By Richard R. Watkins, MD, MS, FACP, FIDSA, FISAC
Professor of Internal Medicine, Northeast Ohio Medical University, Rootstown, OH
SYNOPSIS: A multicenter, retrospective cohort study on patients with gram-negative bacteremia found infectious diseases consultation was associated with a 40% decrease in 30-day mortality.
SOURCE: Shulder S, Tamma PD, Fiawoo S, et al. Infectious diseases consultation associated with reduced mortality in gram-negative bacteremia. Clin Infect Dis 2023;77:1234-1237.
Despite the availability of effective antibiotics, gram-negative bacteremia (GNB) remains a major cause of severe illness and death. Infectious diseases consultation (IDC) has been shown to improve outcomes for bloodstream infections due to Staphylococcus aureus and Candida spp. Shulder and colleagues sought to determine whether IDC could similarly improve 30-day mortality and other outcomes in patients with GNB.
The study was a retrospective, observational, cohort analysis that included 24 hospitals in the United States. All sites had antibiotic stewardship programs (ASP), and none had a mandatory IDC for GNB policy. Adult patients admitted from January 2019 to December 2019 with GNB were included, while patients who died < 48 hours from admission were excluded. A sensitivity analysis was performed by excluding hospitals with standard ASP review of GNB to account for confounding by ASP. Furthermore, a propensity score analysis was conducted to balance the IDC and no-IDC groups.
The primary exposure was IDC vs. no IDC in patients with GNB. The primary outcome was 30-day mortality due to any cause. Secondary outcomes included 30-day readmission due to the index infection, 30-day bacteremia recurrence due to the original bacterial species, and 90-day mortality.
IDC was obtained in 58% (2,814/4,861) of cases of GNB. There were 155 episodes where death occurred < 48 hours from admission, which were excluded from the analysis. IDC happened an average of 1.8 days (standard deviation, 2.2 days) from initial blood culture collection. There were no significant differences (P < 0.05) in any patient characteristic between the two groups. The most common source of GNB in both groups was the urinary tract, followed by intraabdominal and hepatobiliary etiologies. Source control was achieved in 2,055/2,814 (85%) of patients with IDC and 1,491/2,047 (84%) of patients without IDC.
Of the 4,861 cases of GNB, 681 (14%) died within 30 days of diagnosis. Death occurred in 12% of patients with IDC vs. 17% without IDC, a 40% reduction (hazard ratio [HR], 0.6; 95% confidence interval [CI], 0.47-0.77). Mortality also was significantly reduced at 90 days with IDC (HR, 0.7; 95% CI, 0.57-0.86). There was no difference between the two groups in the 30-day readmission rate from recurrent bacteremia, including the original bacterial species. Finally, sites that had routine ASP interventions for GNB had similar findings to those without them.
COMMENTARY
In the largest study to investigate the impact of IDC on GNB, Shulder and colleagues found a 40% reduction in mortality. This is a valuable discovery that demonstrates the substantial effect of IDC on GNB. The authors should be commended for their work, which highlights the benefits of IDC. While the optimization of antibiotic therapy is the most crucial role for the IDC in patients with GNB, there also are other important management decisions. Using sagacious diagnostic skills, infectious diseases (ID) physicians carefully and systematically assess for complications of GNB, such as abscess formation and seeding of prosthetic material. They adjust antibiotics based on cultures and other laboratory data.
ID physicians also determine when it is appropriate to switch to oral therapy. Many non-ID physicians continue empiric broad-spectrum antibiotic therapy due to clinical inertia, which can lead to increased costs and adverse events. In the current study, there was an increased likelihood of IDC if the patient had multiple comorbidities, a severe presentation, or if a complicated pathogen (e.g., Pseudomonas, Acinetobacter, Stenotrophomonas, and multi-drug-resistant organism [MDRO]) was present.
There are some limitations to the study. First, most of the authors are ID physicians, and it was published in an ID journal, making confirmation bias a concern. Second, the retrospective design means the results could have been affected by unmeasured confounding variables. This could explain why previous studies found a reduction in 30-day readmissions and recurrent GNB with IDC. Third, the investigators did not assess whether timing of IDC affected any of the outcomes. Fourth, the rate of MDROs was low. Finally, recurrent GNB is rare in clinical practice, which makes detecting a significant difference a challenge.
ID physicians have an important role to play in the management of GNB. Effort should be made to increase awareness of the value of IDC for GNB within the wider healthcare environment. Further investigation on whether the timing of IDC affects outcomes is warranted.
