Clinical impact of chronic substance abuse in a Norwegian ICU-population.

BACKGROUND: The clinical impact of chronic substance abuse of alcohol and drugs-referred to as substance use disorders (SUD)-is often overlooked in the intensive care (ICU) setting. The aims of the present study were to identify patients with SUD-regardless of cause of admission-in a mixed Norwegian ICU-population, and to compare patients with and without SUD with regard to clinical characteristics and mortality. METHODS: Cross-sectional prospective study of a mixed medical and surgical ICU-population aged ≥18 years in Oslo, Norway. Data were collected consecutively, using a questionnaire including the AUDIT-C test, medical records and toxicology results. Patients classified with SUD were divided into the subgroups alcohol use disorders (AUD) and drug use disorders (DUD). RESULTS: Overall, 222 (26%) of the 861 patients included were classified with SUD; 137 (16%) with AUD and 85 (10%) with DUD. 130/222 (59%) of the SUD-patients had substance abuse-related cause of ICU-admission. Compared to non-SUD patients, DUD-patients were younger (median age 42 vs 65 years) and had lower SAPS II scores (41 vs 46), while AUD-patients had higher SOFA scores (8.0 vs 7.3). Overall, age-adjusted logistic regression analysis showed similar hospital mortality for SUD-patients and non-SUD patients, but AUD was associated with increased mortality among medical patients and in patients with sepsis (OR 1.7 (95% CI 1.0-2.8), and OR 2.6 (95% CI 1.1-6.2)). CONCLUSION: One in four ICU-patients had SUD regardless of cause of admission. Alcohol use disorder was associated with increased mortality in medical patients and in patients with sepsis.

Genetic variation in neutrophil accumulation in mice is not mediated through immigrant regulatory cells.

Genetic variation of induced peritoneal neutrophilia in mice was accompanied by parallel variation in macrophage responses. The timing of the macrophage responses in high responder (C57B1/10) mice indicated a potential role for these cells in mediating the enhanced neutrophil response. However, adoptive transfer of inflammatory macrophages did not induce neutrophilia. Analysis of peritoneal cytokine levels in high and low responder mice further indicated that IL-1, IL-3, GM-CSF, G-CSF and interferon-gamma (IFN-gamma) were not involved in mediating the genetic variation observed. Exogenous tumour necrosis factor-alpha (TNF-alpha) was effective in inducing the high responder phenotype, despite the absence of detectable TNF-alpha in either peritoneal fluid or serum. A role for genetically determined differential expression of endothelial adhesion molecules in high and low responders is suggested.

Effect of genetic variation on induced neutrophilia in mice.

Mice from a variety of strains were injected with a sterile irritant (Brewer's thioglycolate) and killed bacteria (Staphylococcus aureus, Staphylococcus epidermidis, or Escherichia coli) to determine their effect on accumulation of neutrophils in the peritoneal cavity. Peak accumulation occurred around 15 h postinjection and showed significant strain-related variation. C57BL/10 mice were identified as having a high-responder phenotype and BALB/c mice a low-responder phenotype. Inheritance of the high-responder phenotype followed simple Mendelian genetics: (BALB/c x C57BL/10)F1 mice were found to be more responsive than either parental phenotype. Major histocompatibility complex H-2d haplotype was found to convey an augmented neutrophil response in conjunction with B10 background high-responder genes (B10.D2/n) but the H-2d haplotype per se was not the only factor in determining high responsiveness. Gram-positive and gram-negative bacteria appeared to activate different immune mechanisms. Both gram-negative bacteria and lipopolysaccharides (LPS) induced a response similar to, but less potent than, that induced by Brewer's thioglycollate. Neutralization of the LPS content of Brewer's thioglycolate abrogated the response.