Bloodstream infections in hospitalized patients before and during the COVID-19 surge in a community hospital in the South Bronx

Abstract

•Higher incidence of bloodstream infections (BSI) among patients during the COVID-19 surge.•Increased hospital-acquired and central line-associated BSI during the COVID-19 surge.•Enterococcus spp. was the most common microbe causing BSI during the COVID-19 surge. New York City's (NYC) public health system, which predominantly serves lower-income communities, bore the burden of care and had to ramp up services to respond to the rapidly evolving COVID-19 pandemic. An increase in critical-care beds, staffing, and equipment was integral to the response, especially in our hospital in the South Bronx, where the number of intensive care unit (ICU) beds were augmented from 34 to 195 (Uppal et al., 2020Uppal A Silvestri DM Siegler M Natsui S Boudourakis L Salway RJ Parikh M Agoritsas K Cho HJ Gulati R Nunez M Hulbanni A Flaherty C Iavicoli L Cineas N Kanter M Kessler S Rhodes KV Bouton M Wei EK. Critical Care And Emergency Department Response At The Epicenter Of The COVID-19 Pandemic.Health Aff (Millwood). 2020; 39 (AugEpub 2020 Jun 11PMID:32525713): 1443-1449https://doi.org/10.1377/hlthaff.2020.00901Google Scholar). Hospitalized patients with COVID-19 usually have severe/critical infection with acute respiratory distress syndrome (ARDS), shock, coagulopathies, and multiorgan failure (Zaim et al., 2020Zaim S Chong JH Sankaranarayanan V Harky A. COVID-19 and Multiorgan Response.Curr Probl Cardiol. 2020; 45100618https://doi.org/10.1016/j.cpcardiol.2020.100618Google Scholar). Multiple studies have confirmed that a high frequency of coinfections including bloodstream infections (BSI) have been observed in hospitalized patients with COVID-19 infection, like with other respiratory viruses (Lee et al., 2011Lee N Chan PKS Lui GCY et al.Complications and outcomes of pandemic 2009 influenza A (H1N1) virus infection in hospitalized adults: how do they differ from those in seasonal influenza?.J Infect Dis. 2011; 203: 1739-1747Google Scholar; Chertow and Memoli, 2013Chertow D Memoli M. Bacterial coinfection in influenza: a grand rounds review.JAMA. 2013; 309: 275-282Google Scholar; Bhatt et al., 2021 Jun 15Bhatt PJ Shiau S Brunetti L Xie Y Solanki K Khalid S Mohayya S Au PH Pham C Uprety P Nahass R Narayanan N. Risk Factors and Outcomes of Hospitalized Patients With Severe Coronavirus Disease 2019 (COVID-19) and Secondary Bloodstream Infections: A Multicenter Case-Control Study.Clin Infect Dis. 2021 Jun 15; 72 (PMID:33216875PMCID: PMC7717183): e995-e1003https://doi.org/10.1093/cid/ciaa1748Google Scholar). Our study aimed to compare the incidence of BSI, clinical and microbial characteristics of infection among patients with BSI before and during the surge of the COVID-19 pandemic. This is a single institution, retrospective study of adult hospitalized patients with BSI admitted before (Jan to Feb 2020) and during the COVID-19 surge (March to May 2020). All adult patients hospitalized to Medicine including Medical Intensive Care from January to May 2020 were included in the study if they had a laboratory-confirmed bloodstream infection during their stay. CLABSI (central line-associated bloodstream infection), defined according to the NHSN (National Healthcare Safety Network) 2020 criteria, is a BSI in a patient that had a central line in place for 48 or more hours before the development of the BSI and is not related to an infection at another site. Primary BSI was determined if the patient did not have a clear source of infection whereas secondary BSI was defined by the identification of the same microorganism in blood culture and the suspected source of infection. BSI was classified as community-acquired if occurred within 48 hours of hospital admission and hospital-acquired if occurred after 48 hours of admission. Polymicrobial BSI encompasses the identification of more than one species of microorganisms from a single positive blood culture. During the pandemic surge, all the blood cultures were collected by accessing the central line. Peripheral blood cultures were not collected. Hence, in the absence of an alternate source of infection, a positive blood culture was considered a CLABSI. Usual skin commensals were excluded from the analysis. Data regarding the incidence of BSI, clinical and microbial characteristics, various therapeutic interventions including central lines, days to positive blood cultures, DOT (days of antibiotic treatment) previous to BSI, length of stay, patient outcomes, etc. were obtained by retrospective chart review. Interventions associated with COVID-19 care including the use of steroids, mechanical ventilation, proning, inotropes, therapeutic anticoagulation were evaluated for correlation with BSI. Descriptive statistics and chi-square tests were used to compare the characteristics of infection. Univariate Cox regression models were used to evaluate the factors independently associated with the development of BSI during the COVID-19 period. Significant variables from this analysis were included in a multivariate model to determine risk associations for the most prevalent BSI. Of the 148 patients with BSI, 59 were admitted in the pre-COVID-19 period while 89 were admitted during the surge. Baseline characteristics and therapeutic interventions during hospitalization are shown in Table 1. The incidence of BSI was 4.37 per 1000 patient days in the pre-COVID-19 period compared with 8.36 during the surge (p = 0.004). A significant majority of patients during the COVID period had ARDS (40.4%), required mechanical ventilation (57.3%), pressors (46.1%), therapeutic anticoagulation (31.5%), proning (22.5%), rectal tube (29.2%), and steroids (28.1%) in comparison to the pre-COVID-19 period. The median DOT previous to BSI during the COVID-19 surge was 3 days (0 to 20.5) whereas it was <1 day (0 to 6) before the COVID-19 period. Mortality was higher among patients with BSI during the surge (40.4% vs 15.3%, p = 0.001).Table 1Baseline characteristics, comorbidities, risk factors for BSI & characteristics of BSI during the pre-COVID-19 period (Jan to Feb 2020) & COVID-19 period (March to May 2020).Overall n = 148Pre-COVID-19 Period n = 59COVID-19 Period n = 89P valuePatient characteristicsAge, median (IQR)60.0 (48.3 - 69.0)60.0 (45.0-69.0)60.0 (52.0-69.5)0.158Females, no (%)66 (44.6%)31 (52.5%)35 (39.3%)0.113Race0.094Hispanic85 (57.4%)30 (50.8%)55 (61.8%)Black32 (21.6%)10 (16.9%)22 (24.7%)White4 (2.7%)2 (3.4%)2 (2.2%)Asian3 (2.0%)2 (3.4%)1 (1.1%)Others24 (16.2%)15(25.4%)9 (10.1%)Body mass index0.158Underweight9 (6.1%)4 (6.8%)5 (5.6%)Normal55 (37.2%)28 (47.5%)27 (30.3%)Overweight34 (23.0%)12 (20.3%)22 (24.7%)Obese44 (28.4%)15 (25.4%)35 (39.3%)ComorbiditiesCharlson comorbidity index4.0 (2.0-6.0)3.0 (2.0-6.0)4.0 (2.0 -6.0)0.299Hypertension88 (59.5%)29 (49.2%)59 (66.3%)0.038Diabetes mellitus64 (43.2%)20 (33.9%)44 (49.4%)0.062Asthma/COPD29 (19.6%)13 (22.0%)16 (18.0%)0.543Autoimmune disease7 (3.9%)5 (8.5%)2 (2.2%)0.090Chronic kidney disease21 (14.2%)6 (10.2%)15 (16.9%)0.254Dementia7 (4.7%)1 (1.7%)6 (6.7%)0.157Previous history of cancer14 (9.5%)5 (8.5%)9 (13.0%)0.739HIV19 (12.8%)10 (16.9%)9 (10.1%)0.223Smoking history46 (31.1%)24 (40.7%)22 (24.7%)0.040Risk factors for BSIARDS on admission, n (%)41 (27.7%)5 (8.5%)36 (40.4%)0.001Mechanical ventilation, n (%)67 (45.3%)16 (27.1%)51 (57.3%)0.001Days of mechanical ventilation, median (IQR)14.0 (7.0-31.0)12.0 (1.5-46.2)16.0 (8.0 -34.0)0.128Pressor Use during hospitalization54 (36.5%)13 (22.0%)41 (46.1%)0.003Days of pressors, median (IQR)6.5 (2.8-13.0)4.0 (1.5-12.5)7.0 (3.0-13.5)0.400Proning, n (%)21 (14.2%)1 (1.7%)20 (22.5%)0.001Rectal tube, n (%)31 (20.9%)5 (8.5%)26 (29.2%)0.002Anticoagulation use, n (%)32 (21.6%)4 (6.8%)28 (31.5%)0.001Steroid use, n (%)35 (23.6%)10 (16.9%)25 (28.1%)0.118Length of stay- days, median (IQR)13.5 (4.3-29.0)9.0 (4.0 -21.8)17.5 (9.4-36.2)0.136Death, n (%)45 (30.4%)9 (15.3%)36 (40.4%)0.001BSI characteristicsOverall n = 164Pre-COVID-19 Period n = 53COVID-19 Period n = 111P valuePrimary BSI, n (%)87 (53.0%)42 (79.2%)45 (40.5%)0.001Secondary BSI, n(%)77 (47.0%)11 (20.8%)66 (59.5%)CLABSI, n (%)58 (35.4%)5 (9.4%)53 (47.7%)Number of central venous access per patient, median (IQR)1.0 (1.0-2.0)1.0 (1.0-2.0)1.0 (1.0-2.0)Community acquired BSI, n (%)83 (56.1%)43 (81.1%)40 (36.0%)0.001Hospital acquired BSI, n (%)81 (49.4%)10 (18.9%)71 (64.0%)0.001Time to first positive blood culture from admission – days, median (IQR)1.8 (1.5-8.8)1.7 (1.4-2.1)2.7 (1.5-14.2)0.003Days of antibiotic therapy previous to BSI (DOT), median (IQR)0 (0-15.0)0 (0-6.0)3.0 (0-20.5)0.017Polymicrobial BSI15 (9.1%)7 (13.2%)8 (7.2%)0.191 Open table in a new tab Of the 164 BSI events, 53 were pre-COVID-19 while 111 were during the surge. BSI during the COVID-19 period was predominantly monomicrobial (93%) and nosocomial (64%, p = 0.001). In the Pre-COVID-19 era, primary BSI was predominant compared with central line-associated secondary BSI seen during the COVID-19 surge. Enterococcus (22.5%), Staphylococcus aureus (13.5%) and Candida (8.1%) were more common BSI during the COVID-19 surge versus MRSA (15.1%), Escherichia coli (13.2%), and Streptococcus pneumoniae (13.2%) before COVID-19. On multivariate analysis, Enterococcal coinfection was strongly associated with SARS-CoV-2 positivity (odds ratio [OR] 2.685, p = 0.038), mechanical ventilation (OR 8.739, p = 0.002), and chronic obstructive pulmonary disease (COPD)/Asthma (OR 2.823, p = 0.035). Our results highlight the higher incidence of BSI during the COVID-19 surge compared with the pre-COVID-19 period at the NYC public hospital. The increased BSI during the COVID-19 surge in our hospital was higher than the pooled estimated BSI occurrence published in a systematic review (Ippolito et al., 2016Ippolito M. Simone B. Filisina C. Catalanotto F.R. Catalisano G. Marino C. Misseri G. Giarratano A. Cortegiani A. Bloodstream Infections in Hospitalized Patients with COVID-19: A Systematic Review and Meta-Analysis.Microorganisms. 2021; 9: 2016https://doi.org/10.3390/microorganisms9102016Google Scholar), due to our higher burden of critically ill patients requiring mechanical ventilation, pressors, steroids, proning, etc. Empiric antibiotics were used in almost all patients due to a paucity of information about COVID-19 resulting in higher DOT. Data reported by the Centers for Disease Control and Prevention (CDC)/NHSN has confirmed a significant increase in CLABSI throughout the United States in the early months of the pandemic. Due to modification of the infection control practices to accommodate the surge in cases, shortages of personal protective equipment (PPE), staff and supplies occurred. Further adjustments reduced the frequency of contact with patients – decreasing compliance with central line care bundles and disrupting protocols, during proning sessions (Patel et al., 2021Patel PR Weiner-Lastinger LM Dudeck MA et al.Impact of COVID-19 pandemic on central line-associated bloodstream infections during the early months of 2020, National Healthcare Safety Network [published online ahead of print, 2021 Mar 15].Infect Control Hosp Epidemiol. 2021; : 1-4https://doi.org/10.1017/ice.2021.108Google Scholar; McMullen et al., 2020McMullen KM Smith BA Rebmann T. Impact of SARS-CoV-2 on hospital-acquired infection rates in the United States: predictions and early results.Am J Infect Control. 2020; 48: 1409-1411Google Scholar). The increased mortality observed during the COVID-19 surge has been well described during previous influenza pandemics in patients with secondary BSI (Lee et al., 2011Lee N Chan PKS Lui GCY et al.Complications and outcomes of pandemic 2009 influenza A (H1N1) virus infection in hospitalized adults: how do they differ from those in seasonal influenza?.J Infect Dis. 2011; 203: 1739-1747Google Scholar; Chertow and Memoli, 2013Chertow D Memoli M. Bacterial coinfection in influenza: a grand rounds review.JAMA. 2013; 309: 275-282Google Scholar).