The ILC-FoCUS focused echocardiography recommendations are a product of 4 conferences, 33 experts and a Delphi process, giving a useful set of statements on everything from training to equipment to volume status.
July 2014 archive
Early tracheostomy? Definitely maybe.
Tracman said no, but others have shown benefit. Siempos’ meta-analysis of 13 trials suggests thinking again. Depending on which statistical model you use, it appears that early tracheostomy may reduce 30 day mortality, particularly, or exclusively in patients with high baseline risk of death. Less pneumonia and, unsurprisingly, shorter ICU stay is also suggested. However longer …
Surely statins are useful for something on the ICU?
Recent articles on statins for ARDS treatment, VAP treatment and COPD excerbation prevention have all been disappointing. In the lab they are anti-inflammatory, protect coagulation pathways, and inhibit micro-organisms. On the unit there have been suggestions that they prevent the onset of, and reduce mortality in, severe infections and sepsis, but this has been difficult to replicate in …
Be balanced when resuscitating sepsis?
Raghunathan has used some interesting statistics to try and extract the effects of using Saline or Ringer’s lactate in resuscitating sepsis. Elixhauser comborbidities, 5:1 greedy matching, Generalized Estimating Equation models, and “missing-ness” are just a few of the terms I didn’t understand. Nevertheless this appears to be a retrospective study that suggests there’s measurable mortality benefit in using balanced …
Prone who, when and for how long? Are we clear now?
Proning in ICU is 40 years old now. Meta-analyse 11 high-quality studies in a thorough manner (including blinding to authors!) and it appears strikingly good, with an NNT of 11 – definitely better than protective ventilation alone. Do it in the severely hypoxic (P/F <100 mmHg), do it early (within 72 hours), do it for …
- 1
- 2
Which co-morbidities really matter (at least in Scotland!)?
July 30, 2014
For over 1000 Scottish teaching hospital admissions, using univariate logistic regression: IHD, ALD, COPD, obesity, rheumatological disease, functional status and, just barely, age all impact upon unit or hospital death. Surprisingly many conditions such as DM, malignancy, tromboembolic disease, CVA and several others did not.