Anesthetic agents, including opioids can modulate the altered immune function in patients with obesity through mechanisms that involve the expression and release of cytokines. For this reason, anesthetic care in patients with obesity remains controversial. Therefore, the aim of the study was to compare the effect of opioid-containing anesthesia vs opioid-free anesthesia using the Cortínez-Sepúlveda model on serum levels of IL-6, IL-1β and TNF-α before and after surgery in obese patients undergoing bypass surgery.

Methods: A randomized cross-sectional study of 40 unrelated obese adults was performed in the Anesthesiology and Bariatric Surgery Service at the Civil Hospital of Guadalajara "Dr. Juan I. Menchaca". Before undergoing laparoscopic Roux-en-Y gastric bypass, patients were randomly assigned to two anesthesia groups: opiod-containing (n=20) or opioid-free (n=20). The opioid used in the opioid-containing anesthesia group was fentanyl. To characterize the disposition of intravenous propofol for the target-controlled infusion technique in obese patients, the Cortínez-Sepúlveda pharmacokinetic model was used. Body mass was determined to the nearest 0.05kg using a balance scale (Seca 703; Seca, Hamburg, Germany). Blood samples were taken before and immediately after surgery and cytokine serum levels were determined with ELISA kits. Statistical analyses were performed using the IBM Statistical Package for the Social Sciences software package version 20.0 (IBM Corp., Armonk, NY, USA).

Inclusion Criteria:

• Who were scheduled for bypass surgery at the Service of Anesthesiology and Bariatric Surgery of the Civil Hospital of Guadalajara "Dr. Juan I Menchaca", Jalisco, Mexico, were recruited.

Exclusion Criteria:

• Patients with a history of ischemic heart disease, history of drug abuse, and with any known allergy to any of the drugs used during anesthesia. Elimination criteria were: patients who withdrew their consent or with insufficient and poor quality blood samples (coagulated) or other reasons that did not allow sample processing.