It has been suggested that excessive activation of the anti-inflammatory pathways in sepsis may lead to poor outcome of patients with sepsis. The aim of this study was to test the value of histocompatibility leukocyte antigen (HLA)-DR-expression on blood monocytes and plasma levels of interleukin (IL)-4 and -10 in prediction of hospital mortality in patients with sepsis. Sixty-one critically ill patients with sepsis were prospectively enrolled to this study in two university hospital intensive care units. Survivors (n = 41) and nonsurvivors (n = 20) differed significantly in HLA-DR expression at admission: survivors' median 84% (interquartile range 64%-98%) versus nonsurvivors' median 62% (interquartile range 47%-83%, P = 0.025 by Mann-Whitney test). Similarly, the analysis revealed statistically significant differences between survivors and nonsurvivors in admission plasma IL-10 levels and in admission Sequential Organ Failure Assessment (SOFA) and Acute Physiology and Chronic Health Evaluation (APACHE) II scores, but not in IL-4 levels. The areas under receiver operating curves (AUC) showed that both monocyte HLA-DR expression and plasma IL-4 level showed poor discriminative power in prediction of hospital mortality (AUC < 0.70). Only IL-10 levels on days 1 and 2 showed reasonable predictive power (AUCs 0.706 and 0.725, respectively). The highest AUC values were those of APACHE-II (0.786) and admission SOFA score (0.763). In conclusion, APACHE II and SOFA scores on admission showed better discriminatory power than HLA-DR expression and IL-10 and IL-4 levels in prediction of hospital mortality in critically ill patients with sepsis.

Sixty patients scheduled for colonic surgery were randomly allocated to four groups according to postoperative pain medication: I. Control group, the patients received oxycodone intramuscularly (0.15 mg kg-1) on request. II. Epidural bupivacaine (0.25%) continuously administered by infusion pump, 4-6 ml h-1, for 48 h. III. Epidural morphine, 2-6 mg, at the end of operation and repeated on the first and second postoperative mornings. IV. Epidural morphine, 2-6 mg per die, administered for 48 h continuously by infusion pump. All patients received a balanced anaesthesia with enflurane, fentanyl and vecuronium. Postoperatively, intramuscular oxycodone was given on request. There were no significant differences between the groups in changes in peak flow, spirometry and blood-gas analyses postoperatively. Pain intensity (visual analogue scale) was lower in Groups II and III at 3 h and in Group IV at 24 h compared to the control Group I. All the epidurally treated groups needed less additional analgesics than the control Group I. Postoperatively bowel movements occurred on the second day in Group II (bupivacaine) as compared to the fourth day in all other groups (P less than 0.05).

SummaryThe

Sixty-three patients (ASA 1-2), scheduled for elective surgery under general anaesthesia, were randomly given either oral clonidine (225-375 micrograms) + diazepam (5 15 mg), cimetidine (300 mg the night before and 300 mg in the morning) + diazepam or only diazepam for premedication. Anaesthesia was induced with thiopentone and maintained with N2O + O2 (70:30), enflurane and fentanyl. Vecuronium bromide was used as a muscle relaxant. The sleep dose of thiopentone was significantly smaller in the patients pretreated with clonidine than in the other groups. The mean maximal increase in heart rate was lowest in the clonidine-pretreated patients, but there were no significant differences in the mean arterial pressure changes associated with intubation. Before and just after intubation and in the recovery room, the arterial pressures were lowest in the patients pretreated with clonidine. During anaesthesia, marked bradycardia (less than or equal to 45 beats min-1) did not occur more often when clonidine was used, but in the recovery room there were statistically significantly more patients with bradycardia in the clonidine group than in the other groups. On the electrocardiogram (ECG) during the endotracheal intubation, the incidence of bigeminy was higher in the diazepam patients (5/20) than in the cimetidine patients (2/20) and the clonidine patients (0/23). There were significantly more gastric content samples with a pH above 2.5 in the cimetidine group than in the other groups, and clonidine patients did not differ from diazepam patients in this respect. The high incidence of bradycardia with the concomitant hypotension may limit use of this drug to highly selected patients.