S100B as a biomarker of blood-brain barrier disruption after thoracoabdominal aortic aneurysm repair: a secondary analysis from a prospective cohort study.

PURPOSE: The objective of this study was to determine whether the magnitude of the peripheral inflammatory response to cardiovascular surgery is associated with increases in blood-brain barrier (BBB) permeability as reflected by changes in cerebrospinal fluid (CSF)/plasma S100B concentrations. METHODS: We conducted a secondary analysis from a prospective cohort study of 35 patients undergoing elective thoracoabdominal aortic aneurysm repair with (n = 17) or without (n = 18) cardiopulmonary bypass (CPB). Plasma and CSF S100B, interleukin-6 (IL-6), and albumin concentrations were measured at baseline (C0) and serially for up to five days. RESULTS: Following CPB, the median [interquartile range] plasma S100B concentration increased from 58 [32-88] pg·mL-1 at C0 to a maximum concentration (Cmax) of 1,131 [655-1,875] pg·mL-1 over a median time (tmax) of 6.3 [5.9-7.0] hr. In the non-CPB group, the median plasma S100B increased to a lesser extent. There was a delayed increase in CSF S100B to a median Cmax of 436 [406-922] pg·mL-1 in the CPB group at a tmax of 23.7 [18.5-40.2] hr. In the non-CPB group, the CSF concentrations were similar at all time points. In the CPB group, we did not detect significant correlations between plasma and CSF S100B with plasma IL-6 [r = 0.52 (95% confidence interval [CI], -0.061 to 0.84)] and CSF IL-6 [r = 0.53 (95% CI, -0.073 to 0.85)] concentrations, respectively. Correlations of plasma or CSF S100B levels with BBB permeability were not significant. CONCLUSION: The lack of parallel increases in plasma and CSF S100B following CPB indicates that S100B may not be a reliable biomarker for BBB disruption after thoracoabdominal aortic aneurysm repair employing CPB. TRIAL REGISTRATION: www.clinicaltrials.gov (NCT00878371); registered 7 April 2009.

RéSUMé: OBJECTIF: L'objectif de cette étude était de déterminer si l'intensité de la réponse inflammatoire périphérique à la chirurgie cardiovasculaire était associée à une augmentation de la perméabilité de la barrière hémato-encéphalique (BHE), telle que reflétée par des changements dans les concentrations de S100B dans le liquide céphalorachidien (LCR) et le plasma. MéTHODE: Nous avons mené une analyse secondaire à partir d'une étude de cohorte prospective portant sur 35 patients bénéficiant d'une réparation élective d'un anévrisme aortique thoraco-abdominal avec (n = 17) ou sans (n = 18) circulation extracorporelle (CEC). Les concentrations plasmatiques et dans le LCR de S100B, d'interleukine-6 (IL-6) et d'albumine ont été mesurées au départ (C0) et en série jusqu'à cinq jours. RéSULTATS: Après la CEC, la concentration médiane [écart interquartile] plasmatique de S100B est passée de 58 [32­88] pg·mL-1 au départ (C0) à une concentration maximale (Cmax) de 1131 [655­1875] pg·mL-1 sur une période médiane (tmax) de 6,3 [5,9­7,0] h. Dans le groupe sans CEC, la concentration plasmatique médiane de S100B a augmenté dans une moindre mesure. Il y a eu une augmentation retardée de S100B dans le LCR à une Cmax médiane de 436 [406­922] pg·mL-1 dans le groupe CEC à une tmax de 23,7 [18,5­40,2] h. Dans le groupe sans CEC, les concentrations dans le LCR étaient similaires à tous les moments. Dans le groupe CEC, nous n'avons pas détecté de corrélations significatives entre la concentration de S100B dans le plasma et le LCR avec les concentrations plasmatiques d'IL-6 [r = 0,52 (intervalle de confiance [IC] à 95 %, -0,061 à 0,84)] et d'IL-6 dans le LCR [r = 0,53 (IC 95 %, -0,073 à 0,85)], respectivement. Les corrélations entre les taux plasmatiques ou dans le LCR de S100B et la perméabilité de la BHE n'étaient pas significatives. CONCLUSION: L'absence d'augmentations parallèles de la concentration de S100B dans le plasma et le LCR après la CEC indique que la S100B pourrait ne pas être un biomarqueur fiable de la perturbation de la BHE après une réparation d'anévrisme aortique thoraco-abdominal sous CEC. ENREGISTREMENT DE L'éTUDE: www.clinicaltrials.gov (NCT00878371); enregistrée le 7 avril 2009.

An observational study examining the effects of a surgically induced inflammatory response on the distribution of morphine and its metabolites into cerebrospinal fluid.

PURPOSE: Morphine is administered intravenously for pain management in the perioperative period. The effect of the inflammatory response to surgery on morphine distribution across the blood-brain barrier (BBB) in humans was investigated. We hypothesized that a graded surgically induced, systemic inflammatory response alters cerebrospinal fluid (CSF) levels of morphine, morphine-3-glucuronide (M3G), and morphine-6-glucuronide (M6G) through a temporary reduction in BBB drug efflux transporter function. METHODS: We conducted a prospective pharmacokinetic study of the plasma and CSF distribution of the P-glycoprotein (PGP) substrate morphine in 33 patients undergoing open thoracic (n = 18) or endovascular (n = 15) aortic aneurysm repair. Morphine was administered with induction of anesthesia and in the intensive care unit. Plasma and CSF concentrations of interleukin (IL)-6, morphine, M3G, M6G, and albumin were measured prior to surgery (baseline), during surgery, and postoperatively every six hours until removal of the CSF drain. The area under the curve (AUC) was determined for plasma and CSF IL-6, morphine, M3G, and M6G concentrations vs time. The primary endpoint measures were the correlations between the morphine, M6G, and M3G AUC CSF/plasma ratios and systemic inflammation as quantified by the time-normalized IL-6 exposure, which was calculated for each individual by dividing the total exposure (AUC) by time (t). A Bonferroni corrected P < 0.017 indicated a significant correlation. RESULTS: Plasma and CSF IL-6 concentrations increased postoperatively. The median [interquartile range] IL-6 exposures were significantly higher in the open vs endovascular surgical group for plasma (105 [40-256] pg·mL-1 vs 29 [16-70] pg·mL-1, respectively; P = 0.013) and CSF (79 [26-133] pg·mL-1 vs 16 [9-80] pg·mL-1, respectively; P = 0.013). For the primary endpoint, the plasma IL-6 AUC/t did not correlate with the CSF accumulation of morphine (r = -0.009; P = 0.96) or M3G (r = 0.37; P = 0.04) when corrected for surgical procedure, age, and sex. There were insufficient data on CSF concentration to complete the primary analysis for M6G. CONCLUSION: Morphine distribution into the CSF was not significantly altered in patients undergoing thoracic aortic aneurysm repair. This suggests that BBB PGP function may not be affected by the perioperative inflammatory response. TRIAL REGISTRATION: www.clinicaltrials.gov , NCT 00878371. Registered 7 April 2009.

Effect of acute inflammatory brain injury on accumulation of morphine and morphine 3- and 6-glucuronide in the human brain.

OBJECTIVE: In animals, central nervous system inflammation increases drug accumulation in the brain partly due to a loss of central nervous system drug efflux transporter function at the blood-brain barrier. To determine whether a similar loss of active drug efflux occurs in humans after acute inflammatory brain injury. DESIGN: Observational human pharmacokinetic study. SETTING: Medical-surgical-neurosurgical intensive care unit at a university-affiliated, Canadian tertiary care center. PATIENTS: Patients with acute inflammatory brain injury, including subarachnoid hemorrhage (n = 10), intracerebral and/or intraventricular hemorrhage (n = 4), or closed head trauma (n = 2) who received morphine intravenously after being fitted with cerebrospinal fluid ventriculostomy and peripheral arterial catheters. INTERVENTIONS: We correlated the cerebrospinal fluid distribution of morphine, morphine-3-glucuronide, and morphine-6-glucuronide with the cerebrospinal fluid and plasma concentration of the proinflammatory cytokine interleukin-6 and the passive marker of blood-brain barrier permeability, albumin. MEASUREMENTS AND MAIN RESULTS: Acute brain injury produced a robust inflammatory response in the central nervous system as reflected by the elevated concentration of interleukin-6 in cerebrospinal fluid. Penetration of morphine metabolites into the central nervous system increased in proportion to the neuroinflammatory response as demonstrated by the positive correlation between cerebrospinal fluid interleukin-6 exposure and the area under the curve cerebrospinal fluid/plasma ratio for morphine-3-glucuronide (r = .49, p < .001) and morphine-6-glucuronide (r = .51, p < .001). In contrast, distribution of morphine into the brain was not linked with cerebrospinal fluid interleukin-6 exposure (r = .073, p = .54). Albumin concentrations in plasma and cerebrospinal fluid were consistently in the normal range, indicating that the physical integrity of the blood-brain barrier was likely undisturbed. CONCLUSIONS: Our results suggest that central nervous system inflammation following acute brain injury may selectively inhibit the activity of specific drug efflux transporters within the blood-brain barrier. This finding may have significant implications for patients with neuroinflammatory conditions when administered centrally acting drugs normally excluded from the brain by such transporters.