Literature Decoded

Living reprintoriginally published in BMC Medicine 2025CC BY 4.0Read the version of record Opt out

Literature DecodedLiving reprint
Living reprint · journal article Journal version available

Effect of combination of multiple anti-inflammatory drugs strategy on postoperative delirium among older patients undergoing hip fracture surgery: a pilot randomized controlled trial

Randomised trial first published in BMC Medicine (2025), reprinted in full under its CC BY 4.0 licence.

Ayixia Nawan1, Zilong Wu2, Bailin Jiang2*ORCID iD 0000-0002-9012-7472, Geng Wang3, Wenchao Zhang3, Yi Feng4*ORCID iD 0000-0002-3952-1351

1https://ror.org/035adwg89grid.411634.50000 0004 0632 4559Department of Anesthesiology, Peking University People’s Hospital, No.11 Xizhimen South Street, Xicheng District, Beijing, 100044 China; https://ror.org/013xs5b60grid.24696.3f0000 0004 0369 153XDepartment of Anesthesiology, Beijing Jishuitan Hospital, Capital Medical University, Beijing, China2https://ror.org/035adwg89grid.411634.50000 0004 0632 4559Department of Anesthesiology, Peking University People’s Hospital, No.11 Xizhimen South Street, Xicheng District, Beijing, 100044 China3https://ror.org/013xs5b60grid.24696.3f0000 0004 0369 153XDepartment of Anesthesiology, Beijing Jishuitan Hospital, Capital Medical University, Beijing, China4https://ror.org/035adwg89grid.411634.50000 0004 0632 4559Department of Anesthesiology, Peking University People’s Hospital, No.11 Xizhimen South Street, Xicheng District, Beijing, 100044 China; https://ror.org/035adwg89grid.411634.50000 0004 0632 4559Department of Pain Medicine, Peking University People’s Hospital, Beijing, China; https://ror.org/02v51f717grid.11135.370000 0001 2256 9319Key Laboratory for Neuroscience, Ministry of Education, National Health Commission, Peking University, Beijing, China*Corresponding author

Reprinted 2026-10-01 19 min read Living reprint · journal article Version of record: BMC Medicine 2025Licence: CC BY 4.0

Reading mode
In plain languageRandomised trial first published in BMC Medicine (2025), reprinted in full under its CC BY 4.0 licence.

Randomised trial first published in BMC Medicine (2025), reprinted in full under its CC BY 4.0 licence.

No plain-language summary has been written for this reprint yet. The authors' abstract and full text follow, unchanged apart from layout.

Educational summary of research findings; not medical advice. Discuss care decisions with a qualified clinician.

Bottom line

What the evidence supports

How to read it

No plain-language summary has been written for this reprint yet. The authors' abstract and full text follow, unchanged apart from layout.

Limitations

The findings apply to the included study populations and may not generalise to every person or setting.

Disclaimer

Educational summary of research findings; not medical advice. Discuss care decisions with a qualified clinician.

The paper

Full manuscript

Abstract

Background

Postoperative delirium is the most common complication in older patients and is associated with surgery-induced inflammation. Although inflammation plays a key role in delirium, the potential benefits of a comprehensive anti-inflammatory approach to managing perioperative systemic inflammation remain underexplored. This study evaluated whether a perioperative anti-inflammatory bundle strategy, combining dexmedetomidine, glucocorticoids, ulinastatin, and nonsteroidal anti-inflammatory drugs, reduces the risk of postoperative delirium in older patients undergoing hip fracture surgery.

Methods

This dual-center, double-blind, placebo-controlled, parallel-group, pilot study was conducted from August 2023 to January 2024 at two tertiary university hospitals. A total of 132 patients aged ≥ 65 years with an American Society of Anesthesiologists physical status of 2 or 3 scheduled for elective hip fracture surgery were screened and randomized to receive either an anti-inflammatory drug bundle or a placebo. The primary outcome was postoperative delirium, identified within the first three postoperative days. Postoperative blood inflammatory markers and acute pain were measured for mediation analysis.

Results

Of the 132 patients randomized, 123 (93%) completed the trial (mean age, 82 years; 75% women). The prevalence of postoperative delirium was significantly lower in the anti-inflammatory bundle group (15%, 9/62) compared to the placebo group (44%, 27/61) (risk difference, − 30 percentage points [95% CI, − 45 to − 15]; relative risk [RR], 0.33 [95% CI, 0.17 to 0.64]; P = 0.001). No major adverse events were reported in either group. The postoperative CRP level in the anti-inflammatory bundle group was significantly lower (predicted mean difference: − 29.4 [95% CI: − 46.5, − 12.2] mg·L−1; adjusted P < 0.001). Mediation analysis showed a significant indirect association between the anti-inflammatory bundle and postoperative delirium through reduced systemic inflammation (odds ratio [OR], 0.61 [95% CI, 0.26 to 0.87]).

Conclusions

This study demonstrates that a perioperative anti-inflammatory bundle significantly reduces the prevalence of postoperative delirium in older patients undergoing hip fracture surgery, without major side effects. Systemic inflammation mediates the protective effect of the intervention. These findings provide preliminary evidence supporting the anti-inflammatory bundle strategy, paving the way for large-scale multicenter trials to optimize postoperative delirium prevention strategies.

Trial registration

This study was registered in the Chinese Clinical Trial Registry (ChiCTR2300074303) by Ayixia Nawan on August 3, 2023, prior to patient enrollment.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12916-025-03946-x.

Keywords: Aged; Anti-inflammatory; Hip fracture; Perioperative management; Pharmacological interventions; Postoperative delirium

Background

Postoperative delirium (POD) is the most common complication of surgery among older patients1, with an incidence of approximately 50% in those undergoing hip fracture surgeries2–4. POD is linked to cognitive decline, dementia, functional deterioration, and increased mortality1, 5, 6, with healthcare costs reaching $44,291 per patient per year7. Given its preventable nature7, 8, effective strategies are needed, especially for older patients undergoing hip fracture surgeries.

However, effective strategies to prevent POD remain lacking. Current prevention methods9, such as avoiding benzodiazepines, titrating anesthetic depth, and managing pain, are often passive, controversial, and inconsistently effective3, 10–13. Although multidisciplinary non-pharmacologic interventions show promise2, 8, 14, their complexity and variability hinder widespread adoption9, 15. Current pharmacological interventions remain largely focused on pain control, with limited attention to broader inflammatory mechanisms contributing to POD8, 16.

Surgery-evoked systemic inflammation is a key contributor to POD development17, with accumulating evidence suggesting that suppressing inflammation may reduce its incidence18–21. Perioperative drugs such as dexmedetomidine8, 19, 22–24, glucocorticoids20, 21, ulinastatin25, and nonsteroidal anti-inflammatory drugs (NSAIDs)26, 27 have demonstrated anti-inflammatory effects and potential for POD prevention. However, their effectiveness when used individually has been suboptimal, and comprehensive pharmacological approaches targeting systemic inflammation remain underexplored28. Combining these drugs may enhance their anti-inflammatory effects, providing a more holistic strategy to mitigate perioperative inflammation and potentially prevent POD.

Given the complexity and potential risks associated with multidrug regimens, it is critical to conduct a pilot study to assess feasibility and gather preliminary evidence of efficacy. This will lay the groundwork for larger multicenter trials to develop and validate new clinical strategies for POD prevention. In this pilot study, we introduced a perioperative anti-inflammatory bundle strategy combining multiple anti-inflammatory drugs. We hypothesized that this strategy could reduce POD incidence in older patients undergoing hip fracture surgery by mitigating systemic inflammation. Furthermore, we examined the mediation effect of systemic inflammation on the relationship between the anti-inflammatory bundle and POD.

Methods

Study design

This dual-center, double-blind, placebo-controlled, parallel-group study was conducted at two class A tertiary comprehensive hospitals in China. Ethical approval (Approval No.: K2023-139–00) was granted by the Ethics Committee of Beijing Jishuitan Hospital, Beijing, China on May 23, 2023. Written informed consent was obtained from all patients or their legal surrogates. The study was registered in the Chinese Clinical Trial Registry (ChiCTR2300074303) by Ayixia Nawan on August 3, 2023, prior to patient enrollment. This manuscript adheres to the Consolidated Standards of Reporting Trials (CONSORT) guidelines.

Participants

Eligible participants were patients aged 65 years or older, classified as American Society of Anesthesiologists (ASA) physical status 2 or 3, and scheduled for hip fracture surgery. To reduce potential bias from circadian rhythms and ensure adequate intervals between delirium assessments on the day of surgery, only patients whose surgeries were completed in the morning were included. Exclusion criteria included (a) refusal to participate, (b) history of long-term opioid use or alcohol abuse, (c) inability to assess cognitive function, (d) contraindications to study drugs, and (e) cancelations or changes in scheduled surgeries, including intraoperative events leading to significant alterations in the course or duration of the surgery.

Randomization and blinding

The participants were randomly allocated to receive either an anti-inflammatory drug bundle (anti-inflammatory bundle group) or a placebo (control group) in a 1:1 ratio, with randomization stratified by center. A research assistant generated and securely stored the randomization, keeping allocations concealed from participants and investigators until the study’s conclusion. An independent pharmacist team, aware of the group assignments, prepared the trial drugs or placebos before surgery. Patients, anesthesiologists, and outcome assessors were blinded to the group assignments.

Perioperative management and interventions

Patients underwent total hip arthroplasty (THA), hemiarthroplasty, cannulated screw fixation, or proximal femoral nail anti-rotation (PFNA) surgeries at the discretion of the surgeon. All surgeries were performed under routine spinal anesthesia, supplemented with ultrasound-guided fascia iliaca compartment block analgesia. Postoperatively, all participants received patient-controlled intravenous analgesia with sufentanil. Anesthetic techniques and hemodynamic management adhered to local practices, with the use of anticholinergic medications, benzodiazepines, or ketamine prohibited.

In the anti-inflammatory bundle group, patients received perioperative intravenous anti-inflammatory drugs, including (a) a single preoperative dose of dexamethasone (0.1 mg·kg−1); (b) a single preoperative dose of ulinastatin (10,000 U·kg−1); (c) continuous intraoperative and postoperative infusion of dexmedetomidine (0.2–0.5 μg·kg−1·h−1 until the end of surgery and 0.03 μg·kg−1·h−1 for 48 h postoperatively); and (d) a single postoperative dose of flurbiprofen axetil (50 mg) followed by continuous infusion (0.06 mg·kg−1·h−1) for 48 h postoperatively. Dosages and administration methods were determined according to routine clinical practice, with adjustments towards the lower end of the standard range to minimize potential adverse events.

The control group received 0.9% saline and 0.05% lipid emulsion (to mimic flurbiprofen axetil) as a placebo. Pharmacists prepared all trial drugs to ensure that anesthesiologists were blinded to group assignments.

Assessment and outcomes

The primary outcome of this study was POD, defined as delirium identified during the day of surgery (postoperative day 0) and the first 2 days after surgery28. Delirium was identified using the confusion assessment method (CAM). A team of geriatricians, blinded to the group assignments, conducted assessments twice daily on the day of surgery (post-anesthesia care unit and afternoon) and postoperative days 1 and 2 (morning and afternoon). For patients who experienced POD, the Richmond Agitation-Sedation Scale (RASS) and Delirium Rating Scale-Revised-98 (DRS-R-98) severity scale29 were used to assess the motor subtypes and severity of delirium, respectively.

Secondary outcomes included postoperative pain at rest and during activity and the length of hospital stay. A blinded nurse anesthetist team assessed acute pain using the numeric rating scale (NRS) at 12-h intervals during the first 48 h postoperatively30. The time-weighted average NRS16 was calculated for the analysis.

Major adverse events included postoperative acute myocardial infarctions, strokes, gastrointestinal ulcers, acute renal failure, and hypotensive shock.

Baseline clinical data, including ASA physical status, type of fracture and surgery, preoperative pain, preoperative opioid use, anxiety, depression, cognitive function, frailty status, preoperative hemoglobin level, and preoperative cerebrospinal fluid (CSF) lactate concentration, were collected on the day before operation to evaluate their association with POD. Anxiety and depression were assessed using the Hospital Anxiety and Depression Scale (HADS, consisting of two subscales for estimating anxiety [HADS-A] and depression [HADS-D] and both ranging between 0 and 21 points, where HADS-A ≥ 8 indicates anxiety and HADS-D ≥ 8 indicates depression)31. Cognitive function was evaluated using the Montreal Cognitive Assessment (MoCA, ranging between 0 and 30 points, 30 points for best performance). Frailty status was assessed using the FRAIL scale (0 to 5 points, where 0 indicates robust status and ≥ 3 indicates frail status). CSF was attained during spinal anesthesia. Postoperative data that may affect the primary outcomes were recorded to inspect any discrepancies that may still exist after randomization, including intensive care unit (ICU) admission, postoperative rescue opioid use, and postoperative hemoglobin level.

Perioperative blood levels of inflammatory markers, including S100 calcium-binding protein β (S100β), neuron-specific enolase (NSE), and C-reactive protein (CRP), were measured preoperatively and on the operative day and postoperative day 1. For CRP, only the values on postoperative day 1 were analyzed because of missing data on the operative day.

Sample size calculation

According to our previous study on a comparable patient population32, a sample size of 118 patients (59 in each group) was determined based on the assumption that 45% would experience POD and that the anti-inflammatory bundle strategy would reduce the incidence to 20%. It provided 85% power at a 2-sided α level of 0.05 to detect differences. Considering a dropout rate of approximately 10%, we planned to enroll 132 patients (66 patients in each group).

Statistical analysis and mediation analysis

Patients were analyzed according to the randomization group. Multiple imputations were planned for missing data. Statistical analyses included the t-test or Mann–Whitney U test for continuous variables, and the chi-square test or Fisher’s exact test for categorical data. A modified Poisson regression model was used to adjust for the primary outcomes. A subgroup analysis was conducted for the population that developed POD. Sensitivity analyses were established a priori and performed using an alternate definition of the primary outcome which was met if a CAM-identified POD was combined with a DRS-R-98 score greater than 1529. The between-group difference in perioperative inflammatory markers was compared using a two-way analysis of variance (ANOVA) and Sidak’s test.

Least absolute shrinkage and selection operator (LASSO) and logistic regression analyses were used to identify predictors of POD. Parameters related to postoperative pain and inflammatory markers that entered into the final model for predicting POD were used as mediators in mediation analysis. Mediation analysis was conducted using PROCESS (version 4.2 beta; Andrew F. Hayes). Bootstrapping (5000 iterations) was used to estimate the confidence intervals of the indirect effects. Given that the sample size calculation was based on the primary outcome, it may not have provided sufficient power for developing a predictive model or conducting mediation analysis. Therefore, the predictors for POD and the subsequent mediation analysis should be considered preliminary and exploratory.

Statistical significance was set at P < 0.05. IBM SPSS Statistics (version 25.0; IBM Corp., Armonk, NY, USA), Prism (version 8.3.0; GraphPad Software, LLC, Boston, MA, USA), and R software (version 4.2.0; R Foundation for Statistical Computing, Vienna, Austria; https://www.R-project.org) were used for statistical analyses.

Results

Patients

A total of 287 patients were screened for eligibility between August 2023 and January 2024, of whom 132 were randomized. All patients received the assigned intervention, except for six patients whose surgeries were canceled after randomization. During the follow-up period, three patients withdrew consent. Therefore, the final analysis included 123 patients (93%; 62 in the anti-inflammatory bundle group and 61 in the control group) (Fig. 1). No data were missing. The mean age of the patients was 82 ± 7 years, and 92 (75%) were female. Baseline characteristics were comparable between the two groups (Table 1).

Fig. 1
Fig. 1. Flow of participants through the trial and statistical analysis plan. ASA, American Society of Anesthesiologists; POD, postoperative delirium; LASSO, least absolute shrinkage and selection operator
Table 1. Baseline data of patients
CharacteristicAnti-inflammatory bundle (n = 62)Control group (n = 61)
Age (year)84 ± 680 ± 7
Male [n (%)]14 (23)17 (28)
BMI (kg m−2)23 ± 424 ± 3
ASA grade of III or severer [n (%)]58 (94)55 (90)
Type of fracture [n (%)]
Femoral neck fractures52 (84)42 (69)
Intertrochanteric fractures10 (16)19 (31)
Type of surgery [n (%)]
Hemiarthroplasty50 (81)41 (67)
PFNA11 (18)19 (31)
THA1 (2)1 (2)
Pain at rest (NRS)3 [2, 4]3 [2, 4]
Pain on activity (NRS)6 [5, 6]6 [5, 6]
Preoperative opioids [n (%)]30 (48)33 (54)
Anxiety [n (%)]27 (44)26 (43)
Depression [n (%)]18 (29)11 (18)
MoCA26 [25, 27]26 [25, 27]
FRAIL score3 [3, 4]3 [2, 3]
Hemoglobin (g L−1)113 ± 15109 ± 19
CSF lactate (mmol L−1)1.4 [1.0, 2.3]0.9 [0.7, 1.3]
S100β (μg L−1)0.6 [0.5, 0.8]0.7 [0.5, 1.0]
NSE (ng ml−1)11 [9, 15]11 [10, 13]
CRP (mg L−1)9 [ 2, 42]18 [3, 60]

Continuous variables are expressed as mean ± SD or median [25th, 75th percentiles], as appropriate. Categorical data are presented as the number of events (proportion). 

BMI body mass index; ASA American Society of Anesthesiologists, PFNA proximal femoral nail anti-rotation, THA total hip arthroplasty, NRS numeric rating scale, MoCA Montreal Cognitive Assessment, CSF cerebrospinal fluid, NSE neuron-specific enolase, CRP C-reactive protein

Primary outcome

The occurrence of delirium within the first 48 h postoperatively was 15% (9/62) in the anti-inflammatory bundle group and 44% (27/61) in the control group, resulting in a risk difference of − 30 percentage points (95% CI, − 45 to − 15). The anti-inflammatory bundle strategy significantly reduced the risk of POD (RR, 0.33 [95% CI, 0.17 to 0.64]; P = 0.001).

Adjusting for baseline characteristics that could be associated with POD did not change the preventive effect on the primary outcome. The finding was consistent in the sensitivity analysis (Fig. 2, Table 2).

Fig. 2
Fig. 2. Adjusted primary outcome and sensitivity analysis. Patient characteristics included age, sex, and body mass index. Surgical characteristics included type of fracture and surgery. Emotional status encompasses anxiety and depression. Preoperative pain was assessed at rest and during activities. Cognitive function was evaluated using the Montreal Cognitive Assessment (MoCA). Baseline risk factors included MoCA scores, baseline C-reactive protein values, and baseline hemoglobin levels. Sensitivity analysis was performed using an alternate definition of the primary outcome which was met if a CAM-identified POD was combined with a DRS-R-98 score greater than 15. CSF, cerebrospinal fluid; POD, postoperative delirium; RR, risk ratio; CAM, confusion assessment method
Table 2. Outcomes and postoperative data
OutcomesAnti-inflammatory bundle (n = 62)Control group (n = 61)Risk/median differencea [(95% CI), %]Risk ratio (95% CI)P value
Primary outcome
POD [n (%)]9 (14.5)27 (44.3)− 29.7 (− 45.0, − 14.5)0.33 (0.17, 0.64)0.001
Secondary outcomes
Pain at rest (NRS)1.0 [1.0, 1.0]2.0 [1.0, 2.0]− 1.0 (− 1.0, − 0.5)–< 0.001
Pain on activity (NRS)3.0 [3.0, 4.0]4.0 [3.2, 4.0]− 1.0 (− 1.0, − 0.2)–< 0.001
Length of hospital stay (day)4 [4, 5]5 [4, 6]0 [− 1, 0]–0.075
Postoperative data that may affect primary outcome
ICU admission [n (%)]9 (14.5)8 (13.1)1.4 (− 10.8, 13.6)1.11 (0.46, 2.68)1.000
Postoperative opioids [n (%)]5 (8.1)11 (18.0)− 10.0 (− 21.8, 1.8)0.45 (0.17, 1.21)0.115
Hemoglobin (g L−1)108 [100, 115]107 [96, 115]2 (− 3, 7)–0.275

CI confidence interval, POD postoperative delirium, NRS numeric rating scale, ICU intensive care unit

aPseudo-median difference was calculated using the Hodges-Lehmann estimate

In the subgroup of patients who experienced POD, the severity score in the anti-inflammatory bundle group was lower than that in the control group (2518, 33 vs. 3030, 35; median difference, − 6 [95% CI, − 122 to − 0]; P = 0.045). There was no difference in the motor subtypes of delirium between the two groups (Additional file 1: Supplemental Table 1).

Secondary outcomes

The patients in the anti-inflammatory bundle group had lower pain scores at rest (1 [1, 1] vs. 2 [1, 2]; median difference, −1.0 [95% CI, −1.0 to −0.5]) and activity (3 [3, 4] vs. 4 [3, 4]; median difference, −1.0 [95% CI, −1.0 to −0.2]) than those in the control group (Table 2). There was no significant difference between groups regarding length of hospital stay and postoperative characteristics (ICU admission, opioid administration, and hemoglobin level; Table 2). No major adverse events were reported in either of the groups.

Perioperative inflammatory markers

There were no significant differences in S100β and NSE levels between groups (Fig. 3). The postoperative CRP level in the anti-inflammatory bundle group was significantly lower than that in the control group (predicted mean difference: − 29.4 [95% CI: − 46.5, − 12.2] mg·L−1; adjusted P < 0.001; Fig. 3).

Fig. 3
Fig. 3. Perioperative blood levels of inflammatory markers. The data are presented as the median with a 95% confidence interval (CI) and were analyzed using two-way analysis of variance (ANOVA) (with Sidak’s multiple comparisons test when appropriate). A There was no significant difference in S100β levels between groups (F = 0.49, P = 0.483). B There was no significant difference in NSE levels between groups (F = 1.01, P = 0.318). C Postoperative CRP levels in the anti-inflammatory bundle group were significantly lower than those in the control group (predicted mean difference: − 29.4 [95% CI: − 46.5, − 12.2] mg·L.−1, adjusted P < 0.001)

Factors significantly associated with POD

Based on one standard error criterion in the preliminary LASSO regression model, six potential predictors were chosen out of 34 coefficients (Additional file 2: Fig. S1). All six candidates entered the final logistic regression model and were significantly associated with the occurrence of POD (Additional file 1: Supplemental Table 2). These factors included (a) group factor: anti-inflammatory management bundle; (b) baseline factors: preoperative MoCA score, preoperative CRP value, and preoperative hemoglobin level; and (c) postoperative factors: postoperative pain score at rest and postoperative CRP value.

Preliminary mediation analysis

After adjusting for baseline factors predicting POD, a significant indirect association via systemic inflammation was observed between the anti-inflammatory bundle strategy and POD (OR, 0.61 [95% CI, 0.26 to 0.87]; Additional file 2: Fig. S2). However, the indirect association between the anti-inflammatory bundle strategy and POD, mediated by postoperative pain, was not statistically significant (Additional file 1: Supplemental Table 3). In sensitivity analysis, these findings were consistent (Additional file 1: Supplemental Table 4).

Discussion

This study introduced a perioperative anti-inflammatory bundle strategy using routinely used anti-inflammatory drugs, reducing the prevalence of POD to one-third of that in the control group without major adverse events. This significant reduction may be one of the most effective results in POD prevention8, 12, 14, 15, 22, 23, 33. Although the effect could be exaggerated due to the limited sample size, this approach appears promising. Further research involving more centers and patients is warranted.

According to local practice, patients were typically discharged on postoperative day 3 if no adverse events occurred. Therefore, delirium was screened during the first 3 days. Most delirium that occurs after surgery is on postop day 1 and 2. Although this approach meets the recommendation to measure POD for at least 3 days starting on the day of surgery28, the potential underestimation of POD incidence due to the limited follow-up period should be acknowledged. Additionally, preoperative delirium was not assessed in this study. However, screening for delirium preoperatively is often impractical in routine clinical settings, and excluding patients with pre-existing delirium may cause study findings to diverge from real-world conditions. This approach aligned with most reference studies reporting POD prevalence, which similarly did not exclude patients with pre-existing delirium12, 14, 16, 20, 23, 26, 33–35. Therefore, this study focused on the prevalence, rather than the incidence, of delirium, which is more relevant to clinical practice. Although evaluating the effects of an anti-inflammatory strategy on patients with pre-existing delirium was beyond the scope of this study, it remains an area worthy of further investigation.

POD significantly affects quality of life and is strongly associated with increased morbidity and mortality5, 6, especially in older patients2, 23, with rates reaching approximately 50% in those undergoing hip fracture surgeries2–4. Although several strategies have been proposed to prevent POD3, 10–13, non-pharmacological methods are currently recommended by most evidence and guidelines2, 8, 9, 14, 15. However, these complex approaches require multidisciplinary intervention and may result in variable implementation and inconsistent outcomes9, 15, 33.

Pharmacological prophylaxis for preventing POD offers a pragmatic approach, yet current interventions mainly focus on pain control8, 9, 16. Recent literature confirms the pathogenic role of surgery-induced inflammation, highlighting the potential of anti-inflammatory strategies28. Most drugs in this study target postoperative inflammation and alleviate acute pain. Mediation analysis revealed that the association between the interventions and POD is mediated by systemic inflammation rather than acute pain. Both groups achieved effective pain control through fascia iliaca compartment block and patient-controlled intravenous analgesia with sufentanil36. Although there was a significant difference between the groups, acute pain levels remained mild in both. Therefore, the drug bundle’s impact on reducing POD should be interpreted primarily as an anti-inflammatory effect, suggesting that controlling systemic inflammation may help prevent POD and warrants further investigation.

Recent research increasingly highlights inflammation’s role in POD28, with studies documenting changes like circulating cytokines, inflammatory transcripts in the brain, blood–brain barrier impairments, and alterations in microglial cells and astrocytes. While these inflammatory mechanisms are well-established, the broader potential of anti-inflammatory approaches remains underexplored. Combining multiple anti-inflammatory drugs might offer a more effective strategy for managing inflammation and preventing POD. Given the caution required with polypharmacy in older patients, validating the benefits of such combinations through clinical trials is essential. This study found that the combined use of dexmedetomidine, glucocorticoids, ulinastatin, and NSAIDs was significantly more effective in reducing POD prevalence (RR = 0.33). In contrast, when used individually, even the most promising drug, dexmedetomidine, showed weaker effects (RR = 0.60)22, 34. The superior efficacy of the combination likely stems from its ability to target various anti-inflammatory pathways, though further studies are needed to confirm these findings. Additionally, the impact of the duration of dexmedetomidine infusion warrants further investigation.

Future studies should determine the optimal drugs, dosages, and administration methods for the anti-inflammatory bundle. While this pilot study supports the concept, it does not establish the specific drug combination as the definitive formulation. Additional research is needed to identify the most effective and safe drug formulations.

Safety is a key concern with combined drug use. The drugs in this study are routinely used in perioperative practice and are generally safe for patients without contraindications. At our centers, continuous postoperative NSAID infusion has been safely implemented and recommended for years. We followed standard dosing strategies, adjusting to the lower end of the range, and observed no major adverse events. However, further studies are needed to assess detailed adverse event profiles and rates of drug side effects, particularly as new drug formulations are developed.

Anti-inflammatory treatments generally reduce both brain and peripheral inflammation. This study examined blood markers associated with POD, including S100β (a biomarker for blood–brain barrier dysfunction)37, NSE (a marker for neuronal damage)38, and CRP (a common systemic inflammatory marker)39. Only CRP showed significant postoperative changes, suggesting that S100β and NSE levels were likely already elevated preoperatively. This elevation may partly explain the high susceptibility and prevalence of POD in this population. Postoperative CRP levels were significantly lower in the anti-inflammatory bundle group, indicating that the perioperative anti-inflammatory strategy effectively controlled systemic inflammation, thereby reducing POD occurrence. Since the aim of this pilot study was to explore whether systemic inflammation or acute pain mediated the association between the interventions and POD, rather than to identify risk factors for POD, CRP levels were used as a continuous variable in the prediction model and mediation analysis. No attempts were made to establish a threshold for CRP values. The lack of postoperative elevation in S100β and NSE levels could imply the presence of preexisting neuroinflammation and blood–brain barrier dysfunction. However, the strategy may still mitigate POD risk by effectively targeting systemic inflammation under these conditions.

This study focused on older patients undergoing hip fracture surgery, but previous research on the individual drugs used here has shown consistent POD prevention across various surgeries8, 19–23, 25–27. Thus, the perioperative anti-inflammatory bundle strategy could potentially benefit older patients experiencing distinct inflammatory process. However, its effectiveness may vary depending on the surgery, warranting further testing with more diverse samples.

This study has some limitations. First, as a dual-center pilot study with a small sample size, the findings may overestimate the effect. The sample size was not sufficiently powered to identify predictors for POD. However, the primary aim of this pilot study was to evaluate the approach’s effectiveness in preventing POD. The sample size was calculated based on prior analyses32, with a statistical power of 85% to detect a significant effect. Second, the study may have underestimated POD incidence due to the 3-day follow-up period. However, since the most pronounced disturbances usually occur shortly after surgery and the control group had a high prevalence of POD, the primary findings are unlikely to be significantly affected by this limitation. Finally, to reduce potential bias from circadian rhythms and ensure adequate intervals between delirium assessments on the day of surgery, only patients whose surgeries were completed in the morning were included. This limitation precluded analysis of circadian rhythm and sleep quality, factors that may be associated with POD40, potentially restricting the applicability of the findings to situations where these concerns are relevant.

Conclusions

This pilot study demonstrated that a perioperative anti-inflammatory bundle strategy significantly reduced the prevalence of POD by mitigating systemic inflammation, without major adverse events, in older patients undergoing hip fracture surgery. Systemic inflammation was identified as a mediator in the relationship between pharmacological intervention and POD. Larger multicenter trials are needed to confirm these findings and further investigate the role of anti-inflammatory approaches in POD prevention.

Supplementary Information

Supplementary material — available with the version of record.

Supplementary material — available with the version of record.

Abbreviations

POD
Postoperative delirium
NSAID
Nonsteroidal anti-inflammatory drug
CONSORT
Consolidated Standards of Reporting Trials
ASA
American Society of Anesthesiologists
THA
Total hip arthroplasty
PFNA
Proximal femoral nail anti-rotation
CAM
Confusion assessment method
RASS
Richmond Agitation-Sedation Scale
DRS-R-98
Delirium Rating Scale-Revised-98
NRS
Numeric rating scale
CSF
Cerebrospinal fluid
HADS
Hospital Anxiety and Depression Scale
MoCA
Montreal Cognitive Assessment
ICU
Intensive care unit
S100β
S100 calcium-binding protein β
NSE
Neuron-specific enolase
CRP
C-reactive protein
ANOVA
Analysis of variance
LASSO
Least absolute shrinkage and selection operator

Acknowledgements

The authors thank all the anesthesiologists who cared for the patients.

Declarations

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Ayixia Nawan and Zilong Wu contributed equally to this study.

Authors’ contributions

A.N. and Z.W. helped data curation and investigation. B.J. helped data curation, formal analysis, and visualization. B.J. and Y.F. helped conceptualization, methodology, project administration, supervision, and writing. W.Z. and Y.F. helped funding acquisition. G.W. and W.Z. helped project administration, and resources. All authors reviewed the manuscript.

Funding

This work was supported by National Key Research and Development Program of China [grant number 2018YFC2001905]; Beijing Jishuitan Hospital Elite Young Scholar Program [grant number XKGG202116]; and Beijing Municipal Administration of Hospitals Incubating Program [grant number PY202101].

Data availability

The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethical approval (Approval No.: K2023-139–00) was granted by the Ethics Committee of Beijing Jishuitan Hospital, Beijing, China on May 23, 2023. Written informed consent was obtained from all patients or their legal surrogates.

Not applicable.

The authors declare no competing interests.

Sources

References

40 references, in the article's own order. Citation numbers in the text are this list's numbers (39 records are cited).

  1. Six-year cognitive trajectory in older adults following major surgery and delirium

    Kunicki ZJ, Ngo LH, Marcantonio ER, Tommet D, Feng Y, Fong TG, Schmitt EM, Travison TG, Jones RN, Inouye SK

    2023JAMA Intern Med 183(5):442–450 · PMID 36939716doi:10.1001/jamainternmed.2023.0144

  2. Postoperative neurocognitive disorders: a clinical guide

    Dilmen OK, Meco BC, Evered LA, Radtke FM

    2024J Clin Anesth 92:111320 · PMID 37944401doi:10.1016/j.jclinane.2023.111320

  3. Lack of association between perioperative medication and postoperative delirium in hip fracture patients in an orthogeriatric care pathway

    Genet B, Lamy T, Cohen-Bittan J, Glasman P, Verny M, Riou B, Boddaert J, Zerah L

    2022J Am Med Dir Assoc 23(4):623–630.e622 · PMID 34653382doi:10.1016/j.jamda.2021.09.022

  4. Impaired glucose utilization in the brain of patients with delirium following hip fracture

    Titlestad I, Watne LO, Caplan GA, McCann A, Ueland PM, Neerland BE, Myrstad M, Halaas NB, Pollmann CT, Henjum K, et al.

    2024Brain 147(1):215–223 · PMID 37658825doi:10.1093/brain/awad296

  5. Association between postoperative delirium and adverse outcomes in older surgical patients: a systematic review and meta-analysis

    Yan E, Veitch M, Saripella A, Alhamdah Y, Butris N, Tang-Wai DF, Tartaglia MC, Nagappa M, Englesakis M, He D, Chung F

    2023J Clin Anesth 90:111221 · PMID 37515876doi:10.1016/j.jclinane.2023.111221

  6. Postoperative delirium: an independent risk factor for poorer quality of life with long-term cognitive and functional decline after cardiac surgery

    de la Varga-Martínez O, Gutiérrez-Bustillo R, Muñoz-Moreno MF, López-Herrero R, Gómez-Sánchez E, Tamayo E

    2023J Clin Anesth 85:111030 · PMID 36463611doi:10.1016/j.jclinane.2022.111030

  7. One-year medicare costs associated with delirium in older patients undergoing major elective surgery

    Gou RY, Hshieh TT, Marcantonio ER, Cooper Z, Jones RN, Travison TG, Fong TG, Abdeen A, Lange J, Earp B, et al.

    2021JAMA Surg 156(5):430–442 · PMID 33625501doi:10.1001/jamasurg.2020.7260

  8. Which interventions can reduce post-operative delirium in the elderly? Synthesis of multidisciplinary and pharmacological intervention data

    Herman J, Urman RD, Urits I, Kaye AD, Viswanath O

    2022J Clin Anesth 79:110117 · PMID 33144046doi:10.1016/j.jclinane.2020.110117

  9. Evidence-based strategies to reduce the incidence of postoperative delirium: a narrative review

    Swarbrick CJ, Partridge JSL

    2022Anaesthesia 77(Suppl 1):92–101 · PMID 35001376doi:10.1111/anae.15607

  10. The association between midazolam premedication and postoperative delirium - a retrospective cohort study

    Zarour S, Weiss Y, Kiselevich Y, Iacubovici L, Karol D, Shaylor R, Davydov T, Matot I, Cohen B

    2024J Clin Anesth 92:111113 · PMID 37280146doi:10.1016/j.jclinane.2023.111113

  11. Effect of perioperative benzodiazepine use on intraoperative awareness and postoperative delirium: a systematic review and meta-analysis of randomised controlled trials and observational studies

    Wang E, Belley-Côté EP, Young J, He H, Saud H, D'Aragon F, Um K, Alhazzani W, Piticaru J, Hedden M, et al.

    2023Br J Anaesth 131(2):302–313 · PMID 36621439doi:10.1016/j.bja.2022.12.001

  12. Spinal anesthesia with targeted sedation based on bispectral index values compared with general anesthesia with masked bispectral index values to reduce delirium: the SHARP randomized controlled trial

    Brown CH, Edwards C, Lin C, Jones EL, Yanek LR, Esmaili M, Gorashi Y, Skelton R, Kaganov D, Curto R, et al.

    2021Anesthesiology 135(6):992–1003 · PMID 34666346doi:10.1097/ALN.0000000000004015

  13. Pro-con debate: electroencephalography-guided anesthesia for reducing postoperative delirium

    Hao D, Fritz BA, Saddawi-Konefka D, Palanca BJA

    2023Anesth Analg 137(5):976–982 · PMID 37862399doi:10.1213/ANE.0000000000006399

  14. Outcomes of a delirium prevention program in older persons after elective surgery: a stepped-wedge cluster randomized clinical trial

    Deeken F, Sánchez A, Rapp MA, Denkinger M, Brefka S, Spank J, Bruns C, von Arnim CAF, Küster OC, Conzelmann LO, et al.

    2022JAMA Surg 157(2):e216370 · PMID 34910080doi:10.1001/jamasurg.2021.6370

  15. Perioperative optimization of senior health in spine surgery: impact on postoperative delirium

    Pernik MN, Deme PR, Nguyen ML, Aoun SG, Adogwa O, Hall K, Stewart NA, Dosselman LJ, El Tecle NE, McDonald SR, et al.

    2021J Am Geriatr Soc 69(5):1240–1248 · PMID 33382460doi:10.1111/jgs.17006

  16. Effect of acute pain on the association between preoperative cognitive impairment and postoperative delirium: a secondary analysis of three trials

    Ma JH, Liu YF, Hong H, Li CJ, Cui F, Mu DL, Wang DX

    2023Br J Anaesth 130(2):e272–e280 · PMID 35933172doi:10.1016/j.bja.2022.06.033

  17. Resolution of elevated interleukin-6 after surgery is associated with return of normal cognitive function

    Taylor J, Wu JG, Kunkel D, Parker M, Rivera C, Casey C, Naismith S, Teixeira-Pinto A, Maze M, Pearce RA, et al.

    2023Br J Anaesth 131(4):694–704 · PMID 37385855doi:10.1016/j.bja.2023.05.023

  18. Prevention of postoperative cognitive dysfunction by minocycline in elderly patients after total knee arthroplasty: a randomized, double-blind, placebo-controlled clinical trial

    Takazawa T, Horiuchi T, Orihara M, Nagumo K, Tomioka A, Ideno Y, Hayashi K, Yashima H, Araki T, Hatayama K, et al.

    2023Anesthesiology 138(2):172–183 · PMID 36538374doi:10.1097/ALN.0000000000004439

  19. Mitochondrial DNA methylation drift and postoperative delirium in mice

    Liu Y, Song F, Yang Y, Yang S, Jiang M, Zhang W, Ma Z, Gu X

    2022Eur J Anaesthesiol 39(2):133–144 · PMID 34726198doi:10.1097/EJA.0000000000001620

  20. The effect of preoperative methylprednisolone on postoperative delirium in older patients undergoing gastrointestinal surgery: a randomized, double-blind, placebo-controlled trial

    Xiang XB, Chen H, Wu YL, Wang K, Yue X, Cheng XQ

    2022J Gerontol A Biol Sci Med Sci 77(3):517–523 · PMID 34423832doi:10.1093/gerona/glab248

  21. Steroids to reduce the impact on delirium (STRIDE): a double-blind, randomised, placebo-controlled feasibility trial of pre-operative dexamethasone in people with hip fracture

    Kluger MT, Skarin M, Collier J, Rice DA, McNair PJ, Seow MY, Connolly MJ

    2021Anaesthesia 76(8):1031–1041 · PMID 33899214doi:10.1111/anae.15465

  22. Perioperative dexmedetomidine administration to prevent delirium in adults after non-cardiac surgery: a systematic review and meta-analysis

    Qin C, Jiang Y, Lin C, Li A, Liu J

    2021J Clin Anesth 73:110308 · PMID 33930679doi:10.1016/j.jclinane.2021.110308

  23. The effect of peri-operative dexmedetomidine on the incidence of postoperative delirium in cardiac and non-cardiac surgical patients: a randomised, double-blind placebo-controlled trial

    van Norden J, Spies CD, Borchers F, Mertens M, Kurth J, Heidgen J, Pohrt A, Mueller A

    2021Anaesthesia 76(10):1342–1351 · PMID 33960404doi:10.1111/anae.15469

  24. Effects of dexmedetomidine on perioperative brain protection in elderly frail patients

    He Y, Cheng J, Qin H, Shi Y, Yu H, Peng X

    2023J Anesth Transl Med 2(3):29–33doi:10.58888/2957-3912-2023-03-04

  25. Ulinastatin may significantly improve postoperative cognitive function of elderly patients undergoing spinal surgery by reducing the translocation of lipopolysaccharide and systemic inflammation

    Zhang M, Zhang YH, Fu HQ, Zhang QM, Wang TL

    2018Front Pharmacol 9:1007 · PMID 30356711doi:10.3389/fphar.2018.01007

  26. Effect of flurbiprofen axetil on postoperative delirium for elderly patients

    Wang X, Wang Y, Hu Y, Wang L, Zhao W, Wei L, Chen H, Han F

    2019Brain Behav 9(6):e01290 · PMID 31007001doi:10.1002/brb3.1290

  27. Parecoxib supplementation to morphine analgesia decreases incidence of delirium in elderly patients after hip or knee replacement surgery: a randomized controlled trial

    Mu DL, Zhang DZ, Wang DX, Wang G, Li CJ, Meng ZT, Li YW, Liu C, Li XY

    2017Anesth Analg 124(6):1992–2000 · PMID 28525512doi:10.1213/ANE.0000000000002095

  28. Update of the European Society of Anaesthesiology and Intensive Care Medicine evidence-based and consensus-based guideline on postoperative delirium in adult patients

    Aldecoa C, Bettelli G, Bilotta F, Sanders RD, Aceto P, Audisio R, Cherubini A, Cunningham C, Dabrowski W, Forookhi A, et al.

    2024Eur J Anaesthesiol 41(2):81–108 · PMID 37599617doi:10.1097/EJA.0000000000001876

  29. Validation of the delirium rating scale-revised-98: comparison with the delirium rating scale and the cognitive test for delirium

    Trzepacz PT, Mittal D, Torres R, Kanary K, Norton J, Jimerson N

    2001J Neuropsychiatry Clin Neurosci 13(2):229–242 · PMID 11449030doi:10.1176/jnp.13.2.229

  30. A novel methodology to integrate outcomes regarding perioperative pain experience into a composite score: prediction model development and validation

    Jiang B, Liu Y, Wu Y, Mi W, Feng Y

    2022Eur J Pain 26(10):2188–2197 · PMID 36069125doi:10.1002/ejp.2033

  31. Determinants of burnout and other aspects of psychological well-being in healthcare workers during the COVID-19 pandemic: a multinational cross-sectional study

    Denning M, Goh ET, Tan B, Kanneganti A, Almonte M, Scott A, Martin G, Clarke J, Sounderajah V, Markar S, et al.

    2021PLoS One 16(4):e0238666 · PMID 33861739doi:10.1371/journal.pone.0238666

  32. A novel preoperative electroencephalogram-derived index to predict early postoperative delirium in elderly patients after hip fracture surgeries: development of a prediction model

    Nawan A, Wang G, Zhao C, Zhang W, Jiang B, Feng Y

    2024Geriatr Orthop Surg Rehabil 15:21514593241258654 · PMID 38835410doi:10.1177/21514593241258654

  33. The effect of delirium preventive measures on the occurrence of postoperative cognitive dysfunction in older adults undergoing cardiovascular surgery. The DelPOCD randomised controlled trial

    Olotu C, Ascone L, Wiede J, Manthey J, Kuehn S, Scherwath A, Kiefmann R

    2022J Clin Anesth 78:110686 · PMID 35190345doi:10.1016/j.jclinane.2022.110686

  34. Dexmedetomidine for prevention of delirium in elderly patients after non-cardiac surgery: a randomised, double-blind, placebo-controlled trial

    Su X, Meng ZT, Wu XH, Cui F, Li HL, Wang DX, Zhu X, Zhu SN, Maze M, Ma D

    2016Lancet 388(10054):1893–1902 · PMID 27542303doi:10.1016/S0140-6736(16)30580-3

  35. Anaesthetic depth and delirium after major surgery: a randomised clinical trial

    Evered LA, Chan MTV, Han R, Chu MHM, Cheng BP, Scott DA, Pryor KO, Sessler DI, Veselis R, Frampton C, et al.

    2021Br J Anaesth 127(5):704–712 · PMID 34465469doi:10.1016/j.bja.2021.07.021

  36. The influence of involving patients in postoperative pain treatment decisions on pain-related patient-reported outcomes: a STROBE-compliant registering observational study

    Jiang B, Wu Y, Wang X, Gan Y, Wei P, Mi W, Feng Y

    2022Medicine (Baltimore) 101(38):e30727 · PMID 36197159doi:10.1097/MD.0000000000030727

  37. Presence of plasma markers related to brain injury and postoperative delirium in elderly patients: a latent class analysis

    Wang LY, Hu XY, Ji MH

    2024J Clin Anesth 93:111363 · PMID 38134486doi:10.1016/j.jclinane.2023.111363

  38. Elevated neuron-specific enolase level is associated with postoperative delirium and detection of phosphorylated neurofilament heavy subunit: a prospective observational study

    Mietani K, Hasegawa-Moriyama M, Inoue R, Ogata T, Shimojo N, Kurano M, Sumitani M, Uchida K

    2021PLoS ONE 16(11):e0259217 · PMID 34797829doi:10.1371/journal.pone.0259217

  39. Development and validation of an international preoperative risk assessment model for postoperative delirium

    Dodsworth BT, Reeve K, Falco L, Hueting T, Sadeghirad B, Mbuagbaw L, Goettel N, Schmutz Gelsomino N

    2023Age Ageing 52(6):afad086 · PMID 37290122doi:10.1093/ageing/afad086

  40. The effects of dexmedetomidine on postoperative sleep in elderly patients: a systematic review and meta-analysis

    Xiao Y, Tang L, Chen N

    2023Journal of Anesthesia and Translational Medicine 2(3):11–20doi:10.58888/2957-3912-2023-03-02

Article record

The record

Status

Living reprint · journal article Journal version available

Status
Version of record: BMC Medicine 2025
Journal version
BMC Medicine (2025) · doi:10.1186/s12916-025-03946-x
Confirmed
2025-02-21
Licence
CC BY 4.0

This reprint has no DOI of its own: cite the journal article.

The journal's version is the version of record. This page re-typesets it under the article's licence; where they differ, the journal's version is authoritative.

Version history

  1. Published in BMC Medicine2025-02-21
  2. Living reprint on Literature Decoded2026-10-01
  3. Reprinted here2026-10-01
  4. This HTML2026-10-01

Cite this article

Citation
Nawan A, Wu Z, Jiang B, Wang G, Zhang W, Feng Y. Effect of combination of multiple anti-inflammatory drugs strategy on postoperative delirium among older patients undergoing hip fracture surgery: a pilot randomized controlled trial. BMC Med. 2025;23:108. doi:10.1186/s12916-025-03946-x
BibTeX
@article{Nawan2025Effect,
  title     = {Effect of combination of multiple anti-inflammatory drugs strategy on postoperative delirium among older patients undergoing hip fracture surgery: a pilot randomized controlled trial},
  author    = {Ayixia Nawan and Zilong Wu and Bailin Jiang and Geng Wang and Wenchao Zhang and Yi Feng},
  journal   = {BMC Medicine},
  year      = {2025},
  volume    = {23},
  pages     = {108},
  doi       = {10.1186/s12916-025-03946-x},
  pmid      = {39985018},
  publisher = {BMC}
}
RIS
TY  - JOUR
TI  - Effect of combination of multiple anti-inflammatory drugs strategy on postoperative delirium among older patients undergoing hip fracture surgery: a pilot randomized controlled trial
AU  - Ayixia Nawan
AU  - Zilong Wu
AU  - Bailin Jiang
AU  - Geng Wang
AU  - Wenchao Zhang
AU  - Yi Feng
JO  - BMC Medicine
PY  - 2025
VL  - 23
SP  - 108
DO  - 10.1186/s12916-025-03946-x
PB  - BMC
SN  - 1741-7015
UR  - https://doi.org/10.1186/s12916-025-03946-x
ER  -

Cite the original article (the version of record), not this reprint.

Take it with you

One self-contained HTML file: the text, data and figures, readable offline.

Disclaimer and limitations

Educational summary of research findings; not medical advice. Discuss care decisions with a qualified clinician.

The findings apply to the included study populations and may not generalise to every person or setting.

Licence and attribution

This page reproduces Effect of combination of multiple anti-inflammatory drugs strategy on postoperative delirium among older patients undergoing hip fracture surgery: a pilot randomized controlled trial by Ayixia Nawan, Zilong Wu, Bailin Jiang, Geng Wang, Wenchao Zhang, Yi Feng, first published in BMC Medicine 2025;23:108, doi:10.1186/s12916-025-03946-x, PMID 39985018, PMC11846162. © The Author(s) 2025. It is used under the CC BY 4.0 licence.

Changes made:

The words, data, figures and conclusions are the authors’. Literature Decoded is not the publisher of record and does not claim endorsement by the authors or the journal. The version of record is the journal’s; cite it.

Licence statement in the article: “Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.”

Are you an author of this article? Claim it or ask us to take it down.

Cite this article

Living reprint · journal article