In this small single institution study of 159 bacteremic patients, all of which had troponin I sent, 43% were elevated (abstract). Most were minor elevations (96% < 1mcg/L). Not surprisingly, independent risk factors for troponin elevations included renal insufficiency, higher WBC, and septic shock. Troponin elevations were not significantly associated with EKG changes (although not performed in all patients), and did not independently predict mortality. For bacteremic patients, if you check troponin, there is about a coin-toss chance it will be elevated, but given the lack of association with EKG changes or mortality, there is no clear benefit of routinely checking it in the first place.

Current recommendations from the ADA (guideline) and Surviving Sepsis Campaign (guideline) advocate for tight glucose control in ICU patients, although this had been a contentious issue. The landmark RCT of surgical ICU patients found signficantly lower mortality, LOS, and organ dysfunction in tight control (BS goal 80-110 mg/dL) (van den berghe) but a similar study in medical ICU patients did not find a mortality difference overall. It did find higher mortality in those with LOS <3 days and lower mortality in those with LOS > 3days. They also found a 3X higher rate of hypoglycemia in the intensive control group (van den berghe). A recently published meta-analysis of 29 controlled trials from 8432 ICU patients did not find a mortality benefit in tight control versus usual care (abstract). Tight control was associated with lower risk of septicemia (RR 0.76, CI 0.59-0.97) but markedly increased risk of hypoglycemia (BS <40 mg/dL, RR 5.13, CI 4.09-6.43). The NICE-SUGAR trial will randomize >5000 ICU patients to tight (80-110) versus conventioal (140-180) control and will hopefully settle this issue (NIH site). For non-ICU patients, there is no evidence that tight control is superior, and for now we should follow the ADA guidelines of fasting targets <126 and non-fasting targets <180 for hospitalized patients (ADA website)

Here are some interesting tidbits recently published from the largest prospective cohort of infective endocarditis ever collected (abstract). Causative organisms were Gram positive in 81-88% of cases (3-4% Gram negative, 1-2% fungi or yeast, and 8-13% other / culture negative). MSSA accounted for almost 1/3 of cases, and MRSA accounted for 36% of cases in those >65 years old, and 21% of cases in those 18-65 years old. Compared to patients 18-65 years old, those >65 years old were more likely to have a +blood culture (92% vs 86%) but less likely to have vegetations (84% vs 88%) peripheral embolic events (15% vs 26%) or stroke (15% vs 18%). However, they were twice a likely to die (25% vs 13%). Take home messages here are to remember atypical presentations (lack of blood culture or vegetations; hence the need for the modified Duke criteria for diagnosis (Duke criteria)), and the overwhelmingly high rates of MRSA and death in the elderly.

Sporadic influenza activity is being reported in a handful of states. Follow the CDC map to keep tabs on your local activity (cdc map)

The CDC has confirmed a case of transfusion related anaplasmma phagocytophilum (formerly known as human granulocytic erlichiosis), which is a tickborne rickettsial disease (report). Although a rare event, it is important to recognize that the US blood supply is not routinely screened for tickborne diseases, and that these organisms can survive in refridgerated RBCs. We therefore must consider rickettsial diseases in all patients with post-tranfusion fever and thrombocytopenia

In secondary (non-afib) stroke prevention, current guidelines recommend antiplatelet therapy (guideline) but do not give a prefence to which type. In the first head to head trial, the RCT PRoFESS Trial (abstract), randomized 20,332 patients to clopidogrel or ASA-dipyridamole within 4 months of an ischemic stroke. There was no significant difference in recurrent stroke, MI, or death between the groups. This study indicates there is not much difference in efficacy between them, and an accompanying editorial satirically states, in haiku ”For stroke prevention / use an antiplatelet drug. / Treat hypertension.” (editorial).

Current guidelines recommended providing lytics in acute stroke within 3 hours of symptom onset (guidelines). However, 2 new studies provide evidence for extending that window to 4.5 hours. The first observational study found similar 30 day mortality, 30 day functional status, and 24 hour ICH in patients who recieved lytics in <3 hours compared to those in the 3-4.5 hour window (abstract). The second was a RCT of patients presenting between 3-4.5 hours with acute stroke. The lytic group (compared to placebo) had a slightly higher %  favorable 90 day functional score and non-significantly fewer deaths, but significantly more ICH (OR 1.73) (abstract) As an accompanying editorialist notes, “how long do you have to initiate thrombolytic therapy? The correct answer is 1 minute”, indicating that any delay, however short, is too long (editorial). However, in patients presenting between 3-4.5 hours with acute stroke, evidence suggests they should still be evaluated for lytics, as some of them may benefit.

This study sought to answer the question of how much oxyen your patients are inspiring with different devices, with and without tachypea. They simulated tachypnea in healthy volunteers (by chest binding) and measured the Fi02 in the hypopharynx before and after tachypnea (RR>25). They found the Fi02 decreased by 12-24% in rebreathing masks after tachypnea, but did not change in non-rebreathing masks after tachypnea. However, they did find the hypopharynx Fi02 was only 68% in non-rebreathing masks (indicating the term “100% non-rebreather” is a misnomer). It is very important to know the amount of Fi02 your patients are actually inspiring, and to note that it significantly decreases with tachypnea in rebreather devices (abstract).

For patients with chronic renal insufficiency (gfr<60) undergoing coronary angiography, previous small RCTs showed bicarbonate to be superior to saline (abstract) (abstract) (abstract) in preventing CIN, but in a subsequent larger RCT, it was not found to be superior (when combined with N-acetylcysteine) (abstract). A recent meta-analysis found bicarbonate to be superior to saline, but noted signficant study heterogeneity and likely publication bias, calling for a larger RCT (meta-analysis). Now we have a larger RCT, in which 353 high risk patients (gfr <60 with DM, CHF, HTN, or age >75) were randomized to saline or bicarbonate infusions, and N-acetylcysteine was given at the discretion of the treating physician (in about half of patients). There was no difference in incidence of CIN between the groups (even when stratified by DM or receipt of N-acetylcysteine). In summary, it is unlikely that bicarbonate infusions are more beneficial than saline in preventing CIN (abstract)

Oral rivaroxaban has been shown to be safe and effective in VTE prophylaxis in orthopedic patients (abstract) (abstract). This phase II study gives us enthusiasm for its safety and efficacy in DVT treatment as well (abstract). 543 participants were randomized to oral rivaroxaban (20, 30, or 40mg) or standard therapy (LMWH bridge to coumadin). AT 3 months, non-significantly fewer rivaroxaban patients met the primary endpoint (recurrent DVT, PE, or increase in thrombotic burden) (5-7% vs 10% in standard therapy) and non-significantly fewer rivaroxaban patients had bleeding complications (2-6% vs 9% in  standard therapy). Given its apparent equivalency to standard therapy, and that it requires no monitoring, it will likely supplant coumadin therapy in DVT treatment (assuming confirmation in phase III studies).