Secondary Logo

Journal Logo

Geriatric Anesthesia: Original Clinical Research Report

Parecoxib Supplementation to Morphine Analgesia Decreases Incidence of Delirium in Elderly Patients After Hip or Knee Replacement Surgery: A Randomized Controlled Trial

Mu, Dong-Liang MD*; Zhang, Da-Zhi MD; Wang, Dong-Xin MD, PhD*; Wang, Geng MD; Li, Chun-Jing MD*; Meng, Zhao-Ting MD*; Li, Ya-Wei MD*; Liu, Chao MD*; Li, Xue-Ying MS

Author Information
doi: 10.1213/ANE.0000000000002095
  • Free


Delirium is a state of global cerebral dysfunction manifested by acute disturbance of consciousness, attention, cognition, and perception. It develops over a short period of time (usually hours to days) and tends to fluctuate during the course of the day.1 The reported incidence of delirium after joint surgery varied from 3.8% in low-risk patients to 56% in high-risk ones.2–7 The occurrence of postoperative delirium is associated with worse outcomes, including poor functional recovery, prolonged hospital stay, decreased cognitive function, increased health care costs, and elevated mortality rate.8–10

The etiology of postoperative delirium is multifactorial, and pain is considered to be one of the risk factors.11 Indeed, the severity of pain is closely related to the occurrence of postoperative delirium.11,12 In a prospective cohort study, the risk of postoperative delirium was 1.2 times higher for every unit increment in visual analog pain score.13 Similar relationship between the severity of pain and the occurrence of delirium was also found in elderly Chinese patients after hip surgery.14 It is recommended that postoperative pain should be carefully managed to prevent and treat delirium in elderly surgical populations.

Opioids are often used for postoperative analgesia. However, opioids alone are likely insufficient for effective pain relief, especially in patients after orthopedic surgery.15 High-dose opioids are also associated with an increased risk of postoperative delirium16 and other side effects such as respiratory depression and postoperative nausea and vomiting (PONV).17 On the contrary, a prospective observational study of elderly patients undergoing total hip or knee arthroplasty showed that multimodal opioid-sparing analgesia was associated with a reduced incidence of delirium.18

Nonsteroid anti-inflammatory drugs are frequently used as adjuvant to opioids for postoperative analgesia.19 In our previous study, it was found that the supplementation with flurbiprofen axetil (a nonselective cyclooxygenase inhibitor with high binding affinity to the site of lesion) for pain relief significantly improved the quality of analgesia, reduced the consumption of opioid, and decreased the incidence of early postoperative delirium in elderly patients after orthopedic surgery.20 Parecoxib, a selective cyclooxygenase-2 inhibitor, has also been used as a supplement to IV opioids for postoperative pain relief. Studies showed that the administration of parecoxib (40 mg 2 times daily) in addition to morphine significantly improved the quality of analgesia and decreased morphine consumption by 27.8% in patients after total knee arthroplasty and by 40.5% in those after total hip arthroplasty.21,22 We hypothesized that the administration of parecoxib in addition to IV opioid analgesia could also decrease the incidence of postoperative delirium.

The aim of this study was to investigate the effects of parecoxib, when administered as a supplement to patient-controlled IV morphine analgesia, on the incidence of delirium in elderly patients after total hip or knee arthroplasty.


This was a randomized, double-blind, placebo-controlled, parallel-arm, 2-center clinical trial. The study protocol was approved by the local Clinical Research Ethics Committees of Peking University First Hospital (2010–024) and the Institutional Review Board of Chinese Clinical Trial Registry (ChiECRCT-2010022). The study was registered on Chinese Clinical Trial Registry (ChiCTR-TRC-10001101) and was performed in the Departments of Anesthesiology of Peking University First Hospital and Beijing Jishuitan Hospital in Beijing, China. Written inform consent was obtained from all patients enrolled in this study.


Potential participants were screened the day before surgery (or on Friday for those who underwent surgery the next Monday) from the eligible ones by 2 researchers (D.L.M. and D.Z.Z.). The inclusion criteria were elderly patients (age ≥ 60 years) who underwent elective total hip or knee replacement surgery. Patients were excluded if they met any of the following criteria: (1) preoperative delirium; (2) preoperative American Society of Anesthesiologists physical status classification ≥ IV; (3) preoperative New York Heart Association functional classification ≥ III; (4) allergy to nonsteroid anti-inflammatory drugs; (5) severe renal dysfunction (requiring dialysis); (6) severe hepatic dysfunction (Child-Pugh score ≥ 10); (7) previous coronary heart disease (unstable angina and myocardial infarction), cerebrovascular disease (transient ischemic accident and stroke), or peripheral arterial disease (stenosis and occlusion); and (8) active peptic ulcers or inflammatory bowel diseases. The above exclusion criteria, especially those from (3) to (8), were constituted according to the contraindications of parecoxib.

Anesthesia and Postoperative Analgesia

No premedication was administered. Intraoperative monitoring included electrocardiogram, noninvasive blood pressure, pulse oxygen saturation, and urine output. Radial arterial pressure and central venous pressure were monitored when necessary.

Combined spinal-epidural anesthesia was performed for all patients. The L2-3 or L3-4 interspace was selected for epidural and subarachnoid space puncture. Spinal anesthesia was performed by intrathecal injection of 2.5–3 mL of 0.5% hyperbaric bupivacaine (prepared with 2 mL of 0.75% bupivacaine and 1 mL of 10% glucose). Doses of 5–7 mL of 2% lidocaine might be injected into the epidural space for anesthesia maintenance. Propofol (IV infusion at rates from 2 to 4 mg/kg/h) and/or midazolam (IV injection at doses from 10 to 30 μg/kg) could be administered for intraoperative sedation. Fentanyl (IV injection at doses from 20 to 50 μg) was also administered when necessary. The sedative and/or analgesic management in the operating room was left to the discretion of the attending anesthesiologists.

Intraoperative fluid therapy was managed according to routine practice. Packed red blood cells were transfused when necessary to maintain hemoglobin level above 7 g/dL.23 Vasopressors were administered when necessary to maintain systolic blood pressure within 20% of baseline.

For all patients, prophylactic ondansetron (8 mg) was administered at the end of surgery. Prophylactic low-molecular-weight heparin was administered during the first 3 postoperative days to prevent deep venous thrombosis. To avoid the risk of epidural hematoma, the epidural catheter was withdrawn at the end of surgery. Postoperative analgesia was maintained using a patient-controlled IV analgesia pump with morphine for 3 days. The pump was programmed to deliver a 1-mg morphine bolus on demand with a 6-minute lockout interval and a continuous background infusion of 0.5 mg/h of morphine. Supplemental morphine was administered if patient complained of pain after 3 consecutive demands. In such case, 2 mg IV morphine was administrated at 10-minute intervals until the numeric rating scale (NRS) pain score (a 11-point scale where 0 indicates no pain and 10 indicates the worst pain) was less than 4. After the end of patient-controlled IV analgesia, oral morphine was prescribed for additional analgesia when necessary.

Randomization and Study Drug Administration

Randomization codes were generated in a 1:1 ratio, with a block size of 4, and were stratified by the 2 study centers by a biostatistician (X.Y.L.) using the SAS statistical package (SAS 9.2, SAS Institute Inc, Cary, NC).

Patients were randomly assigned into either the parecoxib or control group at the beginning of anesthesia. This was performed by 2 independent researchers (Z.T.M. and C.L.) who prepared the study drugs but did not participate in perioperative patient care or assessment. The study drug was prepared with either parecoxib sodium (Dynastat, Pfizer Inc, New York, NY; 40 mg dissolved in 5 mL normal saline) or placebo (5 mL normal saline) according to the randomization code. All study drugs were identical in appearance and were provided in syringes of the same size and brand. The randomization codes were then sealed in the envelopes again until the end of study.

For patients in the parecoxib group, the first dose of parecoxib 40 mg was administered by attending anesthesiologists at the end of surgery, followed by repeated doses (40 mg per dose) administered by ward nurses at 12-hour intervals until 72 hours after surgery. For those in the control group, placebo (5 mL normal saline) was administered at the same time points.

All patients, health care team members, and study personnel who performed patient recruitment and postoperative assessment (D.L.M. and D.Z.Z.) were unaware of the treatment group allocation throughout the study period.

Outcome Assessment

Prior to the study, 2 researchers (D.L.M. and D.Z.Z.) who performed interview and delirium assessments were trained to follow standard procedures and to use the Confusion Assessment Method (CAM) and the CAM for the Intensive Care Unit (CAM-ICU) by a psychiatrist. Both CAM and CAM-ICU detect 4 features of delirium: (1) acute onset of mental status changes or a fluctuating course; (2) inattention; (3) disorganized thinking; and (4) altered level of consciousness. To be diagnosed as delirious, a patient must display features 1 and 2, with either 3 or 4.24,25

During preoperative interview, demographic characteristics, preoperative comorbidity, admission diagnosis, and other baseline data were obtained. Functional ability was evaluated with the use of the Barthel Index of Activities of Daily Living (score range, 0–100, with 0 representing a totally dependent bed-ridden state, and 100 indicating that a patient is fully independent in physical functioning).26 The Mini-Mental State Examination (MMSE; score range, 0–30, with lower scores indicating poorer performance)27 and the CAM were used to assess cognitive function and delirium, respectively.

Postoperative delirium was assessed with the use of CAM twice daily (from 6 to 8 am and from 6 to 8 pm) during the first 5 days after surgery. For patients with endotracheal intubation in the ICU, delirium assessment was performed with the use of CAM-ICU.25 Immediately before assessing delirium with CAM-ICU, the level of sedation/agitation was assessed using the Richmond Agitation Sedation Scale (RASS).28 If the patient was too deeply sedated or unarousable (−4 or −5 on the RASS), delirium assessment was aborted and the patient was recorded as comatose. If the RASS score was greater than −4 (−3 through +4), delirium was assessed using the CAM-ICU. For patients who withdrew consents from study participation or those who were discharged or died within 5 days after surgery, the results of the last delirium assessment were considered the results of the missing data.

A previous study showed that higher pain score at rest, but not that with movement, was associated with an increased risk of delirium over the first 3 postoperative days.13 Pain score at the rest was therefore assessed with the use of NRS at 24, 48, and 72 hours after surgery. Cumulative morphine consumption was recorded at same time intervals.19–22 Cognitive function was assessed with the use of MMSE on the first, third, and fifth day after surgery. Postoperative cognitive impairment was defined as a reduction of postoperative MMSE score ≥ 2 from preoperative value.29 PONV, which was defined as any nausea, emetic episodes (retching or vomiting), or both occurred during the first 5 days after surgery, was monitored at same time intervals. Other postoperative complications, which were defined as any conditions that required therapeutic intervention, were monitored daily during the first 5 days after surgery and then weekly until 28 days after surgery. Barthel Index was evaluated at hospital discharge. All-cause 28-day mortality was recorded.

Statistical Analysis

Outcome Analysis.

Primary end point was the incidence of delirium during the first 5 days after surgery, ie, the percentage of patients who developed at least 1 episode of delirium during that period. Secondary end points included the NRS pain score and cumulative morphine consumption after surgery, the duration of delirium, the incidence of postoperative cognitive impairment on the first, third, and fifth day after surgery, as well as the Barthel Index at hospital discharge. Additional outcomes included the incidence of PONV, the occurrence of other postoperative complications, the length of stay in hospital after surgery, and all-cause 28-day mortality.

Outcome data and safety were analyzed in the intention-to-treat population. We also performed per-protocol analysis for the primary end point. The tests were used to assess the treatment effect of parecoxib versus placebo. Continuous data with normal distribution were compared using independent sample t test. Continuous data with asymmetric distribution (duration of delirium) and ordinal data (NRS pain score) were compared using the independent-sample Mann-Whitney U test. Categorical data (incidence of postoperative delirium, incidence of PONV, incidence of postoperative cognitive impairment, and occurrence of other postoperative complications) were compared using χ2 test or continuity correction χ2 test. The difference (and 95% confidence interval for the difference) between 2 medians is estimated using the methodology of Hodges-Lehmann. Time-to-event data (postoperative in-hospital stay) was analyzed by Cox regression model.

Statistical analyses were performed with the SPSS 14.0 software (SPSS, Inc, Chicago, IL) and SAS 9.2 software (SAS Institute, Cary, NC). All tests were 2-tailed, and P values of less than .05 were considered to be statistically significant. The Bonferroni adjustment was made to control type I error for multiple testing.

Sample Size Calculation.

In our previous study, delirium occurred in 26.2% of patients in the control group (sufentanil analgesia) and in 16.3% of those in the treatment group (combined sufentanil and flurbiprofen axetil analgesia) after orthopedic surgery.20 We assumed that the addition of parecoxib to morphine analgesia produced a similar effect in reducing the incidence of postoperative delirium when compared to morphine analgesia alone, ie, a 38% reduction from the control group. The calculated sample size that would provide 80% power to detect this difference based on a 2-tailed significance level of 0.05 was 267 patients per group (STATA 10.0 software, StataCorp LP, College Station, TX). Considering an estimated follow-up missing rate of about 15%, the final sample size was 310 in each group.


Patient Population

Table 1.:
Demographics and Baseline Characteristics
Table 2.:
Intraoperative Data
Figure 1.:
Flowchart of the study. ITT indicates intention-to-treat; NSAIDs, nonsteroid anti-inflammatory drugs; NYHA, New York Heart Association; TIA, transient ischemic attack. aAt least 1 dose of nonstudy nonsteroid anti-inflammatory drugs (50 mg flurbiprofen axetil or 40 mg parecoxib sodium) was administered during the first 3 postoperative days.

Between January 2011 and May 2013, 772 patients were screened for study participation; 673 patients met the inclusion/exclusion criteria and were invited to take part; 620 patients gave consents and were randomized into the study. During the study period, 11 patients withdrew consents (study drug administration was stopped in these patients, 7 in the control group and 4 in the parecoxib group) and 27 patients deviated from the study protocol (at least 1 dose of nonstudy nonsteroid anti-inflammatory drugs was administered during the first 3 postoperative days, 18 in the control group and 9 in the parecoxib group). These patients were included in the final intention-to-treat analyses (Figure 1) but excluded from the per-protocol analysis. The 2 groups were well matched for baseline and perioperative variables (Tables 1 and 2).

Effectiveness Analysis

The incidence of delirium during the first 5 postoperative days was reduced from 11.0% (34/310) in the control group to 6.1% (19/310) in the parecoxib group (relative risk [RR] 0.56, 95% confidence interval [CI], 0.33–0.96; P = .031, Table 3, Figure 2). Per-protocol analysis also demonstrated a significant reduction in the incidence of delirium between groups (11.6% [34/292] with placebo versus 6.0% [18/301] with parecoxib; RR 0.51, 95% CI, 0.30–0.89, P = .026). Post hoc analyses showed that daily prevalence of delirium was significantly lower in the parecoxib group than in the placebo group on postoperative days 2 (RR 0.31, 95% CI, 0.12–0.84; P = .021, Figure 2). The incidence of postoperative coma or delirium was lower in the parecoxib group than in the placebo group (11.3% [35/310] with placebo versus 6.1% [19/310] with parecoxib; RR 0.54, 95% CI, 0.32–0.93; P = .023, Table 3).

Table 3.:
Effectiveness Outcomes (Intention-to-Treat Analyses)
Figure 2.:
Daily prevalence of postoperative delirium. The prevalence of delirium on postoperative day 2 (RR 0.31, 95% CI, 0.12–0.84; P = .021) and the overall incidence of delirium during the first 5 postoperative days (RR 0.56, 95% CI, 0.33–0.96; P = .031) was significantly lower in the parecoxib group than in the control group. CI indicates confidence interval; RR, relative risk.

NRS pain scores at rest at 24, 48, and 72 hours after surgery were significantly lower in the parecoxib group than in the control group (all P < .001). Cumulative morphine consumptions at the above time points were also significantly less in the parecoxib group than in the control group (all P < .001). The incidence of postoperative cognitive impairment on the first, third, and fifth postoperative days was significantly lower in the parecoxib group than in the control group (all P < .001). The incidence of PONV was also significantly lower in the parecoxib group than in the placebo group (11.0% [34/310] with placebo versus 6.5% [20/310] with parecoxib; RR 0.59, 95% CI, 0.35–0.99; P = .046, Table 3).

Safety Analyses

Table 4.:
Safety Outcomes

There were no significant differences between the 2 groups with regard to the incidence of postoperative complications (excluding delirium and PONV) during the first 28 days after surgery. One patient in the control group developed sudden cardiac arrest during surgery (before study drug administration) and died on the third day after surgery. No other death occurred during the 28-day follow-up period after surgery (Table 4).


Our study found that, in elderly patients after total hip or knee replacement surgery, parecoxib-supplemented IV morphine analgesia during the first 3 days after surgery decreased the incidence of postoperative delirium. Parecoxib supplementation improved analgesia quality and reduced morphine consumption after surgery, although the differences were small. It also lowered the occurrence of PONV.

In the present study, 40 mg of parecoxib sodium was administered intravenously twice daily for 3 consecutive days after surgery. Previous studies showed that this dosing regimen significantly improves postoperative pain management by ameliorating analgesic effect and reducing opioid requirement;21,22 and dose adjustment is not needed for the elderly (age ≥ 65 years).30 A recent meta-analysis reported that the combination of patient-controlled analgesia with parecoxib sodium (no longer than 3 days) significantly improved pain management up to 72 hours after surgery.31 Our study also found that parecoxib supplementation to morphine analgesia lowered the NRS pain score and decreased the cumulative morphine consumption (a 9.2% decrease compared to the control group). However, although statistically significant, the magnitudes of these decreases were smaller as compared to the previously reported results.21,22 These may be attributed to the following reasons. First, combined spinal-epidural anesthesia was used in our patients. The residual effects might have decreased the severity of pain and morphine consumption. Second, we did not assess pain score during the early postoperative period (ie, within 24 hours after surgery) and thus might have missed results of the most severe pain. Third, a low-dose background infusion was provided in our patient-controlled analgesia regimen, which might have improved analgesic effect and decreased the differences between groups.32

Since most cases of delirium had their onset within the first 3 postoperative days33 and most of our patients were discharged from hospital within 1 week after surgery, we monitored delirium until the fifth postoperative day. Our study found that postoperative delirium occurred in 11.0% of patients in the control group. This is in line with incidences reported in similar patient populations,3,4,6 but is much lower than the result in our previous study, which showed that delirium occurred in 26.2% of elderly patients in the control group after orthopedic surgery.20 Possible reasons that led to a lower incidence of postoperative delirium in the present study may include the following. First, the patients in this study were healthier than those in our previous one. This was evidenced by the facts that the percentage of patients with American Society of Anesthesiology classification ≥ III was lower and the score of preoperative MMSE was higher in patients of the present study than in those of our previous one (4.5% vs 14.2% and [28 ± 2] vs [27 ± 3], respectively).20 This lowers the risk of developing delirium. Second, the combined spinal-epidural anesthesia was performed for all patients in the present study, whereas 47.5% of patients underwent general anesthesia in our previous one.20 A systematic review suggested that regional anesthesia may be associated with decreased risk of early postoperative cognitive complication compared to general anesthesia in patients undergoing arthroplastic surgery.34 Third, all patients underwent elective joint-replacement surgery in the present study, whereas 55.1% of patients underwent surgery other than elective joint replacement (including hip fracture surgery) in our previous study.20 It has been reported that the incidence of postoperative delirium is lower in elective hip-replacement patients than in hip fracture ones.2,3

In a large randomized controlled trial of 1062 patients after major noncardiac surgery, Langford et al35 reported that inclusion of cyclooxygenase-2 inhibitors (parenteral parecoxib for 3 days then oral valdecoxib for 7 days) in the postoperative analgesia regime significantly reduced the incidence of patient-reported confusion on the second postoperative day (1.8% vs 5%; P = .006) compared to placebo. However, they did not evaluate the occurrence of clinically defined delirium. In a small randomized controlled trial of 80 patients undergoing emergent femoral head replacement surgery, Li et al36 reported that administration of parecoxib (20/40 mg every 12 hours for 3 consecutive days) in addition to opioid analgesia significantly reduced the incidence of postoperative delirium (22.5% vs 45.0%; P < .05) and the prevalence of cognitive dysfunction from day 3 to 6 months after surgery when compared to placebo. Our study with a relative large sample size found that, in elderly patients after elective hip or knee replacement surgery, parecoxib supplementation to morphine analgesia produced a significant reduction (estimated at 44%) in the incidence of postoperative delirium. We also found that parecoxib supplementation significantly decreased the incidences of postoperative cognitive impairment on the first, third, and fifth postoperative days.

The possible mechanisms by which parecoxib decreased delirium occurrence and relieved cognitive impairment in elderly patients after surgery remains unclear but may be due to the following ones. First, the use of parecoxib improves postoperative analgesia. Several studies have shown that high pain scores are associated with an increased risk of postoperative delirium and cognitive decline, but these studies were all observational in design and thus the causal role for pain was unclear.11–14 In this study (similar to results of our prior study20), parecoxib supplementation to IV opioid significantly improved analgesia and decreased the incidence of delirium in elderly patients after joint arthroplasty surgery, thus supporting the hypothesis that pain may in fact contribute to postoperative delirium. Second, parecoxib decreases the cumulative consumption of morphine. Studies found that high opioid consumption was associated with increased risk of postoperative delirium,16 whereas the use of nonopioid postoperative pain management techniques and/or multimodal analgesia, which improved analgesia with less opioid exposure, decreased the occurrence of postoperative cognitive complications.37 Third, parecoxib alleviates the inflammatory response provoked by surgery. It is suggested that central inflammation may play an important role in the pathogenesis of delirium and dementia in the elderly,38 whereas parecoxib inhibited surgery-induced central inflammation in a preclinical study.39

In line with the previous studies,22,29 we also found that the supplement of parecoxib to opioid analgesia decreased the incidence of PONV. A major concern in using cyclooxygenase-2 selective inhibitors for postoperative pain relief in elderly patients is a potentially increasing risk of cardiovascular events, especially when they were used in high-risk patients (such as those undergoing coronary artery bypass grafting surgery) for a prolonged duration (IV parecoxib followed by oral valdecoxib for 10 days).40 In the present study, patients with significant preoperative cardiovascular disease were excluded, and the parecoxib therapy was limited to the first 3 days after surgery. Our results showed that the incidence of new-onset atrial fibrillation was slightly increased in the parecoxib group, but the difference was not statistically significant (P = .061). In a previous meta-analysis of 114 randomized trials with 116,094 participants, rofecoxib (another cyclooxygenase-2 selective inhibitor) is associated with increased risk of arrhythmia.41 Therefore, despite the fact that the occurrences of cardiovascular events as well as the overall incidence of postoperative complications were not different between the 2 groups, the risk of arrhythmia deserves further study.

It is worth pointing out that, despite a reduction in delirium and pain, there were no differences in the length of stay in hospital after surgery and the total medical expense in our patients. This may be because that the postoperative length of stay was relatively long in this population (median 6 days, interquartile range 6–8). Since, in our control group patients, 70.6% of delirium developed during the first 2 postoperative days with a median duration of 1 day (interquartile range 1–1.75), the reduction of delirium might have more impact on hospital outcomes where length of stay is shorter. For example, many US-based hospitals have length of stay of about 2–3 days after arthroplasty.42 Reducing delirium during postoperative days 1–2 may increase the possibility of early hospital discharge.

There are several limitations of our study. First, because of the contraindications of parecoxib, a majority of high-risk patients, especially those with cardiovascular disease, were eliminated from the study. And because of the epidemiologic characteristics of the degenerative osteoarthritis, most of our enrolled patients were women. These limited the generalizability of the results of this study. Second, exclusion of high-risk patients also lowered the incidence of postoperative delirium, which may lower the power of the study to detect more significant differences between the 2 groups. However, we did find differences between the 2 groups. Third, among all enrolled patients, 11 withdrew consents after randomization and 27 deviated from the study protocol. These may produce interference on the results. Fourth, clinical practices in the 2 centers where this study was performed might be different from other institutions. For example, the length of hospital stay in our centers was much longer than in U.S. hospitals, indicating differences in some particular practice. These may also limit the generalizability of our study.


Our study found that, in low-risk elderly patients after total hip or knee replacement surgery, multidose parecoxib supplemented to IV morphine analgesia decreased the incidence of postoperative delirium without increasing adverse events.


The authors gratefully acknowledge Dr Xin-Yu Sun (MD, Professor, Department of Psychiatrics, Peking University Sixth Hospital, Beijing, China) for her help in psychiatric consultation and Dr Daqing Ma (MD, PhD, Macintosh Professor, Department of Anaesthetics, Pain Medicine, and Intensive Care, Imperial College London, London, UK) for his critical comments during the manuscript preparation stage.


Name: Dong-Liang Mu, MD.

Contribution: This author helped perform patient recruitment, data collection and data analysis, and helped draft the manuscript.

Name: Da-Zhi Zhang, MD.

Contribution: This author helped perform patient recruitment, data collection and data analysis, and helped draft the manuscript.

Name: Dong-Xin Wang, MD, PhD.

Contribution: This author helped design the study and perform data analysis and final revision of the manuscript. He was Chief Investigator of the project.

Name: Geng Wang, MD.

Contribution: This author participated in the study design. He was Principal Investigator at Beijing Jishuitan Hospital.

Name: Chun-Jing Li, MD.

Contribution: This author helped perform patient recruitment and data collection.

Name: Zhao-Ting Meng, MD.

Contribution: This author helped perform study drug preparation and data collection.

Name: Ya-Wei Li, MD.

Contribution: This author helped perform patient recruitment and data collection.

Name: Chao Liu, MD.

Contribution: This author helped perform study drug preparation and data collection.

Name: Xue-Ying Li, MS.

Contribution: This author helped perform statistical analyses.

All the authors contributed to the critical revision of the manuscript and read and approved the final manuscript.

This manuscript was handled by: Robert Whittington, MD.


1. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 2013.5th ed. Arlington, VA: American Psychiatric Association.
2. Galanakis P, Bickel H, Gradinger R, Von Gumppenberg S, Förstl H. Acute confusional state in the elderly following hip surgery: incidence, risk factors and complications. Int J Geriatr Psychiatry. 2001;16:349–355.
3. Kalisvaart KJ, Vreeswijk R, de Jonghe JF, van der Ploeg T, van Gool WA, Eikelenboom P. Risk factors and prediction of postoperative delirium in elderly hip-surgery patients: implementation and validation of a medical risk factor model. J Am Geriatr Soc. 2006;54:817–822.
4. Priner M, Jourdain M, Bouche G, Merlet-Chicoine I, Chaumier JA, Paccalin M. Usefulness of the short IQCODE for predicting postoperative delirium in elderly patients undergoing hip and knee replacement surgery. Gerontology. 2008;54:116–119.
5. Juliebø V, Bjøro K, Krogseth M, Skovlund E, Ranhoff AH, Wyller TB. Risk factors for preoperative and postoperative delirium in elderly patients with hip fracture. J Am Geriatr Soc. 2009;57:1354–1361.
6. Jankowski CJ, Trenerry MR, Cook DJ, et al. Cognitive and functional predictors and sequelae of postoperative delirium in elderly patients undergoing elective joint arthroplasty. Anesth Analg. 2011;112:1186–1193.
7. Lee HB, Mears SC, Rosenberg PB, Leoutsakos JM, Gottschalk A, Sieber FE. Predisposing factors for postoperative delirium after hip fracture repair in individuals with and without dementia. J Am Geriatr Soc. 2011;59:2306–2313.
8. Edelstein DM, Aharonoff GB, Karp A, Capla EL, Zuckerman JD, Koval KJ. Effect of postoperative delirium on outcome after hip fracture. Clin Orthop Relat Res. 2004;422:195–200.
9. Kat MG, Vreeswijk R, de Jonghe JF, et al. Long-term cognitive outcome of delirium in elderly hip surgery patients. A prospective matched controlled study over two and a half years. Dement Geriatr Cogn Disord. 2008;26:1–8.
10. Kat MG, de Jonghe JF, Vreeswijk R, et al. Mortality associated with delirium after hip-surgery: a 2-year follow-up study. Age Ageing. 2011;40:312–318.
11. Vaurio LE, Sands LP, Wang Y, Mullen EA, Leung JM. Postoperative delirium: the importance of pain and pain management. Anesth Analg. 2006;102:1267–1273.
12. Fong HK, Sands LP, Leung JM. The role of postoperative analgesia in delirium and cognitive decline in elderly patients: a systematic review. Anesth Analg. 2006;102:1255–1266.
13. Lynch EP, Lazor MA, Gellis JE, Orav J, Goldman L, Marcantonio ER. The impact of postoperative pain on the development of postoperative delirium. Anesth Analg. 1998;86:781–785.
14. Nie H, Zhao B, Zhang YQ, Jiang YH, Yang YX. Pain and cognitive dysfunction are the risk factors of delirium in elderly hip fracture Chinese patients. Arch Gerontol Geriatr. 2012;54:e172–e174.
15. Parvizi J. Pain management following total joint arthroplasty: making strides. J Bone Joint Surg Am. 2012;94:1441.
16. Leung JM, Sands LP, Paul S, Joseph T, Kinjo S, Tsai T. Does postoperative delirium limit the use of patient-controlled analgesia in older surgical patients? Anesthesiology. 2009;111:625–631.
17. Mizuno J, Morita S, Hanaue N, Hanaoka K, Yokoyama T. Masui. 2011;60:908–912.
18. Krenk L, Rasmussen LS, Hansen TB, Bogø S, Søballe K, Kehlet H. Delirium after fast-track hip and knee arthroplasty. Br J Anaesth. 2012;108:607–611.
19. Elia N, Lysakowski C, Tramèr MR. Does multimodal analgesia with acetaminophen, nonsteroidal antiinflammatory drugs, or selective cyclooxygenase-2 inhibitors and patient-controlled analgesia morphine offer advantages over morphine alone? Meta-analyses of randomized trials. Anesthesiology. 2005;103:1296–1304.
20. Shen L, Liu P, Zhang P, Zhang X, Cui J. Characterization of periostin expression in human endometrium and endometriotic lesions. Gynecol Endocrinol. 2012;28:815–818.
21. Hubbard RC, Naumann TM, Traylor L, Dhadda S. Parecoxib sodium has opioid-sparing effects in patients undergoing total knee arthroplasty under spinal anaesthesia. Br J Anaesth. 2003;90:166–172.
22. Malan TP Jr, Marsh G, Hakki SI, Grossman E, Traylor L, Hubbard RC. Parecoxib sodium, a parenteral cyclooxygenase 2 selective inhibitor, improves morphine analgesia and is opioid-sparing following total hip arthroplasty. Anesthesiology. 2003;98:950–956.
23. Tian YK, Yue Y, Yao SL, Huang WQ, Ye TH, Wu XM. Chinese Society of Anesthesiology: guideline of perioperative blood transfusion. Lin Chuang Ma Zui Xue Za Zhi. 2009; 3:189–191.
24. 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;113:941–948.
25. Ely EW, Inouye SK, Bernard GR, et al. Delirium in mechanically ventilated patients: validity and reliability of the confusion assessment method for the intensive care unit (CAM-ICU). JAMA. 2001;286:2703–2710.
26. Mahoney FI, Barthel DW. Functional evaluation: the Barthel index. Md State Med J. 1965;14:61–65.
27. 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;12:189–198.
28. Sessler CN, Gosnell MS, Grap MJ, et al. The Richmond Agitation-Sedation Scale: validity and reliability in adult intensive care unit patients. Am J Respir Crit Care Med. 2002;166:1338–1344.
29. Xu T, Bo L, Wang J, et al. Risk factors for early postoperative cognitive dysfunction after non-coronary bypass surgery in Chinese population. J Cardiothorac Surg. 2013;8:204.
30. Instruction of parecoxib sodium for injection (version 20160906). Available at: Accessed January 23, 2017.
31. Wei W, Zhao T, Li Y. Efficacy and safety of parecoxib sodium for acute postoperative pain: a meta-analysis. Exp Ther Med. 2013;6:525–531.
32. White I, Ghinea R, Avital S, Chazan S, Dolkart O, Weinbroum AA. Morphine at “sub-analgesic” background infusion rate plus low-dose PCA bolus control pain better and is as safe as twice a bolus-only PCA regimen: a randomized, double blind study. Pharmacol Res. 2012;66:185–191.
33. Shi CM, Wang DX, Chen KS, Gu XE. Incidence and risk factors of delirium in critically ill patients after non-cardiac surgery. Chin Med J (Engl). 2010;123:993–999.
34. Zywiel MG, Prabhu A, Perruccio AV, Gandhi R. The influence of anesthesia and pain management on cognitive dysfunction after joint arthroplasty: a systematic review. Clin Orthop Relat Res. 2014;472:1453–1466.
35. Langford RM, Joshi GP, Gan TJ, et al. Reduction in opioid-related adverse events and improvement in function with parecoxib followed by valdecoxib treatment after non-cardiac surgery: a randomized, double-blind, placebo-controlled, parallel-group trial. Clin Drug Investig. 2009;29:577–590.
36. Li JZ, Li XZ, Wang XM, et al. Effects of parecoxib sodium analgesia on serum concentrations of neuron-specific enolase and S-100β and postoperative cognitive function of elderly patients undergoing acute replacement of femoral head. Zhonghua Yi Xue Za Zhi. 2013;93:2152–4.
37. Halaszynski TM. Pain management in the elderly and cognitively impaired patient: the role of regional anesthesia and analgesia. Curr Opin Anaesthesiol. 2009;22:594–599.
38. Simone MJ, Tan ZS. The role of inflammation in the pathogenesis of delirium and dementia in older adults: a review. CNS Neurosci Ther. 2011;17:506–513.
39. Peng M, Wang YL, Wang FF, Chen C, Wang CY. The cyclooxygenase-2 inhibitor parecoxib inhibits surgery-induced proinflammatory cytokine expression in the hippocampus in aged rats. J Surg Res. 2012;178:e1–e8.
40. Nussmeier NA, Whelton AA, Brown MT, et al. Complications of the COX-2 inhibitors parecoxib and valdecoxib after cardiac surgery. N Engl J Med. 2005;352:1081–1091.
41. Zhang J, Ding EL, Song Y. Adverse effects of cyclooxygenase 2 inhibitors on renal and arrhythmia events: meta-analysis of randomized trials. JAMA. 2006;296:1619–1632.
42. Duellman TJ, Gaffigan C, Milbrandt JC, Allan DG. Multi-modal, pre-emptive analgesia decreases the length of hospital stay following total joint arthroplasty. Orthopedics. 2009;32:167.
Copyright © 2017 International Anesthesia Research Society