension and delirium after on-pump cardiac surgery. Br J Anaesth. (2015) 115:427–33. doi: 10.1093/bja/aev256,

PubMed Abstract | Crossref Full Text | Google Scholar

28. Tobar, E, Abedrapo, MA, Godoy, JA, Llanos, JL, Díaz, MJ, Azolas, R, et al. Impact of hypotension and global hypoperfusion in postoperative delirium: a pilot study in older adults undergoing open colon surgery. Braz J Anesthesiol. (2018) 68:135–41. doi: 10.1016/j.bjan.2017.10.002,

PubMed Abstract | Crossref Full Text | Google Scholar

29. Wang, Y, Yu, H, Qiao, H, Li, C, Chen, K, and Shen, X. Risk factors and incidence of postoperative delirium in patients undergoing laryngectomy. Otolaryngol Head Neck Surg. (2019) 161:807–13. doi: 10.1177/0194599819864304,

PubMed Abstract | Crossref Full Text | Google Scholar

30. Langer, T, Santini, A, Zadek, F, Chiodi, M, Pugni, P, Cordolcini, V, et al. Intraoperative hypotension is not associated with postoperative cognitive dysfunction in elderly patients undergoing general anesthesia for surgery: results of a randomized controlled pilot trial. J Clin Anesth. (2019) 52:111–8. doi: 10.1016/j.jclinane.2018.09.021,

PubMed Abstract | Crossref Full Text | Google Scholar

31. Maheshwari, K, Ahuja, S, Khanna, AK, Mao, G, Perez-Protto, S, Farag, E, et al. Association between perioperative hypotension and delirium in postoperative critically ill patients: a retrospective cohort analysis. Anesth Analg. (2020) 130:636–43. doi: 10.1213/ane.0000000000004517,

PubMed Abstract | Crossref Full Text | Google Scholar

32. Wesselink, EM, Abawi, M, Kooistra, NHM, Kappen, TH, Agostoni, P, Emmelot-Vonk, M, et al. Intraoperative hypotension and delirium among older adults undergoing transcatheter aortic valve replacement. J Am Geriatr Soc. (2021) 69:3177–85. doi: 10.1111/jgs.17361,

PubMed Abstract | Crossref Full Text | Google Scholar

33. Narayanan, GK, Koshy, RC, Amma, RO, Subramanian, D, and Rajendran, SP. Incidence of postoperative delirium and its association with intraoperative blood pressure fluctuation in elderly undergoing oncosurgery: a prospective cohort study. J Clin Diagn Res. (2022) 16:UC39–45. doi: 10.7860/JCDR/2022/57305.16624

Crossref Full Text | Google Scholar

34. Ushio, M, Egi, M, Fujimoto, D, Obata, N, and Mizobuchi, S. Timing, threshold, and duration of intraoperative hypotension in cardiac surgery: their associations with postoperative delirium. J Cardiothorac Vasc Anesth. (2022) 36:4062–9. doi: 10.1053/j.jvca.2022.06.013,

PubMed Abstract | Crossref Full Text | Google Scholar

35. Duan, W, Zhou, C-M, Yang, J-J, Zhang, Y, Li, Z-P, Ma, D-Q, et al. A long duration of intraoperative hypotension is associated with postoperative delirium occurrence following thoracic and orthopedic surgery in elderly. J Clin Anesth. (2023) 88:111125. doi: 10.1016/j.jclinane.2023.111125,

PubMed Abstract | Crossref Full Text | Google Scholar

36. Mohr, NL, Krannich, A, Jung, H, Hulde, N, and von Dossow, V. Intraoperative blood pressure management and its effects on postoperative delirium after cardiac surgery: a single-center retrospective cohort study. J Cardiothorac Vasc Anesth. (2024) 38:1127–34. doi: 10.1053/j.jvca.2024.01.027,

PubMed Abstract | Crossref Full Text | Google Scholar

37. Zarour, S, Weiss, Y, Abu-Ghanim, M, Iacubovici, L, Shaylor, R, Rosenberg, O, et al. Association between intraoperative hypotension and postoperative delirium: a retrospective cohort analysis. Anesthesiology. (2024) 141:707–18. doi: 10.1097/aln.0000000000005149,

PubMed Abstract | Crossref Full Text | Google Scholar

38. Singh, M, Spence, J, Shah, K, Duncan, AE, Kimmaliardjuk, D, Sessler, DI, et al. Intraoperative high and low blood pressures are not associated with delirium after cardiac surgery: a retrospective cohort study. J Clin Anesth. (2025) 100:111686. doi: 10.1016/j.jclinane.2024.111686,

PubMed Abstract | Crossref Full Text | Google Scholar

39. Wang, Y, Chen, K, Ye, M, and Shen, X. Intraoperative hypotension and postoperative delirium in elderly male patients undergoing laryngectomy: a single-center retrospective cohort study. Braz J Anesthesiol. (2025) 75:844560. doi: 10.1016/j.bjane.2024.844560,

PubMed Abstract | Crossref Full Text | Google Scholar

40. Williams-Russo, P, Sharrock, NE, Mattis, S, Liguori, GA, Mancuso, C, Peterson, MG, et al. Randomized trial of hypotensive epidural anesthesia in older adults. Anesthesiology. (1999) 91:926–35. doi: 10.1097/00000542-199910000-00011

Crossref Full Text | Google Scholar

41. Siepe, M, Pfeiffer, T, Gieringer, A, Zemann, S, Benk, C, Schlensak, C, et al. Increased systemic perfusion pressure during cardiopulmonary bypass is associated with less early postoperative cognitive dysfunction and delirium. Eur J Cardiothorac Surg. (2011) 40:200–7. doi: 10.1016/j.ejcts.2010.11.024,

PubMed Abstract | Crossref Full Text | Google Scholar

42. Xu, X, Hu, X, Wu, Y, Li, Y, Zhang, Y, Zhang, M, et al. Effects of different BP management strategies on postoperative delirium in elderly patients undergoing hip replacement: a single center randomized controlled trial. J Clin Anesth. (2020) 62:109730. doi: 10.1016/j.jclinane.2020.109730,

PubMed Abstract | Crossref Full Text | Google Scholar

43. Hu, A-M, Qiu, Y, Zhang, P, Zhao, R, Li, S-T, Zhang, Y-X, et al. Higher versus lower mean arterial pressure target management in older patients having non-cardiothoracic surgery: a prospective randomized controlled trial. J Clin Anesth. (2021) 69:110150. doi: 10.1016/j.jclinane.2020.110150,

PubMed Abstract | Crossref Full Text | Google Scholar

44. Zhang, Y, Zhang, Y, Zhou, Z, Sang, X, Qin, M, Dai, G, et al. Higher intraoperative mean arterial blood pressure does not reduce postoperative delirium in elderly patients following gastrointestinal surgery: a prospective randomized controlled trial. PLoS One. (2022) 17:e0278827. doi: 10.1371/journal.pone.0278827,

PubMed Abstract | Crossref Full Text | Google Scholar

45. Feng, X, Hu, J, Hua, F, Zhang, J, Zhang, L, and Xu, G. The correlation of intraoperative hypotension and postoperative cognitive impairment: a meta-analysis of randomized controlled trials. BMC Anesthesiol. (2020) 20:193. doi: 10.1186/s12871-020-01097-5,

PubMed Abstract | Crossref Full Text | Google Scholar

46. Wijnberge, M, Schenk, J, Bulle, E, Vlaar, AP, Maheshwari, K, Hollmann, MW, et al. Association of intraoperative hypotension with postoperative morbidity and mortality: systematic review and meta-analysis. BJS Open. (2021) 5:zraa018. doi: 10.1093/bjsopen/zraa018,

PubMed Abstract | Crossref Full Text | Google Scholar

47. Wickham, AJ, Highton, DT, Clark, S, Fallaha, D, Wong, DJN, and Martin, DS. Treatment threshold for intra-operative hypotension in clinical practice-a prospective cohort study in older patients in the UK. Anaesthesia. (2022) 77:153–63. doi: 10.1111/anae.15535,

PubMed Abstract | Crossref Full Text | Google Scholar

48. Yu, Q, Qi, J, and Wang, Y. Intraoperative hypotension and neurological outcomes. Curr Opin Anaesthesiol. (2020) 33:646–50. doi: 10.1097/ACO.0000000000000904,

PubMed Abstract | Crossref Full Text | Google Scholar

49. Donnelly, J, Aries, MJ, and Czosnyka, M. Further understanding of cerebral autoregulation at the bedside: possible implications for future therapy. Expert Rev Neurother. (2015) 15:169–85. doi: 10.1586/14737175.2015.996552,

PubMed Abstract | Crossref Full Text | Google Scholar

50. Reich, DL, and Levin, MA. Chasing the elusive cerebral autoregulation curve in pursuit of intraoperative brain protection. J Thorac Cardiovasc Surg. (2017) 154:1599–600. doi: 10.1016/j.jtcvs.2017.05.055,

PubMed Abstract | Crossref Full Text | Google Scholar

51. Barnes, SC, Beishon, LC, Hasan, MT, Robinson, TG, and Minhas, JS. Cerebral haemodynamics, anaesthesia and the frail brain. Anesthesiol Perioper Sci. (2023) 1:19. doi: 10.1007/s44254-023-00020-8

Crossref Full Text | Google Scholar

52. Chen, X, Xie, W, Li, W, and Gao, J. Cerebral oximetry index-guided blood pressure management during cardiopulmonary bypass reduces postoperative delirium in patients with acute type a aortic dissection. J Cardiothorac Vasc Anesth. (2025) 39:1666–77. doi: 10.1053/j.jvca.2025.01.003,

PubMed Abstract | Crossref Full Text | Google Scholar

53. Yamanoi, T, Suzuki, S, Kaku, R, and Morimatsu, H. Prolonged double-Low time and the incidence of postoperative delirium in surgical ICU patients. Acta Med Okayama. (2023) 77:161–7. doi: 10.18926/AMO/65145,

PubMed Abstract | Crossref Full Text | Google Scholar

54. D'Amico, F, and Landoni, G. Protective hemodynamics: a novel strategy to manage blood pressure. Curr Opin Crit Care. (2024) 30:629–36. doi: 10.1097/MCC.0000000000001205,

PubMed Abstract | Crossref Full Text | Google Scholar