The relationship between intraoperative hypotension and acute kidney injury in elderly and super elderly patients undergoing noncardiac surgery: a retrospective cohort analysis

Abstract

Background

Intraoperative hypotension has been identified as a significant factor associated with postoperative renal dysfunction. While numerous studies have reported on the relationship between incidence of postoperative acute kidney injury (AKI) and intraoperative hypotension, limited research has specifically focused on elderly and super elderly patients.

Methods

This retrospective cohort study comprised subjects aged 60 years and older who underwent elective non-cardiac surgery under general anesthesia. AKI was defined in accordance with established clinical criteria based on changes in serum creatinine concentration. We explored potential harm thresholds for the lowest intraoperative mean arterial pressure (MAP) and the largest MAP reduction from baseline. Logistic regression with restricted cubic splines was conducted to assess the relationships between AKI and both the lowest intraoperative MAP or the largest MAP reduction from baseline across elderly patient subgroups aged 60–69, 70–79, 80–89, and 90 + years.

Results

A total of 10,859 patients were included in the final analysis. The overall incidence of AKI in this population was 2.77% (301 out of 10,859). Specifically, the incidence rates among patients aged 60–69, 70–79, and 80–89 years were 2.10%, 3.13%, and 4.68%, respectively. The lowest MAP and the incidence of postoperative AKI exhibited a non-linear relationship, both univariably and multivariably (p = 0.0005 and p = 0.0215, respectively), suggesting that the absolute MAP was associated with the risk of AKI. The predicted minimum incidence of AKI (probability = 2.31%) occurred at a lowest MAP of 86 mmHg. Within the MAP range from 74 to 102 mmHg, the incidence of AKI was below the population average. The largest reduction in MAP from baseline was not significantly associated with the probability of AKI.

Conclusions

This study shows that the risk of postoperative AKI was strongly correlated with absolute intraoperative hypotension in both the elderly and super elderly populations. In older adults, maintaining a stable absolute intraoperative MAP should be considered to prevent intraoperative renal injury.

Trial registration

This study was registered in Chinese Clinical Trial Registry with the registration number ChiCTR2400094990 on December 31, 2024.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12877-025-06944-z.

Introduction

According to World Health Organization (WHO) citing UNDESA projections (WHO, 2022; UNDESA, 2022), the population aged 60 years and older is increasing. In 2019, the number was 1 billion and is projected to reach 1.4 billion by 2030 and 2.1 billion by 2050. This increase is occurring at an unprecedented pace and is expected to accelerate in the coming decades, particularly in developing countries. The demand for surgery among elderly patients is also increasing [1, 2]. Due to the limited regenerative capacity of the kidney, the impact of acute kidney injury (AKI) in the elderly is significantly greater than that in younger individuals [3]. Older patients with several comorbidities, such as diabetes, hypertension, and heart failure, are more likely to develop AKI [4–7].

Intraoperative hypotension occurs frequently during surgery and has been reported to be associated with organ ischemia, which can lead to AKI [8, 9]. Several critical knowledge gaps lead to uncertainty regarding when and how to intervene to prevent perioperative hypotension-associated acute AKI [10, 11]. Although it is known that the risk of AKI varies with both the severity and duration of hypotensive episodes, ‘safe’ levels of arterial pressure have not yet been established [10, 12].

In a retrospective cohort analysis, mean arterial pressure (MAP) below an absolute threshold of 65 mmHg [13] or a relative threshold of 20% was progressively associated with both myocardial and kidney injury [14]. However, this threshold has not yet been well established [15], and most anesthesiologists may prefer to maintain MAP within 10% of baseline in elderly patients during anesthesia [16]. ‌The extent of hypotension’s contribution to postoperative kidney injury is uncertain; moreover, the influence of aging on this complication remains poorly understood [9, 13, 17]. Research conducted by the Cleveland Clinic shows that when MAP is below 55 mmHg for 5 min, the risk of AKI is 30% higher compared to patients without episodes of hypotension [18]. Intraoperative hypotension may contribute to postoperative AKI and myocardial injury (MI) [15]; however, the relationship between intraoperative hypotension and AKI in individuals aged 60 years and older undergoing noncardiac surgery remains unclear [10, 19, 20].

Therefore, we hypothesized that intraoperative hypotension is associated with postoperative AKI in individuals aged 60 years and older. The primary objective of this retrospective cohort study was to quantify the relationship between absolute or relative intraoperative hypotension and postoperative AKI in individuals aged 60 years and older, with a secondary aim of determining the critical threshold for renal injury.

Materials and methods

Data source and participants

This was a single-center retrospective cohort study approved by the Ethics Committee of the First Affiliated Hospital of Soochow University (LunShen [2024]704) and was registered in the Chinese Clinical Trial Registry (www.medicalresearch.org.cn, ChiCTR2400094990) on December 31, 2024. The study design adhered to the STROBE 2007 statement. ‌Patients who underwent noncardiac surgery under general anesthesia between 2020 and 2024 were included in the study. We used the electronic system of the First Affiliated Hospital of Soochow University (Haitai system, Surgical anesthesia system, Perioperative scientific research data platform) for data collection. Inclusion criteria: patients aged 60 years and older who underwent noncardiac surgery under general anesthesia; intraoperative invasive arterial MAP monitoring; serum creatinine (Scr) measured before surgery and within 7 days postoperatively; surgical duration exceeding 60 min; at least three consecutive MAP measurements obtained before anesthesia induction. Medical conditions were classified according to the International Classification of Diseases (ICD-9 or ICD-10), and medication names were coded using Anatomical Therapeutic Chemical (ATC 2010) classification (Supplementary Material 1).

Exclusion criteria were as follows: (1) preoperative chronic renal insufficiency, defined as an estimated glomerular filtration rate (eGFR) of below 60 ml min^− 1 1.73 m^− 2 or requiring dialysis; (2) urological surgery, including procedures for relief of urinary obstruction, nephrectomy, or renal transplantation; (3) persistent hypotension requiring continuous use of vasopressor agents during the perioperative period; (4) decompensation of other vital organ systems; (5) intravenous administration of high-dose aminoglycosides and other nephrotoxic agents during the perioperative period; (6) total parenteral nutrition during the perioperative period.

In this retrospective cohort analysis, we controlled for potential confounding factors [21, 22] through covariate adjustment. The main indicators related to renal function status (urate, urea, cystatin C) [23] were involved, along with commonly used perioperative antibacterial agents (cephalosporins, penicillins, quinolones, vancomycin), the commonly used antihypertensive agents (beta-blockers, calcium channel blockers, etc.) [24], and prevalent comorbidities in elderly patients during the perioperative period (cardiovascular and cerebrovascular diseases, etc.) [25]. To minimize the risk of omitting potential confounding factors, a mandatory variable inclusion approach was employed.

The Kidney Disease Improving Global Outcomes (KDIGO) defined AKI as follows (ungraded): an increase in Scr of ≥ 0.3 mg dl^{− 1} (≥ 26.5 µmol l^{− 1}) within 48 h, or an increase in Scr to 1.5 times or more above baseline, known or presumed to have occurred within the prior 7 days; or urine output < 0.5 ml kg^{− 1} h^{− 1} for 6 h (Note: Due to unreliable measurement of urine output, this criterion is excluded in many clinical settings). We defined the occurrence of AKI as a dichotomous variable, coded as “YES” or “NO”. Perioperative MAP was recorded in each patient’s anesthesia record both noninvasively and invasively. Before the induction of anesthesia, MAP was collected at intervals of more than 2 min for 3 times to calculate the mean value as a baseline. Invasive arterial MAP was recorded every minute during the operation, and the lowest MAP over 5 consecutive minutes was collected during the procedure. Since there are many definitions of intraoperative hypotension [9, 14, 17, 20], absolute hypotension was defined as the lowest intraoperative MAP sustained for 5 consecutive minutes. In this study, relative hypotension was defined as the maximum percentage reduction in MAP from the preoperative baseline, also maintained for a continuous period of 5 min.

All noninvasive and invasive blood pressure measurements obtained at intervals ranging from 1 to 5 min were recorded in the electronic anesthesia and surgical management system of the research center. An artifact removal algorithm was used to clean the raw blood pressure monitoring data based on the range and abrupt change (Supplementary Material 2). Linear interpolation was applied for each missing blood pressure measurement. Finally, data reliability was ensured through manual correction and expert review. Our primary objective was to investigate the relationships between absolute hypotension or relative hypotension and AKI, with a secondary aim of determining the critical MAP thresholds associated with AKI risk.

Statistical methods

For summary statistics, continuous variables were reported as means ± standard deviations (SDs) or medians [interquartile range: quartile 1, quartile 3], as appropriate; categorical variables were reported as frequencies and proportions. Differences in continuous variables were evaluated using the t-test for normally distributed data or the Mann-Whitney U test for non-normally distributed data. For categorical variables, Pearson’s chi-square test or Fisher’s exact test (expected frequencies < 5) was used. Missing data were removed from the analysis. Summary statistics were conducted in R 4.4.1 using the stats package.

The primary outcome was prespecified in the study protocol (registered at the Chinese Clinical Trials Registry, ChiCTR2400094990, on December 31, 2024) prior to participant enrollment. The primary outcome was the prediction of AKI probability based on the absolute lowest MAP or largest MAP reduction from baseline for 5 consecutive minutes in individuals aged 60 years and older. To reduce noise in the data, a univariate smoothed moving average plot was performed to show the crude AKI probability at smoothed exposure point with a window size of 5 (k = 5). We then fitted the lowest MAP or largest MAP reduction and AKI incidence using a logistic regression model with restricted cubic splines with three knots (10th, 50th, and 90th percentiles), both univariably and multivariably. Adjusted confounding variables include age, sex, BMI, ASA levels, surgery duration, intraoperative use of phenylephrine or norepinephrine, preoperative use of cephalosporins, penicillins, quinolones, aminoglycosides, or vancomycin, preoperative comorbidities (hypertension, cerebral infarction, cardiac insufficiency, Alzheimer’s disease, Parkinson’s disease), preoperative uric acid and urea, blood loss, crystal volume, plasma substitute, and blood transfusion. Due to the importance of age, we divided the individuals into four groups: 60–69 (n = 5456), 70–79 (n = 4152), 80–89 (n = 1090), and 90 or older (n = 161) to study MAP-AKI associations. A smoothed moving average plot was conducted in R 4.4.1 using the zoo package version 1.8–12, along with logistic regression with restricted cubic splines using the R packages splines and rms version 7.0–0. Marginal effect analysis was performed using the R package marginaleffects version 0.30.0.

Post hoc analysis

We conducted a sensitivity analysis on the exclusion criteria for AKI incidence and their effects on the association between hypotension and AKI incidence.

Sample size considerations

A priori, we estimated that a sample size of 1,269 would achieve a significance level (α) of 0.05 and greater than 90% power (1 – β) when using a nonlinear logistic regression model to evaluate the association between AKI and MAP. This calculation was based on an assumed small Cohen’s f² effect size of 0.01. We screened 82,768 patient records; of these, 10,859 met the inclusion/exclusion criteria and had complete primary data, thus forming the final analytic cohort.

We performed a post hoc analysis to evaluate the power of the actual sample size for the AKI-lowest MAP association models. With all individuals aged 60 years and older and the age-stratified groups, we had a power of 0.9 or more, except for the 90 years and older group due to low sample size (n = 161). Power analysis was conducted using the pwr package in R.

Results

Between January 2020 and December 2024, a total of 12,000 patients meeting the inclusion/exclusion criteria for elective noncardiac surgery were screened at the First Affiliated Hospital of Soochow University (Suzhou, China) (Fig. 1). Among them, 1,141 patients were excluded due to missing intraoperative MAP or an unsuitable number of repetitions. Finally, 10,859 patients were included in this analysis. The general characteristics of the patients, preoperative conditions, intraoperative conditions, as well as perioperative concomitant medications and complications were systematically recorded and categorized according to AKI status (AKI group vs. non-AKI group) (Table 1). Age, sex, ASA, surgery duration, blood loss, and preoperative renal function (uric acid, urea, and cystatin) related variables were associated with AKI; baseline MAP was not significantly different between the AKI and non-AKI group. The incidence of AKI following noncardiac surgery in patients aged 60 years and older was 2.77% (301 out of 10,859). The incidence rates among patients aged 60–69, 70–79, and 80–89 years were 2.10%, 3.13%, and 4.68%, respectively.

Fig. 1.

Study flowchart. Patients from 2020–2024 in the First Affiliated Hospital of Soochow University (N = 165,536)

Table 1.

Baseline, intraoperative and postoperative characteristics by AKI (N = 10,859)

	AKI (N = 301)	non-AKI (N = 10,558)	Odds Ratio (95% CI)	P value
Age, yr	72 [67, 78]	69 [65, 75]	2 [1, 3]	< 0.001
Age group				< 0.001
60–69 yr	115 (38.2)	5341 (50.6)	Reference
70–79 yr	130 (43.2)	4022 (38.1)	0.67 [0.52, 0.86]
80–89 yr	51 (16.9)	1039 (9.8)	0.44 [0.31, 0.61]
90 + yr	5 (1.7)	156 (1.5)	0.67 [0.27, 1.67]
Female	110 (36.5)	5259 (49.8)	1.72 [1.35, 2.21]	< 0.001
BMI, kg m− 2	23.4 [21.5, 26.1]	23.3 [21.0, 25.4]	0.46 [0, 0.92]	0.05
ASA classification				< 0.001
I	4 (1.3)	360 (3.4)	Reference
II	188 (62.5)	8339 (79)	0.49 [0.18, 1.33]
III	96 (31.9)	1820 (17.2)	0.21 [0.08, 0.58]
VI	13 (4.3)	39 (0.4)	0.03 [0.01, 0.11]
Preoperative variables
Baseline MAP, mmHg	100 ± 16	100 ± 15	−0.42−2.26, 1.41]	0.649
Baseline noninvasive MAP, mmHg	101 [92, 111]	102 [93, 110]	0 [−2, 2]	0.895
Uric acid, µmol l− 1	316.1 [246.3, 409.3]	303.9 [245.5, 369.8]	16.65 [3.9, 29.4]	0.011
Urea, mmol l− 1	6.6 [5.2, 8.3]	5.7 [4.7, 7]	0.85 [0.6, 1.1]	< 0.001
Creatinine, µmol l− 1	71.9 [57.9, 92.5]	62.2 [52.5, 73.7]	10.1 [7.6, 12.6]	< 0.001
Cystatin C, mg l− 1	1.205 [0.99, 1.5025]	0.99 [0.88, 1.13]	0.2 [0.17, 0.23]	< 0.001
Preoperative antibiotics use
Cephalosporins	166 (55.1)	6794 (64.3)	1.47 [1.16, 1.86]	0.001
Penicillins	55 (18.3)	1032 (9.8)	0.48 [0.36, 0.67]	< 0.001
Quinolones	7 (2.3)	269 (2.5)	1.1 [0.52, 2.78]	0.955
Aminoglycosides	2 (0.7)	19 (0.2)	0.27 [0.06, 2.4]	0.114
Vancomycin	20 (6.6)	604 (5.7)	0.85 [0.54, 1.43]	0.58
Blood pressure medications
Dihydropyridines	8 (2.7)	388 (3.7)	1.4 [0.69, 3.29]	0.44
Sartans	10 (3.3)	326 (3.1)	0.93 [0.49, 1.97]	0.95
ACEI	0 (0)	24 (0.2)	Inf	1
Thiazide diuretics	5 (1.7)	102 (1)	0.58 [0.24, 1.83]	0.224
Beta-blockers	0 (0)	36 (0.3)	Inf	0.626
Preoperative comorbidity
Hypertension	244 (81.1)	8962 (84.9)	1.31 [0.96, 1.77]	0.082
Cerebral infarction	9 (3)	263 (2.5)	0.83 [0.42, 1.85]	0.719
Parkinson’s disease	3 (1)	110 (1)	1.05 [0.35, 5.18]	1
Cardiac insufficiency	5 (1.7)	24 (0.2)	0.13 [0.05, 0.46]	0.001
Alzheimer’s disease	0 (0)	5 (0.05)	Inf	1
Postoperative complications
Coronary heart disease	2 (0.7)	14 (0.1)	0.2 [0.05, 1.81]	0.071
Cerebrovascular disease	7 (2.3)	111 (1.1)	0.45 [0.21, 1.15]	0.046
Gastrointestinal dysfunction	18 (6)	168 (1.6)	0.25 [0.15, 0.45]	< 0.001
Duration of surgery, min	190 [125, 285]	165 [115, 230]	25 [15, 35]	< 0.001
Intraoperative variables
Sufentanil, µg	50 [30, 60]	50 [35, 50]	−0.00001 [−0.00008, 0.00006]	0.979
Ephedrine, µg	0 [0, 0]	0 [0, 0]	0.00001 [−0.00002, 0.00004]	0.731
Phenylephrine, µg	0 [0, 50]	0 [0, 0]	0.00003 [0.00005, 0]	< 0.001
Norepinephrine, µg	33 (11)	303 (2.9)	0.24 [0.16, 0.36]	< 0.001
Atropine, mg	0 [0, 0]	0 [0, 0]	−0.00002 [−0.00005, 0.00001]	0.522
Blood loss, ml	200 [65, 400]	100 [50, 200]	75 [50, 100]	< 0.001
Crystal volume, ml	500 [500, 1000]	500 [500, 1000]	0 [0, 0]	0.035
Plasma substitute, ml	500 [500, 1000]	500 [500, 500]	0 [0, 0]	< 0.001
Blood transfusion volume, ml	0 [0, 600]	0 [0, 0]	0 [0, 0]	< 0.001
Urine volume, ml	300 [187.5, 500]	300 [0, 500]	25 [−0.00004, 50]	0.096

 Data are presented as mean ± SD, median [25th, 75th percentiles] or n (%)

P values are derived from Pearson chi-square test, Fisher chi-square test, Welch’s t-test, or Mann-Whitney U test, as appropriate

AKI Acute kidney injury, BMI Body mass index, ASA American Society of Anesthesiologists, MAP Mean arterial pressure, ACEI Angiotensin-converting enzyme inhibitor

Data are presented as mean ± SD, median [25th, 75th percentiles] or n (%). P values are derived from Pearson chi-square test, Fisher chi-square test, Welch’s t-test, or Mann-Whitney U test, as appropriate. AKI, acute kidney injury; BMI, body mass index; ASA, American Society of Anesthesiologists; MAP, mean arterial pressure; ACEI, angiotensin-converting enzyme inhibitor.

As expected, patients who developed AKI were significantly older than patients without AKI (P < 0.001, Table 1). Univariable logistic regression showed a positive linear correlation between age and AKI probability (P < 0.0001), with an odds ratio and 95% confidence interval: (1.04, 1.02 to 1.06) (Supplementary Figure S1).

We first explored the raw association between absolute and relative MAP and AKI incidence in patients aged 60 years and older using line charts. We observed potential trends univariately (Supplementary Figure S2). We then performed logistic regression with restricted cubic splines to fit MAP and AKI incidence (Fig. 2). Lowest MAP and the incidence of postoperative AKI exhibited a non-linear relationship, both univariately and multivariately (p = 0.0005 and p = 0.0215, respectively), suggesting that the lowest MAP level was associated with the risk of AKI (Table 2, Supplementary Table S1). The predicted minimum incidence of AKI (probability = 2.31%) occurred at a lowest MAP of 86 mmHg. Within the MAP range from 74 to 102 mmHg, the incidence of hypotension was below the population average (Supplementary Table S2). We next performed marginal effect analysis, suggesting that as the lowest MAP (< 86 mmHg threshold) decreased, the absolute change in AKI risk significantly increased. In contrast, when the lowest MAP above the threshold increased, the change in AKI risk was not significantly different (Supplementary Material 3). In addition, the spline terms of the largest MAP reduction from baseline were not statistically significant (Table 2, Supplementary Table S1).

Fig. 2.

Relationship between lowest MAP or largest MAP reduction and AKI in the 60 + elderly. The correlations between (A) lowest MAP or (B) largest MAP reduction from baseline and AKI were generated using a logistic regression model with restricted cubic splines of exposure at the 10th, 50th, and 90th percentiles. Error bars represent 95% confidence interval representing prediction

Table 2.

Associations between lowest MAP or largest MAP reduction from baseline and AKI

	Unadjusted		Adjusted
	Chi-Square	P value	Chi-Square	P value
Lowest MAP (mmHg)	15.2	0.0005	7.68	0.0215
Largest MAP reduction from baseline (%)	3.36	0.189	1.39	0.4985

Results of univariate (unadjusted) and multivariate (adjusted) logistic regression models with restricted cubic splines were calculated using ANOVA. For multivariate logistic regression models, fixed effects include age, sex, BMI, ASA levels, surgery duration, intraoperative use of phenylephrine or norepinephrine, preoperative use of cephalosporins, penicillins, quinolones, aminoglycosides, or vancomycin, preoperative comorbidities (hypertension, cerebral infarction, cardiac insufficiency, Alzheimer’s disease, Parkinson’s disease), preoperative uric acid and urea, blood loss, crystal volume, plasma substitute, and blood transfusion

AKI Acute kidney injury, MAP Mean arterial pressure

Results of univariate (unadjusted) and multivariate (adjusted) logistic regression models with restricted cubic splines were calculated using ANOVA. For multivariate logistic regression models, fixed effects include age, sex, BMI, ASA levels, surgery duration, intraoperative use of phenylephrine or norepinephrine, preoperative use of cephalosporins, penicillins, quinolones, aminoglycosides, or vancomycin, preoperative comorbidities (hypertension, cerebral infarction, cardiac insufficiency, Alzheimer’s disease, Parkinson’s disease), preoperative uric acid and urea, blood loss, crystal volume, plasma substitute, and blood transfusion. AKI, acute kidney injury; MAP, mean arterial pressure.

Unadjusted smooth moving-average plots for the lowest MAP are shown for AKI by different age ranges in Fig. 3. We observed potential associations as age ranges increase. We performed logistic regression using restricted cubic splines by age group (Fig. 4). With the increase in age levels, the nonlinear relationship between the lowest MAP and AKI incidence became more significant (except for the 90 + group, possibly due to a small sample size of 161), indicating the importance of controlling the lowest MAP in the elderly population (Supplementary Table S3, Supplementary Material 4). The largest MAP reduction from baseline was not significantly associated with AKI incidence across different age groups (Supplementary Figure S2, Supplementary Table S4).

Fig. 3.

Smooth moving average plot of AKI incidence and lowest MAP for a cumulative 5 min stratified by age group

Fig. 4.

Univariate logistic regression model with restricted cubic splines of exposure at the 10th, 50th, and 90th percentiles stratified by age group

Post hoc analysis

Given the low incidence of AKI observed in our dataset, we conducted a sensitivity analysis to evaluate the exclusion criteria. These criteria include preoperative chronic renal insufficiency, urological surgery, persistent hypotension requiring continuous vasopressor use during the perioperative period, decompensation of other vital organ functions, intravenous administration of high-dose aminoglycosides and other nephrotoxic agents during the perioperative period, and receipt of total parenteral nutrition during the perioperative period. We found that chronic renal insufficiency, among the exclusion criteria, was the most influential factor potentially contributing to the underestimation of AKI incidence (Supplementary Material 5). After additional adjustment for chronic renal insufficiency, the results remained consistent with the primary analysis, showing a significant nonlinear correlation between the lowest MAP and AKI incidence in both univariate and multivariate models, whereas the largest MAP reduction from baseline was not significantly associated with the probability of AKI (Supplementary Material 6).

Discussion

The occurrence of AKI severely impairs postoperative recovery as well as long-term health in older patients. Here we conducted a retrospective cohort study comprised patients aged 60 years and older who underwent elective non-cardiac surgery under general anesthesia to explore the relationship between intraoperative blood pressure and postoperative AKI incidence. Using logistic regression with restricted cubic splines, we demonstrated a non-linear association between lowest MAP and AKI incidence, with a minimum AKI incidence at 86 mmHg. Largest MAP reduction from baseline was not associated with AKI incidence. Thus, maintaining a stable absolute intraoperative MAP is important for preventing intraoperative renal injury in elderly and super elderly patients.

The overall incidence of AKI following noncardiac surgery was found to be 2.77% (301 out of 10,859) in individuals aged 60 years and older in our center. In a meta-analysis examining elderly patients who underwent hip replacement surgery, the incidence of postoperative AKI was 6.3% [18]. The observed discrepancy may be attributed to several factors, including differences in baseline characteristics, preoperative status, and MAP management. To avoid confounding factors in our analysis, we excluded several independent risk factors for postoperative AKI, including chronic kidney disease, urological surgery, unstable preoperative BP, and patients receiving high-dose nephrotoxic medications (Fig. 1). In a large multicenter cohort study with 112,912 qualified surgeries that met al.l inclusion and exclusion criteria, the overall incidence of AKI in the study population was 2.8% (n = 3146) [26]. It is consistent with our research finding.

Our study demonstrated a significant relationship between intraoperative absolute hypotension (lowest MAP) and postoperative AKI in individuals aged 60 years and older. MAP reduction from baseline was not statistically significant, suggesting that instead of relative hypotension (largest MAP reduction), absolute hypotension thresholds may have greater clinical relevance in this population. Schacham et al. [27] reported there is no relationship between hypotension (absolute or relative) and AKI in pediatric populations. These differences may reflect age-related divergences in renal autoregulation, as elderly patients exhibit diminished vascular compliance and impaired compensatory mechanisms, making the elderly population more vulnerable to absolute pressure drops. A multicenter study across eight hospitals also reported the importance of absolute hypotension in AKI risk among adult patients undergoing noncardiac surgery, particularly with higher ASA class and longer surgery duration [28].

Age-stratified analysis revealed a stronger nonlinear relationship between lowest MAP and AKI incidence with age increasing (except in the 90 + group), likely due to small sample size [n = 161], highlighting the importance of controlling absolute hypotension in older adults. Aging acts as an effect modifier, necessitating age-specific intraoperative BP targets.

The lowest predicted AKI incidence (2.31%) occurred at a MAP of 86 mmHg. Within the lowest MAP range from 74 to 102 mmHg, the incidence of AKI was below the population average. For individuals aged 60 and older, a supraphysiological intraoperative MAP may not be necessary if adequate perfusion is maintained. Observations from ICU-based studies [29, 30] have suggested that MAP alone might not serve as an adequate therapeutic target for AKI. In patients with AKI progression, mean perfusion pressure can be a potential risk factor for exacerbating disease progression. In intensive care settings, this threshold appears to be considerably higher-around 90 mmHg. The threshold for harm initiation on the surgical ward remains uncertain [30], likely falls within an intermediate range. These results suggest that transient relative hypotension during surgery in elderly patients may not require urgent vasopressor use, as vasoactive drugs can directly impair renal function directly, and renal protection should focus on maintaining absolute blood pressure above critical thresholds.

The small sample size of super elderly patients (≥ 90 years old) (n = 161) may affect the generalizability of the results. Similar study [31] also has an insufficient sample size for the elderly subgroup, which illustrates that this is a common problem in the field. The low proportion of the super elderly patients(≥ 90 years old) may reflect the target population’s characteristics (e.g., fewer surgical patients in this age group), not sampling bias. While our study assessed the relationship between 5-minute hypotensive episodes and AKI risk, the cumulative effects of prolonged intraoperative hypotension (e.g., > 10 min) on postoperative AKI risk in adults aged 60 and older remain uncertain [32]. Methodologically, a ROC analysis showed the 10-minute window has a slightly higher AUC (0.89) than the 5-minute window (0.82) [33]. A 5-minute window’s advantage lies in capturing brief but critical blood pressure fluctuations. It is important to highlight that the 5-minute protocol may be considered for surgeries lasting over 60 min and in patients excluding ASA grade IV [21, 34].

Confounding and bias are inherent limitations of any observational analysis despite our efforts to adjust for potential confounders. As a single-center retrospective analysis, our study confirmed the relationships between intraoperative hypotension and postoperative AKI risk. However, like general single-center studies, they are usually limited by specific geographical, population characteristics (especially ≥ 90 years old), single sample source and limited coverage, which leads to limited extrapolation of conclusions.

Future multicenter studies should expand sample sizes and hypotension duration thresholds, and incorporate advanced hemodynamic phenotyping (e.g., MAP variability indices, tissue oxygen saturation), novel urinary biomarkers (e.g., NGAL/KIM-1), and multiorgan injury profiling [35]. As a laboratory diagnostic parameter for AKI, serum creatinine also has limitations because it is dependent on muscle mass and may lead to incorrect or delayed diagnosis in certain patient groups [36]. Furthermore, integrating pharmacogenomic data with machine learning-based waveform analysis will enable precise delineation of dose-dependent renal vulnerability to intraoperative hypotension. Future research should also focus on identifying patient-specific and organ-specific hypotension harm thresholds and optimal treatment strategies for intraoperative hypotension including choice of vasopressors.

Conclusion

In conclusion, this retrospective cohort study revealed that the risk of postoperative AKI was strongly correlated with absolute intraoperative hypotension in elderly and super elderly populations. In older adults, maintaining a stable absolute intraoperative MAP should be considered to prevent intraoperative renal injury.

Abbreviations

ASA

American Society of Anesthesiologists

AKI

Acute Kidney Injury

BMI

Body Mass Index

GFR

Glomerular Filtration Rate

HR

Heart Rate

ICD

International Classification of Diseases

KDIGO

Global Committee for the Improvement of Kidney Disease

MAP

Mean Arterial Pressure

MI

Myocardial Injury

Scr

Serum Creatinine

WHO

World Health Organization

Authors’ contributions

All authors helped to design the work. J.C. and Y.P. collected data. R.L. and J.Q. performed data interpretation and analysis. R.L. and J.Q. wrote the paper. J.C., Y.P., J.W., F.J., and S.Z. revised the paper. Correspondence should be addressed to Ju Qian (qianjusuda@126.com).

Funding

This work was supported by National Natural Science Foundation of China (grant number: 82072130).

Data availability

Data can be obtained from the corresponding author upon reasonable request.

Declarations

Ethics approval

The study followed the Consolidated Standards of Reporting Trials statement and the Declaration of Helsinki. This work was registered in the Chinese Clinical Trials Registry (ChiCTR2400094990) on December 31, 2024, and The Ethics Committee of First Affiliated Hospital of Soochow University (LunShen [2024]704) approved the study.

Competing interests

The authors declare no competing interests.