Endocrine DOI 10.1007/s12020-017-1399-x LETTER TO THE EDITOR Increase in the circulating endocannabinoid 2-arachidonoylglycerol is associated with gabapentin use in septic ICU patients Hershel Raff 1,2 ● David J Kramer 3 ● Cecilia J. Hillard 4 Received: 19 July 2017 / Accepted: 17 August 2017 © Springer Science+Business Media, LLC 2017 The endocannabinoid signaling system consists of two G- protein-coupled receptors (CBRs), named CB1 and CB2, and two well-characterized ligands, the endocannabinoids (eCBs) 2-arachidonoylglycerol (2-AG) and N-arachido- noylethanolamine (AEA) [1]. Although the roles of these molecules in the circulation are not well understood, their concentrations change under the conditions of physiological and psychological stress [1]. In particular, 2-AG and AEA concentrations have been reported to be increased in patients with endotoxic shock, infection, and following cardiac surgery [1]. To further our understanding of the roles of endocannabinoids in human disease states asso- ciated with severe physiological stress, we compared cir- culating concentrations of 2-AG and AEA in blood obtained from septic patients within 24 h of admission to the ICU to the same measures in healthy controls. We report plasma eCB results in previously character- ized healthy controls (six women/two men; age 57–80 years) and septic ICU patients (five women/four men; age 58–88 years; APACHE score 56.8 [SE 6.2]) [2]. Written consent was obtained from patients and/or their designated family member. All procedures were approved by the Aurora Health Care Human Institutional Review Board (IRB). Causes of sepsis [[N]] were pneumonia [[3]], urinary tract infection [[2]], C. difficile colitis [[1]], diabetic foot ulcer [[1]], cholecystitis [[1]], and pyelonephritis [[1]]. Exclusions included septic shock, use of vasopressors, prior (within 48 h) or current glucocorticoid therapy, or albumin administration. Blood samples were drawn in the morning, immediately placed on ice, processed to plasma in the cold, and frozen within 60 min. Plasma 2-AG and AEA were measured by LC-MS/MS as described previously [3]. Data were evaluated by Mann–Whitney Rank Sum test and linear correlation (Sigmaplot 12.5). Patient characteristics were previously published in this journal [2]. There were no differences in sex distribution, age, or ethnicity between the groups. In the septic ICU patients, 2-AG and AEA concentrations [median (25–75% confidence intervals)] were 16.6 (5.9–130.4) and 1.4 (1.0–3.0) pmol/ml, respectively. The 2-AG concentrations were not normally distributed and were heteroscedastic. In healthy controls, 2-AG and AEA concentrations were 17.3 (10.1–20.3) and 1.4 (1.1–1.5) pmol/L, respectively, and were similar to previous studies of healthy controls [1]. Septic patients receiving gabapentin for pain control had significantly higher circulating concentrations of 2-AG but not AEA (Fig. 1). There were no associations between eCB measurements and indices of the severity of illness (APACHE scores) or other relevant drugs administered in the ICU. In the patients not receiving gabapentin, one was receiving oxycodone and fentanyl and one was receiving metoclopramide. Interestingly, the patient receiving meto- clopramide had previously been treated with pregabalin. In the patients receiving gabapentin, one was receiving * Hershel Raff hraff@mcw.edu 1 Endocrine Research Laboratory, Aurora St. Luke’s Medical Center, Aurora Research Institute, Milwaukee, WI 53215, USA 2 Department of Medicine, Surgery, and Physiology, Medical College of Wisconsin, Milwaukee, WI 53226, USA 3 Aurora Critical Care Service, Aurora St. Luke’s Medical Center, Milwaukee, WI 53215, USA 4 Neuroscience Research Center and Department of Pharmacology and Toxicology, Medical College of Wisconsin, Milwaukee, WI 53226, USA