Abstract
Background:
Tourniquet is widely used in total knee arthroplasty (TKA), but its impact on postoperative delirium (POD) remains unclear. The purposes of this study were to investigate the impact of tourniquet application on POD in elderly patients TKA and to explore the possible mechanisms associated with POD.
Materials and methods:
In this prospective, single-center randomized clinical trial study, 313 patients scheduled for single TKA under general anesthesia were randomly assigned to groups with or without a limb tourniquet. The primary outcome was the incidence of POD within postoperative 7 days. Secondary outcomes were the levels of hypoxia-inducible factor-1α (HIF-1α), tumor necrosis factor-alpha, superoxide dismutase (SOD), and S100β protein (S100β) measured postoperative 30 minutes and 24 hours, blood loss, pain score, the time to first postoperative ambulation, postoperative hospitalization, postoperative complications, and adverse events.
Results:
The incidence of POD in tourniquet group was significantly higher than that in no tourniquet group (19.1% vs. 9.6%, relative risk 1.12, 95% confidence interval: 1.02–1.23, P = 0.018). The serum level of HIF-1α was higher at postoperative 30 minutes and 24 hours and SOD was lower at postoperative 24 hours in tourniquet group compared with no tourniquet group. The incidence of postoperative complications and adverse events were comparable in both groups.
Conclusion:
We concluded that tourniquet application during TKA increased the incidence of POD within postoperative 7 days in older patients, and increased serum level of HIF-1α and decreased level of SOD, which indicated that they may be potential targets for preventing and treating POD.
Keywords: general anesthesia, mechanism, postoperative delirium, total knee arthroplasty, tourniquet
Introduction
Tourniquet is widely used during total knee arthroplasty (TKA) for decreasing intraoperative blood loss, enhancing visualization, reducing operating time, and optimizing cementation[1,2]. At the same time, it is reported that tourniquet may lead to tissue swelling and pain, thromboembolic events, infection, ischemia-reperfusion injury, and distant organs injury[3,4]. Its benefits and harms are still intensely under debate in the field of orthopedics. It is worth noting that a study found that the use of tourniquet in TKA was associated with the increased risk of postoperative delirium (POD), but its underlying mechanisms have not been fully elucidated due to complex physiological changes caused by the tourniquet[5].
HIGHLIGHTS
Tourniquet application during total knee arthroplasty (TKA) increased the incidence of postoperative delirium (POD).
The hypoxia-inducible factor-1α may be associated with POD.
Tourniquet-free during TKA can facilitate postoperative early ambulation and reduce hospital stay.
POD is an acute but transient neurocognitive disorder that usually occurs within 1 week after surgery and is characterized by symptoms including inattention, fluctuating levels of consciousness, and an acute change in cognitive function[6]. The median incidence of POD following lower limb major arthroplasty is 14.8%[7], usually resulting in delayed recovery, extended hospital stays, long-term cognitive impairments, and even increasing mortality rates[8–10]. However, no definitively effective preventive measures and treatments for POD have been recommended to date, so it prompted efforts to identify novel potential targets for preventing or treating POD.
The mechanisms of POD remain not fully understood, though various hypotheses have been proposed in the past few decades, including neuroinflammation, oxidative stress-related injury, and central cholinergic system dysfunction[11–13]. Recent studies have reported that the levels of hypoxia-inducible factor-1α (HIF-1α) within the body may correlate with cognitive function impairment which supports the hypothesis that HIF-1α may be associated with POD[14–16]. The use of tourniquet precisely increased the level of HIF-1α which underscores the importance of investigating the HIF-1α as a measure to understand the mechanisms of tourniquet on POD[17].
We hypothesize that tourniquet usage is associated with the increased risk of POD, and the molecular mechanism of HIF-1α plays an important role in this process. This study aims to investigate the impact of tourniquet application during TKA on the POD in older patients and to examine the serum levels of HIF-1α, superoxide dismutase (SOD), tumor necrosis factor-alpha (TNF-α), and S100β protein (S100β) to explore the possible mechanisms associated with POD.
Materials and methods
Study design
This prospective, single-center randomized clinical trial was conducted from June 20, 2024, to November 30, 2024. The Ethics Committee provided ethics approval for this study, and it was registered at the China Clinical Trial Center. Written informed consent was obtained from each participant before their enrollment. The study followed the CONSORT reporting guideline[18] and the TITAN Guidelines 2025[19].
Patients recruitment
The inclusion criteria were: ≥60 years old, ASA classification I–III, scheduled for single TKA under general anesthesia. The exclusion criteria were: history of communication disorders, preoperative cognitive dysfunction[20,21], history of psychiatric or neurological disorders (including depression, schizophrenia, epilepsy, and Parkinson’s or Alzheimer’s disease), severe arrhythmia, myocardial infarctions or poor cardiac function, severe liver or renal dysfunction, severe anemia or abnormal coagulation function, contraindications for tourniquet use, preoperative rheumatic activities, received surgery within 6 months before admission, allergic to the medication used in the study. The elimination criteria included operation canceled, tourniquet duration less than 1 hour, and intraoperative adverse events such as pulmonary embolism and anaphylactic shock.
Anesthesia procedure
Upon admission to the operating room, patients were given ECG, SpO2, and invasive arterial blood pressure monitoring. 0.05 mg/kg midazolam, 0.3 mg/kg etomidate, 0.5 µg/kg sufentanil, and 0.9 mg/kg rocuronium were used for anesthesia induction, and the patient was placed into a laryngeal mask after disconsciousness. Ultrasound-guided femoral nerve block was performed using 20 ml of 0.375% ropivacaine. Adjust tidal volume and respiratory rate to maintain end-expiratory carbon dioxide partial pressure between 35 and 45 mm Hg. Anesthesia was maintained using 1% sevoflurane inhalation combined with propofol 4–6 mg/kg/h and remifentanil 0.1–0.3 µg/kg/min to maintain BIS values between 50 and 60. Vasoactive drugs were used to keep blood pressure fluctuating within 20% of baseline according to intraoperative management protocol (Appendix 1). A postoperative patient-controlled intravenous analgesia pump was applied with 1.5 µg/kg sufentanil and 6 mg tropisetron. The pump was set with a continuous infusion rate of 2 ml/h, a self-controlled dose of 0.5 ml, and a lockout interval of 15 minutes. All the patients were then transferred to the PACU and antagonized with flumazenil and sugammadex sodium. The laryngeal mask was removed once patients regained consciousness from anesthesia and demonstrated adequate tidal volume and stable hemodynamic parameters. Finally, the patients were transferred to the surgical ward if all vital signs were stable after being observed for 30 minutes, otherwise, they would be admitted to the intensive care unit (ICU) for further surveillance and management. The implementation of postoperative rescue analgesia was based on patients’ pain score (Numeric Rating Scale [NRS] >4) assessed by nurses not involved in the study.
Intervention
A tourniquet was positioned at the upper thigh in both groups before the surgical procedure. The tourniquet was continuously utilized with the inflation pressure of 250 mm Hg from the onset of the incision to the completion of skin suturing in the tourniquet group. In contrast, the no tourniquet group had their tourniquet in place but without inflation. Every subject received an administration of intravenous tranexamic acid and antibiotics before the skin incision. The surgeries were performed by the same team of experienced surgeons who have well-established standardized surgical procedures consistency.
Randomization and blinding
Randomization was done using IBM SPSS Statistics, version 26.0 (IBM Corp, Armonk, NY) and allocation was concealed via sequentially numbered, opaque sealed envelopes by an independent investigator not involved in the outcome evaluation and data analysis. Eligible patients were randomly assigned in a 1:1 ratio to tourniquet group or no tourniquet group. The envelopes were handed to circulating nurses independent of the study after the patient entered the operating room. Blinding the surgeon was not feasible. Investigators responsible for the outcomes assessment and data analysis were unaware of the group assignments.
Outcome measurement
Primary outcome
The primary outcome was the incidence of POD during the first seven postoperative days, which was diagnosed with the Confusion Assessment Method (CAM) on the surgical wards in the hospital (telephone interviews were conducted after discharge) or CAM-intensive care unit for patients admitted to ICU by investigators who were trained and unaware of the group allocation[22]. The POD assessment was performed twice daily at 10.00 AM and 5.00 PM. Given the fluctuating characteristics of delirium, researchers also inquired with the patient’s family for insights into their symptoms to minimize the rate of missed diagnosis.
Secondary outcomes
Secondary outcomes included the assessment of delirium subtypes and severity utilizing the Delirium Rating Scale Revised 98[23,24] (from 0 [no delirium] to 39 [most severe level of delirium]), delirium duration (the days from delirium onset to the last day delirium recorded); The levels of HIF-1α, TNF-α, SOD, and S100β in venous blood collected at multiple points were recorded. Other secondary outcomes included incidence of red blood cell transfusion (according to blood transfusion criteria[25]), total blood loss (calculated by formula[25]), and change in hemoglobin from preoperative to the first postoperative day, worst pain scores both at rest and motion (assessed by the NRS), the incidence of rescue analgesia within 24 hours after surgery, time to first ambulation; ICU admission rates, durations of postoperative hospital care (defined as the time from the end of surgery to discharge). Other complications included deep vein thrombosis (lower limb deep vein ultrasound result), periprosthetic joint infections[26], acute kidney injury[27], and pulmonary complications. Other adverse events were also recorded.
Sample collection and timing
For each patient, 5 ml of peripheral venous blood sample from the upper extremity was collected before induction of general anesthesia (T0), 30 minutes after tourniquet release (T1), and 24 hours after tourniquet release (T2), and then centrifuged at 2500 rpm for 15 minutes to obtain serum, which would be stored at −80°C for subsequently testing. Serum concentrations of HIF-1α, TNF-α, SOD, and s100β were assessed by enzyme-linked immunoassay kits according to the manufacturer’s instructions. Biomarker levels below the limit of detection were set to one-half of that level[28].
Statistical analysis
Sample size
Based on the result of a previous study, the incidence of POD among the old patients undergoing TKA was 15.2% in the tourniquet group and 5.4% in the no tourniquet group[5]. The PASS 15.0 was used to calculate sample size and we used a two-sided significance test with study power of 0.80 and test level of 0.05. Then 148 patients per group were needed, assuming the possible 5% dropout rate, 156 patients in each group were needed to detect the difference in POD incidence between the two groups.
Data analysis
Continuous variables were described as mean and standard deviation or median and interquartile range (IQR) depending on the normality distribution which was tested using the Shapiro–Wilk test. Normally distributed data were analyzed with independent sample t-tests, whereas data were analyzed with Mann–Whitney U test. Categorical variables were presented as frequencies and proportions n (%) and compared using χ2 test or Fisher’s exact test. For data gathered at multiple time points, generalized estimating equation with Bonferroni correction was employed for analysis. Statistical analysis for the primary and secondary outcomes was conducted with the intention-to-treat (ITT) principle. Time until POD onset within 7 days after operation was evaluated with Kaplan–Meier survival curves.
To handle missing data, multiple imputation by chained equations were used under the missing-at-random assumption. As a post hoc sensitivity analysis, a logistic regression model was used to account for potential confounders by adjusting for variables related to delirium, including age, FRAIL, MMSE, and hypotension during operation. Statistical analyses were conducted using SPSS 26.0 (IBM Corp, Armonk, NY), and analysis items with two-sided P < 0.05 were considered statistically significant.
Results
Patient characteristics and perioperative variables
A total of 389 patients were considered for enrollment in the study. Twenty patients did not meet the inclusion criteria due to being under 60 years of age, 11 patients declined to participate in the study, and 45 patients met the exclusion criteria. Finally, 313 patients were enrolled and 157 were randomized to tourniquet group and 156 to no tourniquet group. In the tourniquet group, two patients were eliminated for tourniquet duration of less than 1 hour. The flow diagram of participants is presented in Figure 1. The baseline characteristics of the participants were well balanced (Table 1). In terms of intraoperative data, the estimated blood loss during operation in tourniquet group (100 [50–110]) was significantly lower than no tourniquet group (200 [150–250], P < 0.001). The incidence of hypertension and then urapidil use (8.3% vs. 2.6%, P = 0.026) were higher in the tourniquet group. Despite the more usage of phenylephrine in the no tourniquet group, the incidence of intraoperative hypotension showed no significant difference (12.1% vs. 14.1%, P = 0.600) between the two groups (Table 2).
Figure 1.
Study flow diagram.
Table 1.
Baseline participant characteristics
| Characteristics | Tourniquet group (n = 157) | No tourniquet group (n = 156) |
|---|---|---|
| Age, mean (SD), year | 69.2 (5.5) | 69.3 (6.3) |
| Sex, n (%) | ||
| Male | 47 (29.9) | 42 (26.9) |
| Female | 110 (70.1) | 114 (73.1) |
| Body mass index, mean (SD), kg/m2 | 26.1 (3.2) | 25.8 (2.9) |
| Current smoker, n (%) | 19 (12.1) | 13 (8.3) |
| Alcohol drinking, n (%) | 16 (10.2) | 16 (10.3) |
| Educational level, n (%) | ||
| Illiteracy | 54 (34.4) | 44 (28.2) |
| Primary school | 51 (32.5) | 50 (32.1) |
| Junior high school | 50 (31.9) | 62 (39.7) |
| University or higher | 2 (1.3) | 0 (0) |
| ASA classification, n (%) | ||
| I | 1 (0.6) | 1 (0.6) |
| II | 108 (68.8) | 102 (65.4) |
| III | 48 (30.6) | 53 (34) |
| Medical history, n (%) | ||
| Hypertension | 96 (61.1) | 99 (63.5) |
| Diabetes | 29 (18.5) | 26 (16.7) |
| Coronary artery disease | 16 (10.2) | 23 (14.7) |
| Heart failure | 0 (0) | 1 (0.6) |
| History of stroke | 31 (19.7) | 43 (27.6) |
| Hepatic dysfunction | 4 (2.5) | 4 (2.6) |
| COPD | 5 (3.2) | 4 (2.6) |
| Age-adjusted Charlson Comorbidity Index, median (IQR), score | 3.0 (2.0–4.0) | 3.0 (3.0–4.0) |
| Mini-Mental State Examination, median (IQR), score | 22.0 (20.0–24.0) | 22.0 (20.0–24.0) |
| Self-Rating Anxiety Scale, median (IQR), score | 3 (3–4) | 3 (2–4) |
| FRAIL, n (%) | ||
| Robust | 56 (35.7) | 59 (37.8) |
| Prefrail | 76 (48.4) | 68 (43.6) |
| Frail | 25 (15.9) | 29 (18.6) |
| Numeric Rating Scale for pain, median (IQR), score | ||
| At rest | 2 (2–3) | 2 (2–3) |
| At mobilizationa | 5 (5–6) | 5 (4–6) |
| NSAIDS user, n (%) | 92 (58.6) | 99 (63.5) |
| Preoperative systolic blood pressure, mean (SD), mm Hg | 141.2 (17.0) | 139.0 (16.1) |
| Preoperative diastolic blood pressure, mean (SD), mm Hg | 85.3 (11.8) | 82.9 (8.5) |
| Preoperative heart rate, mean (SD), beats/min | 78.1 (9.5) | 76.2 (11.3) |
| Preoperative laboratory values | ||
| Hemoglobin, mean (SD), g/dl | 132.6 (14.3) | 133.6 (14.2) |
| Platelet count, mean (SD), ×103/μl | 226.4 (51.8) | 232.4 (67.3) |
| Albumin, median (IQR), g/L | 42.7 (40.4–44.7) | 43.0 (41.0–45.3) |
| hs-CRP, median (IQR), mg/L | 1.1 (0.5–2.4) | 1.0 (0.6–2.3) |
| eGFR, mean (SD), ml min−1 1.73 m−2 | 110.1 (13.0) | 113.1 (12.0) |
| Creatinine, median (IQR), mg/dl | 55 (47–62) | 54 (47–66) |
ASA, American Society of Anesthesiologists; COPD, Chronic Obstructive Pulmonary Disease; eGFR, estimated glomerular filtration rate; FRAIL, Fatigue, Resistance, Ambulation, Illness and Loss of weight; hs-CRP, high-sensitivity C-reactive protein; IQR, interquartile range; NSAIDs, non-steroidal anti-inflammatory drugs; SD, standard deviation.
At mobilization: pain after flexing knee to 45° 24 h after operation.
Table 2.
Intraoperative data between the groups
| Characteristics | Tourniquet group (n = 157) | No tourniquet group (n = 156) | P-value |
|---|---|---|---|
| Duration of operation, median (IQR), min | 80 (70–90) | 83 (75–90) | 0.588 |
| Duration of anesthesia, median (IQR), min | 100 (95–110) | 108 (95–119) | 0.058 |
| Fluid administered, median (IQR), ml | 1250 (1225–1750) | 1500 (1050–1500) | 0.728 |
| Estimated blood loss, median (IQR), ml | 100 (50–110) | 200 (150–250) | <0.001 |
| Red blood cell transfusion, no (%) | 1 (0.6) | 0 (0) | 1.000 |
| Propofol, median (IQR), mg | 200 (185–250) | 200 (195–250) | 0.961 |
| Sufentanil, median (IQR), μg | 30 (25–30) | 30 (25–30) | 0.255 |
| Remifentanil, median (IQR), mg | 1.7 (1.5–2.0) | 1.7 (1.4–1.8) | 0.321 |
| Phenylephrine, median (IQR), μg | 120 (80–200) | 160 (120–400) | <0.001 |
| Ephedrine, median (IQR), mg | 3.0 (0–6.0) | 3.0 (0–6.0) | 0.238 |
| Intraoperative hypertensiona, no (%) | 52(33.1) | 28(17.9) | 0.020 |
| Urapidil, no (%) | 13 (8.3) | 4 (2.6) | 0.026 |
| Episodes of hypotensionb, no (%) | 19 (12.1) | 22 (14.1) | 0.600 |
A 30% increase in systolic blood pressure (SBP) ≥ 3 min.
A mean arterial pressure (MAP) ≤65 mm Hg for ≥1 min.
Incidence of POD
In the ITT analysis, POD occurred in 45 of 313 participants (14.4%) during the initial postoperative 7 days. The incidence of POD in the tourniquet group was 19.1% (n = 30) and 9.6% (n = 15) in the no tourniquet (relative risk [RR] 1.12, 95% CI: 1.02–1.23, P = 0.018), the absolute risk difference was 9.5%, resulting a number needed to treat to avoid one POD event of 11 (Table 3 and Fig. 2). The median duration of POD in tourniquet group and no tourniquet group were all 2.0 days (IQR: 1.0–2.0 days, P = 0.848). The severity of POD was comparable between the two groups (P = 0.971). Within postoperative 7 days, the hypoactive type of POD occurred in 60.0% (27/45), hyperactive symptoms in 24.4% (11/45), and mixed one in 15.6% (7/45). The incidence of all three subtypes of delirium was lower in the no tourniquet group (P = 0.125).
Table 3.
Primary and secondary outcomes
| Outcomes | Tourniquet group (n = 157) | No tourniquet group (n = 156) | RR or MD (95% CI) | P-value |
|---|---|---|---|---|
| Postoperative delirium, no (%) | 30 (19.1) | 15 (9.6) | 1.12 (1.02–1.23) | 0.018 |
| Type of delirium, no (%) | 0.125 | |||
| Hypoactive | 18 (11.5) | 9 (5.8) | NA | |
| Hyperactive | 7 (4.5) | 4 (2.6) | NA | |
| Mixed | 5 (3.2) | 2 (1.3) | NA | |
| Days with delirium, median (IQR), days | 2 (1–2) | 2 (1–2) | 0.00 (0.00–0.000) | 0.848 |
| Delirium severity, median (IQR), score | 18 (16–20) | 18 (17–20) | 0.00 (−2.00 to 1.00) | 0.971 |
| Change in hemoglobina, median (IQR), g/dl | 16 (10–21) | 17 (12–22) | −2 (−4 to 1) | 0.189 |
| Total blood loss, median (IQR), ml | 559.0 (348.2–754.5) | 577.0 (423.3–800.8) | −63.4 (−147.9 to 19.8) | 0.127 |
| Maximum pain at rest, median (IQR), score | 2 (1–2) | 2 (1–2) | 0.00 (0.00–0.00) | 0.057 |
| Maximum pain at mobilization, median (IQR), score | 3 (3–5) | 3 (3–4) | 0.00 (0.00–0.00) | 0.908 |
| Thigh pain, no (%) | 112 (71.3) | 65 (41.7) | 2.04 (1.54–2.69) | <0.001 |
| Rescue analgesia within 24 h, no (%) | 47 (29.9) | 34 (21.8) | 1.12 (0.98–1.27) | 0.100 |
| Time to first ambulation, median (IQR), h | 40 (35–48) | 40 (34–44) | 2.00 (0.00–4.00) | 0.042 |
| Admission to ICU, no (%) | 0 (0) | 1 (0.6) | 0.99 (0.98–1.01) | 0.498 |
| Other complications, no (%) | ||||
| Deep vein thrombosis | 60 (38.2) | 44 (28.2) | 1.16 (0.99–1.36) | 0.060 |
| Acute kidney injury | 1 (0.6) | 0 (0) | 1.01 (0.99–1.02) | 1.000 |
| Pneumoniab | 0 (0) | 1 (0) | 0.99 (0.98–1.01) | 0.498 |
| Periprosthetic joint infections | 0 (0) | 0 (0) | NA | NA |
| Duration of postoperative stay, mean (SD), day | 4.27 (0.86) | 3.50 (0.55) | 0.77 (0.61–0.93)c | <0.001 |
MD, median difference; NA, not applicable; RR, risk ratio.
HbPreoperative—Hbpostoperative 1 day.
Pneumonia was defined as a positive sputum culture or presence of a postoperative pulmonary infiltrate with systemic signs of infection (temperature >38.6°C or white blood cell count >12.0 × 109/L) and the use of parenteral antibiotics or documentation of the diagnosis by the patient’s physician.
Mean difference.
Figure 2.
Kaplan–Meier curve showing intention-to-treat analysis of the cumulative incidence of postoperative delirium during postoperative days 1–7 between two groups.
Serum levels of HIF-1α, S100β, SOD, and TNF-α
The measurements of serum levels of HIF-1α, S100β, SOD, and TNF-α are shown in Figure 3. There were no differences in the preoperative concentrations of the four indicators between groups. On T1 and T2, the level of HIF-1α was significantly higher in tourniquet group compared with no tourniquet group (Fig. 3A). The level of SOD was lower in the tourniquet group on T2 (Fig. 3C). Nevertheless, no significant differences were noted in levels of TNF-α (Fig. 3B) and s100β (Fig. 3D) on T1 and T2 between the two groups.
Figure 3.
Changes in serum S100β, TNF-α, SOD, and HIF-1α levels during perioperative periods. HIF-1α, hypoxia-inducible factor-1α; TNF-α, tumor necrosis factor-alpha; SOD, superoxide dismutase; S100β, S100β protein; T0, preoperative baseline; T1, postoperative 30 min; T2, postoperative 24 h. Compared between the two groups, *P < 0.05.
Other outcomes and adverse events
The measured intraoperative blood loss was higher in no tourniquet group (200 [150–250]) compared with tourniquet group (100 [50–110], P < 0.001), but the intraoperative transfusion of blood products was rare and similar between groups (0% vs. 0.6%, P = 1.000). Also, there were no significant differences in calculated total blood loss (MD, −63.4, 95% CI: −147.9 to 19.8, P = 0.127) and changes in hemoglobin levels (MD, −2, 95% CI: −4 to 1, P = 0.189) between two groups. The worst pain scores at rest and motion, as well as the incidence of rescue analgesia (RR, 1.12, 95% CI: 0.98–1.27, P = 0.100) within postoperative 24 hours, were similar between tourniquet and no tourniquet groups. However, the rate of thigh swelling reported was significantly higher in tourniquet group (RR, 2.04, 95% CI: 1.54–2.69, P < 0.001). The time to first postoperative ambulation was later in the tourniquet group compared to the no tourniquet group (MD, 2.0, 95% CI: 0–4.00, P = 0.042). In terms of ICU admission rates, no difference was observed between the two groups (RR, 0.99, 95% CI: 0.98–1.01, P = 0.498). Only one patient in no tourniquet group was unplanned admitted to the ICU due to hypoxia in the PACU and transferred to the general ward the second day. The duration of postoperative hospital stay in tourniquet group was longer than no tourniquet group (MD, 0.77, 95% CI: 0.61–0.93, P < 0.001, Table 3). There was no difference in other complications between the two groups. The incidence of adverse events between the two groups during the postoperative stay was comparable (Supplemental Digital Content Table 1, available at: http://links.lww.com/JS9/E415).
Sensitivity analysis
We identified variables that may influence POD, including age, FRAIL, MMSE, and intraoperative hypotension, and the tourniquet still demonstrated the disadvantage of increasing risk of POD after adjusting for these potential confounding factors (OR, 2.51, 95% CI: 1.24–5.08, P = 0.010) (Supplemental Digital Content Figure 1, available at: http://links.lww.com/JS9/E415).
Discussion
In this study, we described the impact of intraoperative use of tourniquet on POD in older patients undergoing TKA and we found that without tourniquet did reduce the incidence of POD compared with tourniquet group within postoperative 7 days. In total, 9.6% (15 of 156) of patients who were randomized to no tourniquet group and 19.1% (30 of 157) of patients in tourniquet group experienced POD. Moreover, our findings supported that HIF-1α and SOD were associated with tourniquet-induced POD, which provided potential targets for preventing or treating POD.
Our result differed from previous studies which suggested that tourniquet application did not worsen or even improved postoperative cognitive function scores[29,30], which may be attributed to their smaller sample sizes and the fact that cognitive scores were only assessed as secondary outcomes. Moreover, they failed to analyze the underlying reasons for their findings. In contrast, our finding was supported by another study[5], which indicated that tourniquet use increased the incidence of POD. However, it was limited to traditional inflammatory mechanisms as potential explanations. Additionally, we observed a higher overall POD incidence in our study, which may be due to the lower baseline MMSE scores in our study population compared to it.
The underlying mechanisms of POD are intricate and not fully understood, yet it is hypothesized to encompass neuroinflammation, oxidative stress, imbalances in neurotransmitter levels, and metabolic homeostasis[31–33], and recent animal model studies have found that level of HIF-1α may be associated with cognitive function[34,35]. In TKA, the application of tourniquet can increase the level of HIF-1α in the body and exacerbate the inflammatory response and oxidative stress levels after deflation[17,36]. The results of our study demonstrated that serum HIF-1α was significantly higher at 30 minutes and 24 hours after operation and the level of SOD was lower at postoperative 24 hours in tourniquet group. POD development might be linked to HIF-1α elevation because HIF-1α activation can stimulate the production of inflammatory cytokines and contribute to mitochondrial dysfunction, activation of neuroglial cells, and neuronal damage[16]. Additionally, dysregulation of the HIF-1α signaling pathway may disrupt synaptic plasticity and neuronal survival[35], all of which may contribute to POD. And preclinical study indicated that treatment with HIF-1α inhibitor contributed to decreased levels of HIF-1α protein and gene expression, and then improved cognitive function in septic mice[14]. The evidence all supported that HIF-1α was associated with the occurrence of POD. Also, a lower level of SOD in tourniquet group was associated with a higher incidence of POD which is consistent with the finding in cardiac surgery[32]. Oxidative stress products can directly damage neuronal cell membranes and protein functions through lipid peroxidation, leading to neuronal injury and POD[37]. However, no significant differences were observed in serum levels of inflammatory marker TNF-α and the BBB injury biomarker S100β at these two postoperative time points. The inconsistency between the incidence of POD and TNF-α changes may be due to that inflammatory response in the central nervous system occurs independently of peripheral inflammation, and these inflammatory differences may exist in the cerebrospinal fluid instead of serum which should be detected in future studies[38]. Another possible reason is that the duration of the tourniquet usage is too short to cause a significant inflammatory response difference in the body or the difference in inflammatory marker may reflect in the inflammatory response peak on the third postoperative day, but the trend of changes over a longer period was not detected in our study[39]. While no significant difference in S100β levels was observed between groups, the incidence of POD was higher in tourniquet group. The possible reason is that more intense oxidative stress response and higher HIF-1α levels in tourniquet group lead to amplified neuronal damage through injured BBB under conditions of comparable BBB disruption.
In terms of blood loss, our results indicated that there were no significant differences in total blood loss and hemoglobin reduction between the two groups within 24 hours postoperatively, which was consistent with a prior research finding[40]. A previous meta-analysis indicated that TKA with tourniquet was associated with higher postoperative pain at postoperative 24 hours[3]. However, the results of this study indicated that the application of tourniquet did not increase the worst pain at rest or movement within 1 day after surgery, nor did it significantly affect the rate of rescue analgesia. This may be due to that our study distinguished between thigh swelling and pain at the incision site, which could improve the accuracy of pain assessment and provide more effective pain management but usually might be ignored in previous studies. Indeed, the rate of reported thigh swelling was significantly higher in the tourniquet group compared to the other group in this research. Thigh swelling may increase patients’ concerns about falling in terms of early ambulation, thereby delaying their time to first ambulation postoperatively. The results of this study demonstrated that patients in the no tourniquet group get earlier postoperative ambulation and the findings will hold significant value for the refinement of the Enhanced Recovery After Surgery concept and the early recovery strategies for patients undergoing day-case TKA[41]. A higher incidence of delirium and delayed ambulation postoperatively may contribute to prolonged hospitalization in tourniquet group[42]. In this study, we did not observe any difference in the incidence of postoperative complications and adverse events between the two groups.
This study represents the largest randomized controlled trial to date investigating the impact of tourniquet application on POD in elderly patients. Furthermore, our study involves several markers associated with POD, providing novel insights into its pathogenesis and revealing potential molecular targets for POD prevention and therapeutic intervention. The trial also has several limitations. First, the study was limited to the single hospital, which may limit the generalizability and have been unpowered to detect the difference in some rare complications and adverse events. Second, we attempted to blind the surgeons, but they might have deduced the patient’s allocation based on intraoperative blood loss, however, we believed that this limitation had little impact on the outcomes assessment as outcomes assessors and data analysts were all well blinded. Third, our study was conducted under general anesthesia, and further studies are warranted to determine if the conclusions apply to TKA under neuraxial anesthesia and nerve block anesthesia. Additionally, the uniform tourniquet pressure and narrow range of operative duration in our study prevented analysis of pressure and duration effects on POD, which should be investigated in future studies. Furthermore, considering patients’ satisfaction and compliance, we did not collect blood samples at additional multiple time points postoperatively, which limited the observation of long-term trends of biomarkers. The last, we did not follow up for longer duration to assess the impact of tourniquet usage on postoperative long-term cognitive function in patients.
In conclusion, our study found that tourniquet application increased the incidence of POD in elderly patients undergoing TKA within postoperative 7 days and influenced serum levels of HIF-1α and SOD, which suggested that they are associated with POD. In addition, we found that the omission of tourniquet during operation can facilitate postoperative early ambulation and reduce postoperative hospital stay, which is beneficial for postoperative recovery in patients undergoing TKA.
Footnotes
X.C., X.W., and Y.L. contributed equally to this work.
Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article.
Supplemental Digital Content is available for this article. Direct URL citations are provided in the HTML and PDF versions of this article on the journal’s website, www.lww.com/international-journal-of-surgery.
Published online 20 June 2025
Contributor Information
Xiaoyi Chen, Email: 2406729984@qq.com.
Xinghe Wang, Email: wangxh-77@163.com.
Yucheng Liu, Email: Yuchengliu0422@163.com.
Jie Guo, Email: jieguo0222@163.com.
Tianyu Yin, Email: 15366881536@163.com.
Junda Chen, Email: jundachen0920@163.com.
Yangyang Chen, Email: 1667328322@qq.com.
Keqin Wang, Email: wkq4689@163.com.
Yuqi Zhang, Email: yuqizhang1016@163.com.
Ziqian Zhang, Email: ziqian1307@163.com.
Linlin Zhao, Email: lin1675@163.com.
Su Liu, Email: liusu-xzmu@163.com.
Ethical approval
The Ethics Committee provided ethics approval for this study (April 29, 2024).
Consent
Written informed consent was obtained from each participant before their enrollment.
Sources of funding
This work was supported by the Natural Science Research Fund of Higher Education Institutions in Jiangsu Province (22KJA320007), the Science and Technology Project of Xuzhou Health Commission (XWKYHT20230056), Construction Project of High Level Hospital of Jiangsu Province (GSPSJ20240803) and Paired Assistance Scientific Research Project by The Affiliated Hospital of Xuzhou Medical University (FXJDBF2024219).
Author contributions
Conceptualization, data curation, formal analysis, investigation, methodology, project administration, writing-original draft, writing-review and editing: X.C.; data curation, formal analysis, methodology, writing-original draft, writingreview and editing: X.W.; data curation, formal analysis, investigation, writingoriginal draft, writing-review and editing: Y.L.; data curation, investigation, methodology, project administration: J.G.; data curation, formal analysis, project administration, methodology: T.Y.; data curation, formal analysis, project administration, methodology: J.C.; data curation, project administration, investigation: Y.C.; data curation, project administration, methodology: K.W.; data curation, project administration, methodology: Y.Z.; data curation, project administration: Z.Z.; investigation, methodology: L.Z.; conceptualization, supervision: Y.W.; conceptualization, funding acquisition, supervision: S.L.
Conflicts of interest disclosure
Not applicable.
Research registration unique identifying number (UIN)
http://www.chictr.org.cn/index.aspx (ChiCTR2400085505).
Guarantor
Su Liu.
Provenance and peer review
Not commissioned, externally peer reviewed. Our paper was not invited.
Data availability statement
Datasets included in this study are available upon reasonable request.
Acknowledgements
Not applicable.
References
- [1].Johnsen M, Havik S, Husby VS, et al. The use of tourniquet in total knee arthroplasty does not impact the functional outcome: a randomised controlled study. J Orthop Surg Res 2024;19:704. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [2].Kunes JA, El-Othmani MM, LaVelle M, Santos WM, Geller JA, Shah RP. Tourniquet-free minimally invasive total knee arthroplasty is associated with early aseptic loosening. Knee 2024;46:19–26. [DOI] [PubMed] [Google Scholar]
- [3].Ahmed I, Chawla A, Underwood M, et al. Tourniquet use for knee replacement surgery. Cochrane Database Syst Rev 2020;12:CD012874. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Leurcharusmee P, Sawaddiruk P, Punjasawadwong Y, Chattipakorn N, Chattipakorn SC. The possible pathophysiological outcomes and mechanisms of tourniquet-induced ischemia-reperfusion injury during total knee arthroplasty. Oxid Med Cell Longev 2018;2018:8087598. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [5].Ran W, Li S, Yuan R, Luo H, Li P, Gao J. Effect of tourniquet technique on postoperative delirium in elderly patients with total knee arthroplasty: a randomized single-blind controlled trial. BMC Anesthesiol 2022;22:396. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [6].Inouye SK, Westendorp RGJ, Saczynski JS. Delirium in elderly people. Lancet 2014;383:911–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [7].Kitsis P, Zisimou T, Gkiatas I, et al. Postoperative delirium and postoperative cognitive dysfunction in patients with elective hip or knee arthroplasty: a narrative review of the literature. Life (Basel) 2022;12:314. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [8].Witlox J, Eurelings LSM, de Jonghe JFM, Kalisvaart KJ, Eikelenboom P, van Gool WA. Delirium in elderly patients and the risk of postdischarge mortality, institutionalization, and dementia: a meta-analysis. JAMA 2010;304:443–51. [DOI] [PubMed] [Google Scholar]
- [9].Brown CH, Probert J, Healy R, et al. Cognitive decline after delirium in patients undergoing cardiac surgery. Anesthesiology 2018;129:406–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].Meyer M, Götz J, Parik L, et al. Postoperative delirium is a risk factor for complications and poor outcome after total hip and knee arthroplasty. Acta Orthopaedica 2021;92:695–700. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [11].Wilson JE, Mart M, Cunningham C, et al. Delirium. Nat Rev Dis Primers 2020;6:90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [12].Lopez MG, Pandharipande P, Morse J, et al. Intraoperative cerebral oxygenation, oxidative injury, and delirium following cardiac surgery. Free Radic Biol Med 2017;103:192–98. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [13].Müller A, Olbert M, Heymann A, et al. Relevance of peripheral cholinesterase activity on postoperative delirium in adult surgical patients (CESARO): a prospective observational cohort study. Eur J Anaesthesiol 2019;36:114–22. [DOI] [PubMed] [Google Scholar]
- [14].Zhao L, Song Y, Zhang Y, et al. HIF-1α/BNIP3L induced cognitive deficits in a mouse model of sepsis-associated encephalopathy. Front Immunol 2022;13:1095427. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [15].Prabhakar NR, Peng YJ, Nanduri J. Hypoxia-inducible factors and obstructive sleep apnea. J Clin Invest 2020;130:5042. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [16].Choi YK. Detrimental roles of hypoxia-inducible factor-1α in severe hypoxic brain diseases. Int J Mol Sci 2024;25:4465. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [17].García-de-la-Asunción J, Perez-Griera J, Moreno T, et al. Limb ischemic conditioning induces oxidative stress followed by a correlated increase of HIF-1α in healthy volunteers. Ann Vasc Surg 2020;62:412–19. [DOI] [PubMed] [Google Scholar]
- [18].Butcher NJ, Monsour A, Mew EJ, et al. Guidelines for reporting outcomes in trial reports: the CONSORT-outcomes 2022 extension. JAMA 2022;328:2252–64. [DOI] [PubMed] [Google Scholar]
- [19].Agha RA, Mathew G, Rashid R, et al. Transparency in the reporting of Artificial INtelligence – the TITAN guideline. Premier J Sci 2025;10:100082. [Google Scholar]
- [20].Li H, Jia J, Yang Z. Mini-mental state examination in elderly Chinese: a population-based normative study. J Alzheimers Dis 2016;53:487–96. [DOI] [PubMed] [Google Scholar]
- [21].Folstein MF, Folstein SE, McHugh PR. “Mini-mental state.” A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12:189–98. [DOI] [PubMed] [Google Scholar]
- [22].Inouye SK, van Dyck CH, Alessi CA, Balkin S, Siegal AP, Horwitz RI. Clarifying confusion: the Confusion Assessment Method. A new method for detection of delirium. Ann Intern Med 1990;113:941–48. [DOI] [PubMed] [Google Scholar]
- [23].Huang MC, Lee CH, Lai YC, Kao YF, Lin HY, Chen CH. Chinese version of the delirium rating scale-revised-98: reliability and validity. Compr Psychiatry 2009;50:81–85. [DOI] [PubMed] [Google Scholar]
- [24].Bowman EML, Cunningham EL, Page VJ, McAuley DF. Phenotypes and subphenotypes of delirium: a review of current categorisations and suggestions for progression. Critical Care 2021;25:334. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [25].Wei Z, Yu M, Xu Y, Weng X, Feng B. Impact of intraosseous regional administration of tranexamic acid in total knee arthroplasty on perioperative blood loss: a protocol for a randomised controlled trial. BMJ Open 2024;14:e077393. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [26].McNally M, Sousa R, Wouthuyzen-Bakker M, et al. The EBJIS definition of periprosthetic joint infection. Bone Joint J 2021;103-B:18–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [27].Kellum JA, Lameire N. KDIGO AKI Guideline Work Group. Diagnosis, evaluation, and management of acute kidney injury: a KDIGO summary (part 1). Crit Care 2013;17:204. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [28].Lindholm D, Lindbäck J, Armstrong PW, et al. Biomarker-based risk model to predict cardiovascular mortality in patients with stable coronary disease. J Am Coll Cardiol 2017;70:813–26. [DOI] [PubMed] [Google Scholar]
- [29].Dong J, Min S, He KH, Peng LH, Cao J, Ran W. Effects of the nontourniquet combined with controlled hypotension technique on pain and long-term prognosis in elderly patients after total knee arthroplasty: a randomized controlled study. J Anesth 2019;33:587–93. [DOI] [PubMed] [Google Scholar]
- [30].Li X, Liu J, Wang H, Ding Y. Controlled hypotension technology can improve patient recovery in the early postoperative period after total knee arthroplasty: a prospective, randomized controlled clinical study. Jt Dis Relat Surg 2024;35:36–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [31].Alam A, Hana Z, Jin Z, Suen KC, Ma D. Surgery, neuroinflammation and cognitive impairment. EBioMedicine 2018;37:547–56. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [32].Karlidag R, Unal S, Sezer OH, et al. The role of oxidative stress in postoperative delirium. Gen Hosp Psychiatry 2006;28:418–23. [DOI] [PubMed] [Google Scholar]
- [33].Cerejeira J, Nogueira V, Luís P, Vaz-Serra A, Mukaetova-Ladinska EB. The cholinergic system and inflammation: common pathways in delirium pathophysiology. J Am Geriatr Soc 2012;60:669–75. [DOI] [PubMed] [Google Scholar]
- [34].Yao P, Wu L, Yao H, Shen W, Hu P. Acute hyperglycemia exacerbates neuroinflammation and cognitive impairment in sepsis-associated encephalopathy by mediating the ChREBP/HIF-1α pathway. Eur J Med Res 2024;29:546. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [35].Porel P, Bala K, Aran KR. Exploring the role of HIF-1α on pathogenesis in Alzheimer’s disease and potential therapeutic approaches. Inflammopharmacology 2025;33:669–78. [DOI] [PubMed] [Google Scholar]
- [36].Kim SH, Kim DH, Shin S, Kim SJ, Kim TL, Choi YS. Effects of dexmedetomidine on inflammatory mediators after tourniquet-induced ischemia-reperfusion injury: a randomized, double-blinded, controlled study. Minerva Anestesiol 2019;85:279–87. [DOI] [PubMed] [Google Scholar]
- [37].Lopez MG, Hughes CG, DeMatteo A, et al. Intraoperative oxidative damage and delirium after cardiac surgery. Anesthesiology 2020;132:551–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [38].Gigase FAJ, Smith E, Collins B, et al. The association between inflammatory markers in blood and cerebrospinal fluid: a systematic review and meta-analysis. Mol Psychiatry 2023;28:1502–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [39].Ali KA, He L, Deng X, Pan J, Huang H, Li W. Assessing the predictive value of pre- and post-operative inflammatory markers in patients undergoing total knee arthroplasty. J Orthop Surg Res 2024;19:614. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [40].Zhang W, Li N, Chen S, Tan Y, Al-Aidaros M, Chen L. The effects of a tourniquet used in total knee arthroplasty: a meta-analysis. J Orthop Surg Res 2014;9:13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [41].Jenny JY, Godet J. Resurg, SOFCOT Fast-Track and Ambulatory TKA Working Group, SOFCOT, de Ladoucette A. Complication rates are not higher after outpatient versus inpatient fast-track total knee arthroplasty: a propensity-matched prospective comparative study. J Arthroplasty 2024;39:2458–65. [DOI] [PubMed] [Google Scholar]
- [42].Yan E, Veitch M, Saripella A, et al. Association between postoperative delirium and adverse outcomes in older surgical patients: a systematic review and meta-analysis. J Clin Anesth 2023;90:111221. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Datasets included in this study are available upon reasonable request.