Comparison of the T2Dx instrument with T2Candida assay and automated blood culture in the detection of Candida species using seeded blood samples

Using the number needed to treat to assess appropriate antimicrobial therapy as a determinant of outcome in severe sepsis and septic shock

OBJECTIVE

To assess appropriate antimicrobial therapy as an outcome determinant in severe sepsis and septic shock using the number needed to treat.

DESIGN

Single-center cohort study (January 2008 to December 2012).

SETTING

One thousand two hundred fifty-bed academic hospital.

PATIENTS

Two thousand five hundred ninety-four patients with positive blood culture.

INTERVENTIONS

We retrospectively identified patients with severe sepsis or septic shock. Inappropriate antimicrobial treatment was defined as an antimicrobial regimen that lacked in vitro activity against the isolated pathogen. Information regarding demographics, severity of illness, comorbidities, microbiology, and antimicrobial treatment was recorded. Logistic regression was used to identify risk factors for hospital mortality and inappropriate treatment.

MEASUREMENTS AND MAIN RESULTS

Seven hundred eighty-seven patients (30.3%) were nonsurvivors. Inappropriate antimicrobial treatment had the greatest adjusted odds ratio for hospital mortality (adjusted odds ratio, 3.4; 95% CI, 2.8-4.1; p < 0.001). Multivariate logistic regression analysis identified resistance to cefepime, resistance to meropenem, presence of multidrug resistance, nonabdominal surgery, and prior antibiotic use as being independently associated with the administration of inappropriate antimicrobial treatment. For the entire cohort, the number needed to treat with appropriate antimicrobial therapy to prevent one patient death was 4.0 (95% CI, 3.7-4.3). The prevalence-adjusted pathogen-specific number needed to treat (PNNT) with appropriate antimicrobial therapy to prevent one patient death was lowest for multidrug-resistant bacteria (PNNT = 20) followed by Candida species (PNNT = 34), methicillin-resistant Staphylococcus aureus (PNNT = 38), Pseudomonas aeruginosa (PNNT = 38), Escherichia coli (PNNT = 40), and methicillin-susceptible S. aureus (PNNT = 47).

CONCLUSIONS

Our results support the importance of appropriate antimicrobial treatment as a determinant of outcome in patients with severe sepsis and septic shock. Our analyses suggest that improved targeting of empiric antimicrobials for multidrug-resistant bacteria, Candida species, methicillin-resistant S. aureus, and P. aeruginosa would have the greatest impact in reducing mortality from inappropriate antimicrobial treatment in patients with severe sepsis and septic shock.

Effects of empiric antifungal therapy for septic shock on time to appropriate therapy for Candida infection: a pilot study

PURPOSE

Inappropriate initial therapy for Candida-related septic shock is common and associated with a high mortality rate. This before-after pilot study was conducted to determine the feasibility of using empiric therapy for reducing the time to appropriate antifungal therapy in patients with Candida-related septic shock.

METHODS

Patients aged 18-99 years with septic shock presenting to Barnes-Jewish Hospital, St. Louis, Missouri, in 2012-2013 were assigned to 1 of 2 groups. Patients presenting between January 1, 2012, and December 31, 2012, were managed according to local standard of care for patients with septic shock, to include antifungal therapy at the discretion of the treating physician (standard therapy group). Patients presenting between January 1, 2013, and December 31, 2013, received empiric antifungal therapy (primarily micafungin 100 mg/d or fluconazole 800 mg on day 1, followed by 400 mg/d), facilitated by a clinical pharmacist in the medical intensive care unit, until microbiologic cultures were available to determine the cause of septic shock (empiric therapy group). The primary outcome was time to appropriate therapy after shock onset.

FINDINGS

A total of 28 patients were enrolled (mean age, 56.3 [15.1] years [range, 30-92 years]; 16 [57.1%] men). The time to appropriate therapy after shock onset was statistically shorter with empiric therapy (n = 13) compared with standard therapy (n = 15) (10.6 [15.8] vs 40.5 [26.0] hours; P = 0.001). Patients receiving empiric therapy were more likely to have received appropriate therapy within 12 hours (69.2% vs 6.7%; P = 0.001) and within 24 hours (76.9% vs 40.0%; P = NS) of shock onset. In an analysis to determine the number of septic shock patients needed to be treated with empiric antifungal therapy for 1 patient with Candida-related septic shock to receive appropriate treatment, 256 patients without Candida infection received a total of 687 doses of empiric antifungal therapy (mean, 2.7 doses per patient) compared with 136 patients who received 382 doses of standard antifungal therapy (mean, 2.8 doses per patient); the number needed to treat was 19.6.

IMPLICATIONS

The present pilot study demonstrated that the use of empiric antifungal therapy for Candida-related septic shock was associated with a statistically shorter time to administration of appropriate treatment. ClinicalTrials.gov identifier.

Clinical Practice Patterns in Hospitalized Patients at Risk for Invasive Candidiasis

Background: Rapid diagnostic tests for Candida are becoming available that may supplement traditional microbiological identification. Objective: Assess clinical practice patterns in patients with or at risk of candidiasis who may benefit from the use of rapid diagnostic tests. Methods: This was a prospective cohort study of patients with candidemia or receiving systemic antifungals conducted at a university-affiliated tertiary care hospital. Time to initiation of therapy, Candida species, time to identification, and indications for antifungal use were assessed. Results: A total of 162 patients with candidemia aged 58 ± 17 years were identified. Average time to yeast identification yeast was 2.2 ± 1.3 days and varied by Candida species (range = 0.6-7.9 days). Average time for patient to start antifungal therapy was 3.5 ± 2.1 days. In Monte Carlo simulations, average time to initiation of antifungal therapy was 0.6 ± 0.2 days for T2Candida, 2.6 ± 1.3 days for PNA-FISH (fluorescence in situ hybridization using peptide nucleic acid probes), and 2.5 ± 1.4 days for MALDI-TOF (matrix-assisted laser desorption/ionization time of flight). Use of T2Candida on the day of the blood culture collection resulted in 3136 to 6078 fewer doses of echinocandins annually per 5000 patients. Conclusion: Many interventions are possible for antifungal stewardship programs to improve care of patients at risk for systemic candidiasis, including rapid identification of yeast species and limiting unnecessary antifungal agents. Technology enabling rapid diagnosis of Candida will be paramount to appropriate, cost-effective treatment of patients with or at risk for candidiasis.