Epidemiology and treatment of invasive Bartonella spp. infections in the United States.

OBJECTIVES: Bartonella spp., renowned for cat-scratch disease, has limited reports of dissemination. Tissue and blood cultures have limitations in detecting this fastidious pathogen. Molecular testing (polymerase chain reaction, PCR) and cell-free DNA have provided an avenue for diagnoses. This retrospective observational multicenter study describes the incidence of disseminated Bartonella spp. and treatment-related outcomes. METHODS: Inclusion criteria were diagnosis of bartonellosis via diagnosis code, serology testing of blood, polymerase chain reaction (PCR) of blood, 16/18S tests of blood or tissue, cultures of blood or tissue, or cell-free DNA of blood or tissue from January 1, 2014, through September 1, 2021. Exclusions were patients who did not receive treatment, insufficient data on treatment course, absence of dissemination, or retinitis as dissemination. RESULTS: Patients were primarily male (n = 25, 61.0%), white (n = 28, 68.3%), with mean age of 50 years (SD 14.4), and mean Charlson comorbidity index of 3.5 (SD 2.1). Diagnosis was primarily by serology (n = 34, 82.9%), with Bartonella henselae (n = 40, 97.6%) as the causative pathogen. Treatment was principally doxycycline with rifampin (n = 17, 41.5%). Treatment failure occurred in 16 (39.0%) patients, due to escalation of therapy during treatment (n = 5, 31.3%) or discontinuation of therapy due to an adverse event or tolerability (n = 5, 31.3%). CONCLUSIONS: In conclusion, this is the largest United States-based cohort of disseminated Bartonella spp. infections to date with a reported 39% treatment failure. This adds to literature supporting obtaining multiple diagnostic tests when Bartonella is suspected and describes treatment options.

Variability in Antimicrobial Prescribing Across Five Ambulatory Settings Within a Large Integrated Health System.

Background: Lack of access to timely, detailed antibiotic use data has limited ambulatory antibiotic stewardship efforts. Antibiotic utilization is tracked across ambulatory care sites and emergency departments (ED) within a large integrated health system. Methods: This is a retrospective cohort analysis from June 1, 2019 to May 31, 2020 comparing antibiotic prescribing for all patients with ICD-10 diagnosis codes for cystitis, otitis media, pharyngitis, sinusitis, and upper respiratory tract infections (URTIs) among five ambulatory care departments across northeast Ohio and southeast Florida locations: ED, Urgent Care (UC), On-Demand Telehealth (TEL), Pediatrics (PED), and Primary Care (PC). Results: A total of 261,947 encounters were included (ED:56,766, UC:92,749, TEL:8,783, PED:29,151, PC:74,498) for the treatment of cystitis (30,932), otitis media (22,094), pharyngitis (59,964), sinusitis (53,693), or URTI (95,264). The population was 63% female with a median age of 34.2 years [12.8-56.3]. A total of 17% of patients had documented penicillin allergies and 18% of patients with pharyngitis received Group A Streptococcus (GAS) testing. Antibiotics were prescribed in 44% of encounters (ED:21,746 [38%], UC:45,652 [49%], TEL:4,622 [53%], PED:10,909 [37%], PC:33,547 [45%]; P < 0.001). Guideline concordant antibiotics were prescribed in 65% of encounters (ED:14,338 [66%], UC:31,532 [69%], TEL:3,869 [84%], PED:8,212 [75%], PC:17,263 [51%]; P < 0.001). Conclusions: Observed rates of antibiotic and guideline concordant antibiotic prescribing were similar to national published rates of antibiotic prescribing in the ambulatory setting. The variability in antibiotic prescribing demonstrates opportunities for targeted outpatient stewardship efforts. Timely antibiotic tracking tools can facilitate ambulatory antimicrobial stewardship activities.

Clinical impact of a multiplex rapid diagnostic pneumonia panel in critically ill patients.

Objective: To evaluate the clinical impact of the BioFire FilmArray Pneumonia Panel (PNA panel) in critically ill patients. Design: Single-center, preintervention and postintervention retrospective cohort study. Setting: Tertiary-care academic medical center. Patients: Adult ICU patients. Methods: Patients with quantitative bacterial cultures obtained by bronchoalveolar lavage or tracheal aspirate either before (January-March 2021, preintervention period) or after (January-March 2022, postintervention period) implementation of the PNA panel were randomly screened until 25 patients per study month (75 in each cohort) who met the study criteria were included. Antibiotic use from the day of culture collection through day 5 was compared. Results: The primary outcome of median time to first antibiotic change based on microbiologic data was 50 hours before the intervention versus 21 hours after the intervention (P = .0006). Also, 56 postintervention regimens (75%) were eligible for change based on PNA panel results; actual change occurred in 30 regimens (54%). Median antibiotic days of therapy (DOTs) were 8 before the intervention versus 6 after the intervention (P = .07). For the patients with antibiotic changes made based on PNA panel results, the median time to first antibiotic change was 10 hours. For patients who were initially on inadequate therapy, time to adequate therapy was 67 hours before the intervention versus 37 hours after the intervention (P = .27). Conclusions: The PNA panel was associated with decreased time to first antibiotic change and fewer antibiotic DOTs. Its impact may have been larger if a higher percentage of potential antibiotic changes had been implemented. The PNA panel is a promising tool to enhance antibiotic stewardship.

Retrospective evaluation of a delirium order set utilizing nonpharmacologic and pharmacologic interventions for the treatment of delirium in medical intensive care unit patients.

PURPOSE: To describe the use of a medical intensive care unit (MICU) delirium order set pilot and its associated impact on utilization of nonpharmacologic and pharmacologic interventions, pharmacologic continuation at transitions of care, and resolution of ICU delirium. METHODS: This was a retrospective cohort analysis of MICU patients who received delirium management using an order set pilot compared to standard care. Patients 18 years of age or older admitted to the MICU between May 2019 and January 2020 who received an antipsychotic or valproic acid for the treatment of delirium were included. RESULTS: Pharmacologic treatment continuation past ICU discharge occurred in 30% of patients in the pilot cohort (n = 50) compared to 54% of patients receiving standard care (n = 50; P = 0.027). On treatment days 1 through 7, utilization of deliriogenic medications was significantly lower in the pilot cohort (78% vs 96%, P = 0.007). No differences were observed between the groups in delirium resolution, delirium recurrence, hospital and ICU length of stay, or mortality. CONCLUSION: A MICU order set prioritizing nonpharmacologic management and limiting the duration of pharmacologic agents for delirium may aid providers in the management of ICU delirium and reduce exposure to pharmacologic interventions.