Investigation of the Anti-inflammatory Effect of Ketamine in Cardiac Surgery

Participants and Study Design

This was a single-centre, randomised, controlled, single-blind clinical trial conducted at Necmettin Erbakan University Meram Faculty of Medicine Hospital, Türkiye. The study included adults aged 18-80 years who were scheduled to undergo elective cardiac surgery, were able to provide written informed consent, and could reliably communicate their symptoms to the research team.

Exclusion criteria included contraindications to anaesthesia; cognitive impairment or communication difficulties; end-stage renal disease requiring haemodialysis; pregnancy or breastfeeding; known allergy to ketamine; left ventricular ejection fraction below 35%; emergency or repeat cardiac surgery; significant cardiac arrhythmia; morbid obesity; psychiatric disorders; hepatic insufficiency; use of anti-inflammatory medications; preoperative white blood cell count ≥15 × 10³/µL; C-reactive protein (CRP) ≥30 mg/L; and refusal to participate.

Between 30 October 2022 and 30 March 2023, 177 patients were assessed for eligibility. Eleven individuals were excluded before providing written research consent and before randomisation. Consequently, 166 patients provided written informed consent and were randomised to the ketamine group (n = 83) or the control group (n = 83).

Six randomised patients (three in each group) died intraoperatively or within the first 24 postoperative hours, before postoperative day 1 inflammatory biomarker measurements could be obtained. Therefore, the primary biomarker analyses included 160 patients (80 per group). All 166 randomised patients were included in the 28-day all-cause mortality analysis.

The original study protocol (Version 1.0, dated 5 September 2022) received ethics committee approval on 14 September 2022 (decision no. 2022/926). Authorisation to initiate the clinical trial was granted by the Turkish Medicines and Medical Devices Agency on 26 October 2022 (application no. 22-AKD-189).

The study was retrospectively registered at ClinicalTrials.gov on 30 March 2026, after completion of recruitment and data collection.

Randomisation and Blinding

Patients were randomly allocated to the ketamine or control group using a sealed, opaque envelope method involving the drawing of lots, as specified in the original study protocol. Allocation was performed by an investigator who was not otherwise involved in the study.

Patients were blinded to their treatment allocation. The anaesthesia team administering the study intervention was not blinded.

Anaesthesia Protocol

Upon arrival in the operating room, all patients received intravenous premedication with midazolam (1 mg) and fentanyl (50 µg). Standard monitoring, including electrocardiography and peripheral oxygen saturation, was applied. Invasive arterial blood pressure monitoring was established by radial artery catheterisation using a 20-gauge cannula under local anaesthesia.

All weight-based drug doses were calculated using adjusted body weight.

Anaesthesia was induced with intravenous midazolam (0.1 mg/kg), fentanyl (3 µg/kg), and rocuronium (0.6 mg/kg) in both groups.

In addition to the standard induction regimen, patients allocated to the ketamine group received an intravenous ketamine bolus of 1 mg/kg at induction, followed by a continuous intravenous infusion of 2.4 mg/kg/h throughout the intraoperative period. Patients allocated to the control group received standard anaesthesia without additional ketamine administration.

Following adequate muscle relaxation and induction of anaesthesia, endotracheal intubation was performed. A central venous catheter was subsequently inserted through the right internal jugular vein.

Intraoperative analgesic and haemodynamic management was guided by haemodynamic parameters and Surgical Pleth Index (SPI) monitoring. Baseline heart rate, systolic arterial pressure, diastolic arterial pressure, and mean arterial pressure were recorded before induction after premedication. Additional haemodynamic measurements were obtained immediately before intubation, every minute during the first five minutes after intubation, and subsequently at 10-minute intervals throughout surgery.

Anaesthesia was maintained using intravenous remifentanil (0.1-0.3 µg/kg/min) and inhaled sevoflurane (0.5-1 minimum alveolar concentration). An SPI target range of 40-60 was used together with haemodynamic variables to guide intraoperative management. Changes of approximately 20-30% from baseline haemodynamic values were considered clinically relevant.

Hypotension was managed using ephedrine or norepinephrine, as clinically indicated. Hypertension was managed by adjusting the administration of remifentanil and sevoflurane.

Approximately one hour before the end of surgery, all patients received intravenous morphine (0.1 mg/kg) for postoperative analgesia. Intravenous paracetamol (10-15 mg/kg) was administered every eight hours during the postoperative period.

Following surgery, patients were transferred to the intensive care unit (ICU) under mechanical ventilation. Postoperative sedoanalgesia was maintained using remifentanil (0.05-0.1 µg/kg/min) and dexmedetomidine (0.5-1 µg/kg/h) infusions until extubation criteria were met.

Patients were assessed for weaning from mechanical ventilation according to clinical status, blood gas findings, haemodynamic stability, and normothermia. Extubation was performed when the relevant clinical criteria had been fulfilled.

Post-extubation rescue analgesia consisted of intravenous tramadol (1 mg/kg, generally 50-100 mg), administered when required.

Outcome Measures

The original ethics-approved study protocol, dated 5 September 2022, specified two primary outcomes and five secondary outcome measures.

Prespecified primary outcomes

1. Neutrophil-to-lymphocyte ratio (NLR), measured on the morning of postoperative day 1.
2. C-reactive protein (CRP) concentration, measured on the morning of postoperative day 1.

NLR was calculated by dividing the absolute neutrophil count by the absolute lymphocyte count. CRP was measured in mg/L using routine laboratory testing.

Prespecified secondary outcomes

1. Postoperative pain intensity assessed using the Numeric Rating Scale (NRS).
2. Postoperative pain intensity assessed using the Prince Henry Hospital Pain Score (PHHPS).
3. Time to tracheal extubation.
4. Length of ICU stay.
5. Length of hospital stay.

Postoperative pain was assessed using NRS and PHHPS during the early postoperative period. The NRS ranges from 0 (no pain) to 10 (worst imaginable pain). PHHPS is a categorical pain assessment scale ranging from 0 to 4.

Extubation time was recorded in hours, ICU length of stay in hours, and hospital length of stay in days.

Post hoc exploratory outcomes and analyses

Additional inflammatory indices, derived biomarker measures, and selected clinical outcomes were evaluated in analyses introduced after the original study protocol had been established.

Routine preoperative and postoperative day 1 laboratory measurements were used to assess white blood cell, neutrophil, lymphocyte, and platelet counts, together with derived inflammatory indices including platelet-to-lymphocyte ratio (PLR), systemic inflammatory response index (SIRI), and systemic immune-inflammation index (SII).

The derived indices were calculated as follows:

• NLR = neutrophil count / lymphocyte count
• PLR = platelet count / lymphocyte count
• SIRI = neutrophil count × monocyte count / lymphocyte count
• SII = platelet count × neutrophil count / lymphocyte count

Perioperative changes in inflammatory markers were calculated as postoperative minus preoperative values (delta values). Biomarker indices were calculated as postoperative-to-preoperative ratios.

Additional exploratory clinical assessments included intraoperative haemodynamic variables, vasoactive and inotropic support, additional postoperative analgesic requirements, and 28-day all-cause mortality.

Intraoperative vasoactive and inotropic support was quantified using the vasoactive-inotropic score (VIS), calculated as:

VIS = dopamine dose + dobutamine dose + 100 × epinephrine dose + 100 × norepinephrine dose + 10 × milrinone dose + 10,000 × vasopressin dose.

Catecholamine and milrinone doses were expressed in µg/kg/min, and vasopressin in U/kg/min. VIS during cardiopulmonary bypass and at the end of surgery was recorded.

The additional exploratory outcomes and derived analyses were not prespecified as primary outcomes in the original ethics-approved protocol and should be interpreted as hypothesis-generating.

Sample Size Calculation

The original sample size calculation was performed using G*Power before patient recruitment. The documented calculation, dated 15 June 2022, assumed a standardised effect size (Cohen's d) of 0.5537, a one-sided significance level of 0.05, and 95% statistical power, yielding a minimum required sample size of 144 participants (72 per group).

The original study protocol (Version 1.0, dated 5 September 2022) specified a target sample size of 160 participants (80 per group). Ultimately, 166 participants were randomised.

No formal pilot study was conducted to determine the sample size. The source used to derive the assumed effect size of 0.5537 and the specific primary outcome underlying the original calculation could not be verified retrospectively. The previously entered registry statement indicating that the calculation was based on a pilot study of 10 participants per group was incorrect and has been corrected retrospectively.

No adjustment for multiple testing of the two co-primary outcomes was incorporated into the documented sample size calculation.

Statistical Analysis

The original statistical analyses were performed using IBM SPSS Statistics (version 23.0; IBM Corp., Armonk, NY, USA). Additional post hoc analyses and statistical verification were conducted using Python with the SciPy and statsmodels packages.

Continuous variable distributions were assessed using the Shapiro-Wilk test. Normally distributed variables were presented as mean ± standard deviation, whereas non-normally distributed variables were reported as median (interquartile range). Categorical variables were presented as counts and percentages.

Between-group comparisons were performed using independent-samples t-tests for normally distributed continuous variables and Mann-Whitney U tests for non-normally distributed variables. Categorical variables were compared using Pearson's chi-square test or Fisher's exact test, as appropriate.

The reported hypothesis tests were two-sided, with p < 0.05 considered statistically significant in the original analyses. Although the original sample size calculation used a one-sided significance level, the reported outcome analyses were conducted using two-sided tests.

The original protocol specified postoperative day 1 NLR and CRP as the two primary outcomes without a prespecified adjustment for multiple testing. A Bonferroni-adjusted threshold of p < 0.025 was subsequently considered as a post hoc sensitivity analysis.

Additional post hoc analyses of covariance (ANCOVA) were conducted separately for postoperative NLR and CRP, using treatment group as the independent variable and the corresponding preoperative biomarker value as a covariate. Adjusted between-group differences and 95% confidence intervals were estimated using heteroskedasticity-robust (HC3) standard errors.

Further sensitivity analyses were performed using natural logarithm-transformed postoperative biomarker values, adjusted for the corresponding natural logarithm-transformed preoperative values. Treatment coefficients from these models were exponentiated and expressed as adjusted ketamine-to-control ratios with 95% confidence intervals.

Additional inflammatory indices, derived measures, and clinical outcomes introduced after data collection were analysed on an exploratory basis. Their findings were interpreted as hypothesis-generating rather than confirmatory.

The analyses of postoperative day 1 NLR and CRP were restricted to 160 patients with available measurements (80 per group). Six randomised patients died before these measurements could be obtained. Because these missing biomarker outcomes arose from death, they could not be assumed to be missing at random.

All 166 randomised patients (83 per group) were included in the 28-day mortality analysis.

Retrospective Registry Clarification

This study was retrospectively registered on 30 March 2026, after completion of recruitment and data collection. The initial registry record included outcome measures and classifications that were not fully aligned with the original ethics-approved protocol dated 5 September 2022.

The primary and secondary outcome classifications have been clarified with reference to the original protocol. Additional inflammatory indices, derived measures, and exploratory clinical analyses were not prospectively registered as primary or secondary outcomes.

These clarifications describe the differences between the original protocol and the retrospectively entered registry information. They do not constitute prospectively registered amendments to the study design.