Journal Club

In this feature, a panel of IDSA members identifies and critiques important new studies in the current literature that have a significant impact on the practice of infectious diseases medicine.

• Saving SIRS? Discernment of Sepsis from Non-Infectious Syndromes in the ED
• Current Epidemiology of Bacterial Infections in Young Infants
• Cost and Length of Stay Savings with Molecular Testing for Tuberculosis among U.S. Inpatients

Click here for the previous edition of Journal Club. For a review of other recent research in the infectious diseases literature, see “In the Literature,”  by Stanley Deresinski, MD, FIDSA, in each issue of Clinical Infectious Diseases.

Saving SIRS? Discernment of Sepsis from Non-Infectious Syndromes in the ED

Reviewed by Kelly Cawcutt, MD

Upon patients’ arrival to the emergency department (ED), determining whether they do or do not have sepsis is difficult. Because of this, many patients receiving antibiotics ultimately are found to have a noninfectious cause of their syndrome. In order to try and improve detection of sepsis in the ED, Mearelli et al. completed a multicenter prospective study at five university hospitals in Italy and reported their results in a recent article in Critical Care Medicine.

The study included two cohorts: an inception cohort for the development of a predicted algorithm and a validation cohort to determine positive and negative predictive value of the algorithm based on pretest probability of infection with the aim to differentiate sepsis syndromes from noninfectious inflammatory states. Despite its faults, criteria for systemic inflammatory response syndrome (SIRS) were used to enroll patients and then categorize them based on whether an infectious diagnosis was ultimately identified. Biomarkers utilized in the algorithm included procalcitonin, soluble phospholipase A₂ group IIA (sPLA_2G^IIA), presepsin, soluble interleukin-2 receptor α (sCD24), and soluble triggering receptor expressed on myeloid cells (sTREM-1).

The area under the curve (AUC) for the algorithm in the validation cohort was an impressive 0.95 (95 percent confidence interval [CI] 0.82-0.90). The primary impact of this study may be for ruling out sepsis and septic shock given that only five of 700 (0.7 percent) patients were misclassified when having sepsis or septic shock compared to the usual approach of clinical concern for infection, fever, leukocytosis, and increased C-reactive protein.

Although imperfect, the algorithm does appear to be better than current diagnostics used for determining if a patient has sepsis or septic shock. Primary concerns for future implementation of this are the few patients that had sepsis or septic shock with a negative SIRS screen, the feasibility of rapid multimodal test results for screening versus potential delay of therapy, and the overall cost of obtaining such labs.

(Mearelli et al. Crit Care Med. 2018;46(9):1421-1429.)

Back to Top

Current Epidemiology of Bacterial Infections in Young Infants

Reviewed by Terri Stillwell, MD

Concern for possible bacterial infection is one of the most common reasons young infants are evaluated in local emergency departments. Oftentimes, these patients are treated empirically, pending infectious work-up. In a recent study, published in The Journal of Pediatrics, the authors describe the epidemiology of invasive bacterial infections, providing helpful information for clinicians. In a multi-center, retrospective study, pathogens from infants less than 60 days old who were diagnosed with bacteremia and/or bacterial meningitis were evaluated. 

Nearly 20,900 blood cultures and 10,650 cerebrospinal fluid (CSF) cultures were assessed; 2.0 percent of the blood cultures were positive for a true pathogen, with 0.7 percent of the CSF cultures being positive. Four hundred forty-two infants were classified as having an invasive bacterial infection: 79.9 percent with bacteremia, 14.5 percent with bacterial meningitis with bacteremia, and 5.7 percent with bacterial meningitis without bacteremia. Interestingly, 29.4 percent of patients also had urinary tract infections in addition to one of the other invasive bacterial infections. 

The peak incidence of infection occurred during the second week of life. For infants less than or equal to 28 days of age, E. coli was found to be the most frequent cause of bacteremia, whereas Group B Streptococcus (GBS) caused more bacterial meningitis. For those 29-60 days of age, GBS was the leading cause of all invasive infection. Staphylococcus aureus was the third most common pathogen found.

With regards to antimicrobial susceptibilities, nearly 96 percent of all pathogens were sensitive to a combination of ampicillin plus gentamicin or a third-generation cephalosporin. However, approximately 11 percent of the pathogens were resistant to a third-generation cephalosporin alone. 

The empiric treatment of young infants in whom there is concern for invasive bacterial infection remains a daily challenge for many pediatric physicians. The results of this study may help aid in their clinical decision making.

(Woll et al. J Pediatr. 2018;200:210-217.e1.)

Back to Top

Cost and Length of Stay Savings with Molecular Testing for Tuberculosis among U.S. Inpatients

Reviewed by Michael T. Melia, MD

Multiple studies support molecular testing (GeneXpert [Xpert] MTB/RIF; Cepheid) strategies for tuberculosis (TB). A recent analysis in JAMA Internal Medicine investigates the impact of molecular testing on the duration of respiratory isolation for patients with possible TB.

The investigators conducted a prospective, cohort, before-and-after implementation study of the molecular testing strategy introduced at Zuckerberg San Francisco General Hospital in January 2015. Six hundred twenty-one consecutive inpatients undergoing evaluation for pulmonary TB were included; patients with positive rapid TB test results were excluded. In the pre-implementation period, 233/301 (77 percent) patients completed the rapid TB testing evaluation, as compared with 259/320 (81 percent) in the post-implementation period. Pre-implementation, all possible TB patients were required to stay in isolation until at least two sputa over two separate days tested smear negative. Post-implementation, completed testing was defined based upon clinical probability of TB via an institutional algorithm developed through multidisciplinary input. Time spent in isolation was calculated from the time between orders for its initiation and discontinuation. Serial sputum mycobacterial cultures were the gold standard.

Eight (3.4 percent) and seven (2.7 percent) patients had positive TB cultures pre- and post-implementation, respectively. One culture-positive patient initially tested smear positive but molecular assay negative; the patient subsequently had a positive assay as per the algorithm. One culture-positive patient tested assay positive but smear negative. Median hospital length of stay decreased from 6.0 days (interquartile range [IQR], 3.8-10.9) before implementation to 4.9 days (IQR, 2.9-8.9) after (P = 0.003). Median duration of respiratory isolation decreased from 2.9 days (IQR, 2.0-3.7) to 2.5 days (IQR, 1.7-3.4) (P = 0.001). Mean hospital costs per patient with negative rapid test results decreased from $46,921 to $33,574 (average savings of $13,347 per patient).

This study shows that molecular testing strategies for TB are safe, effective, and cost-saving, even with a fairly conservative algorithm for discontinuing airborne precautions. Widespread adoption of molecular testing strategies has the potential to benefit patients, providers, and health care systems alike.

(Chaisson et al. JAMA Intern Med. Published online: August 27, 2018.)

Back to Top