To the Editor:
Hospitalized patients, particularly those who are critically ill, are at risk for “secondary” infections (1, 2). Initial reports of patients hospitalized with coronavirus disease (COVID-19) indicate that 10–33% develop bacterial pneumonia (3, 4) and 2–6% develop bloodstream infection (BSI) (5, 6). Few studies have reported patient characteristics or the impact of intensive care unit (ICU) admission on secondary infections (3, 6–8). We conducted a descriptive study to identify the prevalence, microbiology, and outcomes of secondary pneumonias and BSIs in patients hospitalized with COVID-19.
Methods
The Emory University Institutional Review Board approved this study. Patients :18 years old with a positive severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) real-time polymerase chain reaction assay admitted to four academic hospitals in Atlanta, Georgia, from February 15 to May 16, 2020, were included. Data were extracted from the electronic medical record (Cerner Millennium) through June 16, 2020, including comorbidities (identified by International Classification of Diseases, 10th revision codes).
Blood cultures were incubated in BACT/ALERT 3D (bioMérieux, Inc.), and respiratory cultures were inoculated on 5% sheep blood, chocolate, and MacConkey agars. Organisms were identified by matrix-assisted laser desorption ionization-time of flight mass spectrometry (bioMérieux, Inc.). Susceptibility testing was performed by Vitek 2 (bioMérieux, Inc.).
We used the U.S. Centers for Disease Control and Prevention (CDC) criteria to determine ventilator-associated events (VAEs), including infection-related ventilator-associated complications (IVACs) and possible ventilator-associated pneumonia (PVAP) (9).
Blood cultures were considered contaminated if one of two sets grew a typically nonpathogenic organism (e.g., coagulase-negative staphylococci) or if the clinical team determined the organism a contaminant. Two of three infectious diseases physicians (M.W.A., D.R.B., and A.B.) reviewed BSIs to determine clinical source and a third (J.T.J.) arbitrated disagreements. Infections were attributed to skin if there was a clinically infected wound or peripheral intravenous line but no central line.
We assessed in-hospital mortality, comparing patients with and without infections using the χ2 test. SAS University Edition (SAS Institute) was used for data analysis.
Results
Patients
Among 774 patients hospitalized with COVID-19, the median age was 62 years (interquartile range, 50–73), 49.7% were female, and 66.6% were Black (Table 1). Hypertension (75.5%) and diabetes mellitus (45.7%) were the most common comorbidities. Three hundred thirty-five (43.3%) required ICU admission, 238 (30.7%) required mechanical ventilation, and 120 (15.5%) died.
Table 1.
Characteristics of 774 adults hospitalized with COVID-19 in Atlanta, Georgia
| Characteristic | Value |
|---|---|
| Age, median (IQR), yr | 62 (50–73) |
| Female sex, n (%) | 385 (49.7) |
| Hispanic ethnicity, n (%) | 48 (6.2) |
| Body mass index, median (IQR), kg/m2 (n = 759) | 29 (24–34) |
| Elixhauser comorbidity index, median (IQR) (n = 773) | 5 (3–8) |
| Race, n (%) | |
| Black | 514 (66.4) |
| White | 156 (20.2) |
| Unknown | 86 (11.1) |
| Asian | 17 (2.2) |
| American Indian/Alaska Native | 1 (0.1) |
| Comorbidities, n (%) | |
| Hypertension | 584 (75.5) |
| Diabetes mellitus | 354 (45.7) |
| Chronic kidney disease | 242 (31.2) |
| Cancer | 130 (16.8) |
| Liver disease | 93 (12.0) |
| Immunocompromised | 56 (7.2) |
| Laboratory results, median (IQR) | |
| WBC, ×103/µl | 7.7 (5.9–10.8) |
| Lymphocyte count, ×103/µl (n = 673) | 1.2 (0.9–1.7) |
| Hemoglobin, g/dl | 12.8 (11.5–14.1) |
| Platelets, ×103/µl | 247 (189–322) |
| Creatinine, mg/dl (n = 773) | 1.2 (0.9–1.9) |
| AST, U/L (n = 760) | 38 (25–57) |
| ALT, U/L (n = 760) | 25 (16–42) |
| Total bilirubin, mg/dl (n = 760) | 0.6 (0.4–0.8) |
| D-dimer, ng/dl (n = 579) | 1360 (755–3663) |
| CRP, mg/L (n = 590) | 131 (70–206) |
| ESR, mm/h (n = 78) | 58 (36–85) |
| LDH, U/L (n = 569) | 338 (259–452) |
| Ferritin, ng/ml (n = 385) | 433 (184–1019) |
| Troponin-I, ng/ml (n = 596) | 0.03 (0.03–0.08) |
| IL-6, pg/ml (n = 166) | 9 (5–20) |
| Medications administered, n (%) | |
| Dexamethasone | 54 (7.0) |
| Any steroid | 146 (18.9) |
| Outcomes | |
| Intensive care unit admission, n (%) | 335 (43.2) |
| Mechanical ventilation, n (%) | 238 (30.7) |
| Duration of mechanical ventilation, median (IQR), d | 10 (5–16) |
| Died, n (%) | 120 (15.5) |
Definition of abbreviations: ALT = alanine aminotransferase; AST = aspartate aminotransferase; COVID-19 = coronavirus disease; CRP = C-reactive protein; ESR = erythrocyte sedimentation rate; IL-6 = interleukin-6; IQR = interquartile range; LDH = lactate dehydrogenase; WBC = white blood cells.
Respiratory infections
Among 238 intubated patients, 201 (84.5%) had at least one respiratory culture sent, and 65 (27.3%) had a positive respiratory culture, with a total of 84 potential pathogens (Table 2). The most common bacteria were Staphylococcus aureus (29/84; 34.5%), Pseudomonas aeruginosa (16/84; 19.0%), and Klebsiella spp. (14/84; 16.7%), with only one Aspergillus spp. Mortality did not differ between intubated patients with an identified bacterial respiratory pathogen and those without (41.5% vs. 35.3%, P = 0.37). Forty-six patients (19.3%) had a CDC-defined VAE (15.3 VAEs per 1,000 ventilator-days), 16 (6.7%) had an IVAC (5.3 IVACs per 1,000 ventilator-days), and 5 (2.1%) had a PVAP (1.7 PVAPs per 1,000 ventilator-days). Eleven (23.9%) patients with a VAE required a tracheostomy and 25 (54.3%) died. None of the five patients with PVAP died.
Table 2.
Characteristics of potential respiratory pathogens (n = 84) in 774 patients hospitalized with COVID-19
| Characteristic | n (%) |
|---|---|
| Organism | |
| S. aureus | 29 (34.5) |
| P. aeruginosa | 16 (19.1) |
| Klebsiella species | 14 (16.7) |
| Streptococcus species | 4 (4.8) |
| E. cloacae | 3 (3.6) |
| H. influenzae | 2 (2.4) |
| B. cepacia | 2 (2.4) |
| A. baumannii | 1 (1.2) |
| A. fumigatus | 1 (1.2) |
| Other* | 12 (14.3) |
| Multidrug resistance | |
| None | 60 (71.4) |
| Methicillin resistance | |
| S. aureus | 14 (16.7) |
| S. lugdunensis | 1 (1.2) |
| Extended-spectrum β-lactamase | |
| Klebsiella species | 3 (3.6) |
| Enterobacter cloacae | 1 (1.2) |
| Carbapenem resistance | |
| P. aeruginosa | 3 (3.6) |
| K. aerogenes | 1 (1.2) |
Definition of abbreviations: A. baumannii = Acinetobacter baumannii; A. fumigatus = Aspergillus fumigatus; B. cepacia = Burkholderia cepacia; C. indologenes = Chryseobacterium indologenes; COVID-19 = coronavirus disease; E. cloacae = Enterobacter cloacae; H. alvei = Hafnia alvei; H. influenzae = Haemophilus influenzae; K. aerogenes = Klebsiella aerogenes; P. aeruginosa = Pseudomonas aeruginosa; S. aureus = Staphylococcus aureus; S. lugdunensis = Staphylococcus lugdunensis; S. marcescens = Serratia marcescens.
Corynebacterium spp. (n = 7), S. lugdunensis (n = 2), C. indologenes (n = 1), H. alvei (n = 1), S. marcescens (n = 1).
Among 536 (69.3%) nonintubated patients, 186 (34.7%) had Legionella urine antigens sent, and two (0.4%) were positive. Sixty-nine (12.9%) had at least one respiratory culture sent, and one was positive (for S. aureus). No other bacterial or fungal respiratory pathogens were identified in nonintubated patients.
Bloodstream infections
Of 774 patients, 588 (76.0%) had at least one blood culture sent; 48 (6.2%) had contaminated blood cultures and 36 (4.7%) had BSI, including 5 with polymicrobial BSI (42 isolates total) (Table 3). The majority of BSIs (24/36; 66.7%) had ICU onset. The most common organisms were S. aureus (7/42; 16.7%), Candida spp. (7/42; 16.7%), and coagulase-negative staphylococci (5/42; 11.9%); 12 (28.6%) were gram negative. The most common source was central line–associated BSI (CLABSI, 17/36; 47.2%), followed by skin (6/36; 16.7%), lungs (5/36; 13.9%), and urine (4/36; 11.1%). Overall, mortality was 50% in patients with BSI versus 13.8% in those without (P < 0.0001). Among intubated patients, mortality was 51.9% in patients with BSI versus 35.1% in those without (P = 0.09).
Table 3.
Characteristics of bloodstream infections (n = 36) in 774 patients hospitalized with COVID-19
| Characteristic | n (%) |
|---|---|
| Source | |
| Central line | 17 (47.2) |
| Skin | 6 (16.7) |
| Pulmonary | 5 (13.9) |
| Urine | 4 (11.1) |
| Gastrointestinal | 3 (8.3) |
| Unknown | 1 (2.8) |
| Organism (n = 42)* | |
| S. aureus | 7 (16.7) |
| Candida species | 7 (16.7) |
| Streptococcus species | 6 (14.3) |
| Coagulase negative staphylococci | 5 (11.9) |
| Enterococcus species | 5 (11.9) |
| E. coli | 3 (7.1) |
| Klebsiella species | 3 (7.1) |
| Other gram-negative organisms† | 6 (14.3) |
| Multidrug resistance (n = 42)* | |
| None | 30 (71.4) |
| Methicillin resistance | |
| S. aureus | 4 (9.5) |
| Coagulase-negative staphylococci | 2 (4.8) |
| Vancomycin resistance | |
| Enterococci | 3 (7.1) |
| Carbapenem resistance | |
| Enterobacterales | 2 (4.8) |
| P. aeruginosa | 1 (2.4) |
| Epidemiologic class‡ | |
| Hospital-onset | 26 (72.2) |
| Healthcare-associated, community-onset | 8 (22.2) |
| Community-onset | 2 (5.6) |
| Collection location | |
| Intensive care unit | 24 (66.7) |
| Emergency department | 10 (27.8) |
| Ward | 2 (5.6) |
| Outcomes | |
| Persistent bloodstream infection§ | 0 |
| Mortality (in-hospital) | 18 (50.0) |
Definition of abbreviations: COVID-19 = coronavirus disease; E. cloacae = Enterobacter cloacae; E. coli = Escherichia coli; P. aeruginosa = Pseudomonas aeruginosa; P. melaninogenica = Prevotella melaninogenica; P. mirabilis = Proteus mirabilis; S. aureus = Staphylococcus aureus; S. marcescens = Serratia marcescens.
Five patients had polymicrobial bacteremia; there were 42 total isolates.
Acinetobacter sp., E. cloacae, P. melaninogenica, P. mirabilis, P. aeruginosa, S. marcescens.
Using U.S. Centers for Disease Control and Prevention guidelines (15).
Positive blood cultures for the same organism at least 72 hours apart despite appropriate antibiotics.
Discussion
In this cohort of 774 patients hospitalized with COVID-19, nearly one-third required mechanical ventilation, of whom 27% had positive respiratory cultures and 2% had ventilator-associated pneumonia. Thirty-six patients (5%) developed BSI, of whom 50% died. Secondary infections were associated with traditional risk factors for healthcare-associated infections, including indwelling medical devices, and predominantly in the ICU.
Secondary bacterial pneumonia, commonly due to S. pneumoniae and S. aureus, complicates 20–30% of influenza (10), and methicillin-resistant S. aureus pneumonia has been reported with severe acute respiratory syndrome coronavirus 1 (SARS-CoV-1) (11). In our cohort of patients with COVID-19, the risk of secondary bacterial pneumonia was much lower: only three nonintubated patients had microbiologic evidence of bacterial pneumonia. In contrast, intubated patients had a high proportion of cultures positive for S. aureus, P. aeruginosa, and Klebsiella spp., which concurs with a UK multicenter study of patients with COVID-19 with suspected ventilator-associated pneumonia (7). Importantly, respiratory bacterial pathogens isolated from patients with COVID-19 are similar to pathogens that cause hospital-acquired pneumonia in patients without COVID-19 (2, 12), with no change from pre–COVID-19 ICU-level resistance rates. These results suggest that hospitalization and intubation are more important than are COVID-19–specific effects in conferring susceptibility to specific pathogens. We did not observe a mortality difference in intubated patients with or without positive respiratory cultures. This may be due to significant empiric antibiotic use and warrants future investigation.
BSIs in our cohort were also largely related to risk factors and pathogens associated with hospitalization. The majority (66.7%) were ICU-onset and nearly half (47.2%) were CLABSIs, which may be related to an increase above the pre–COVID-19 baseline in central line–days among ICU patients (J. Jacob, unpublished results). As with respiratory pathogens, the BSI pathogens (S. aureus, Candida spp., and coagulase-negative staphylococci) were typical for healthcare-associated BSI in patients without COVID-19 (12). The observed high proportion of contaminants—severalfold higher than the baseline contamination rate of 0.4–1.8% in study hospitals during January and February 2020—may be due to obtaining blood cultures through central lines to minimize exposure for phlebotomists. Importantly, there was no change in observed daily bathing and central line maintenance practices observed by infection prevention teams. Maintaining central access in critically ill patients with COVID-19 can minimize emergent central line placement and potential healthcare worker SARS-CoV-2 exposure. Clinicians should continue to balance the need for central venous access with risk of CLABSI.
Our study has several limitations. First, these results are from academic hospitals in the southeastern United States and may not be generalizable to other settings. Second, we used International Classification of Diseases, 10th revision codes to determine comorbidities, which may be less accurate than reviewer adjudication. Third, we defined ventilator-associated pneumonia using CDC surveillance definitions (9), which may not represent clinical practice and have not been validated for use in patients with COVID-19. To mitigate this limitation, we presented data on all respiratory pathogens to inform empiric antimicrobials for patients with COVID-19 with suspected ventilator-associated pneumonia. Fourth, our study period was before publication of the RECOVERY (Randomised Evaluation of COVID-19 Therapy) dexamethasone trial (13), and corticosteroids were not routinely administered; only 7% of this cohort received any dexamethasone. Future studies should assess the risk of corticosteroids on secondary infections in patients with COVID-19. Finally, results from this descriptive study are not risk-adjusted and therefore should not be used to infer risk associated with different interventions (e.g., risk of BSI in patients with COVID-19 with vs. without central lines).
Overall, few of the 744 patients hospitalized with COVID-19 developed secondary bacterial pneumonia (2%) or BSI (5%). This is consistent with findings from a living meta-analysis reporting a 7% prevalence of bacterial secondary infections in patients with COVID-19 (14). Our analysis adds to this previous literature by demonstrating that the risk factors for these infections (intubation and central lines, respectively) and causative pathogens reflect healthcare delivery and not a COVID-19–specific effect. Our results suggest that clinicians, especially ICU clinicians, should adhere to standard best practices for preventing and empirically treating secondary infections in patients hospitalized with COVID-19.
Acknowledgments
Acknowledgment
The authors thank their colleagues in Emory Healthcare and the Emory Critical Care Center who have worked so hard to provide excellent clinical care during this global pandemic. They also thank the Emory Medical Laboratory personnel, who continued their usual excellence despite a significant increase in specimens from patients with COVID-19.
Footnotes
Supported by the National Institute of Allergy and Infectious Diseases grants K23AI144036 (M.H.W.) and K23AI134182 (S.C.A.) and the National Center for Advancing Translational Sciences grant UL1TR002378, both at the National Institutes of Health. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Author Contributions: M.W.A., D.R.B., A.C.H.-R., A.B., M.H.W., C.R., and J.T.J. developed the study design. M.W.A., D.R.B., A.C.H.-R., A.B., D.J.M., S.C.A., C.S.K., and J.T.J. performed data analysis and/or interpretation. M.W.A., D.R.B., A.B., S.C.A., and J.T.J. wrote the manuscript.
Author disclosures are available with the text of this letter at www.atsjournals.org.
Contributor Information
Collaborators: the Emory COVID-19 Quality and Clinical Research Collaborative members, Max W. Adelman, Scott Arno, Sara C. Auld, Theresa Barnes, William Bender, James M. Blum, Gaurav Budhrani, Stephanie Busby, Laurence Busse, Mark Caridi-Scheible, David Carpenter, Nikulkumar Chaudhari, Craig M. Coopersmith, Gordon Dale, Lisa Daniels, Johnathan A. Edwards, Jane Fazio, Babar Fiza, Eliana Gonzalez, Ria Gripaldo, Charles Grodzin, Robert Groff, Alfonso C. Hernandez-Romieu, Max Hockstein, Dan Hunt, Craig S. Jabaley, Jesse T. Jacob, Colleen S. Kraft, Greg S. Martin, Samer Melham, Nirja Mehta, Chelsea Modlin, David J. Murphy, Jung Park, Deepa Patel, Cindy Powell, Amit Prabhakar, Jeeyon Rim, Ramzy Rimawi, Chad Robichaux, Nicholas Scanlon, Milad Sharifpour, Bashar Staitieh, Michael Sterling, Jonathan Suarez, Colin Swenson, Nancy Thakkar, Alexander Truong, Hima Veeramachaneni, Alvaro Velasquez, Aimee Vester, Michael Waldmann, Max Weinmann, Thanushi Wynn, and Joel Zivot
References
- 1. De Santis V, Gresoiu M, Corona A, Wilson AP, Singer M. Bacteraemia incidence, causative organisms and resistance patterns, antibiotic strategies and outcomes in a single university hospital ICU: continuing improvement between 2000 and 2013. J Antimicrob Chemother. 2015;70:273–278. doi: 10.1093/jac/dku338. [DOI] [PubMed] [Google Scholar]
- 2. Ibn Saied W, Mourvillier B, Cohen Y, Ruckly S, Reignier J, Marcotte G, et al. OUTCOMEREA Study Group. A comparison of the mortality risk associated with ventilator-acquired bacterial pneumonia and nonventilator ICU-acquired bacterial pneumonia. Crit Care Med. 2019;47:345–352. doi: 10.1097/CCM.0000000000003553. [DOI] [PubMed] [Google Scholar]
- 3.Crotty MP, Akins RL, Nguyen AT, Slika R, Rahmanzadeh K, Wilson MH, et al. Investigation of subsequent and co-infections associated with SARS-CoV-2 (COVID-19) in hospitalized patients [preprint] medRxiv 2020. [accessed 2021 Jul 13]. Available from: https://www.medrxiv.org/content/10.1101/2020.05.29.20117176v2
- 4.Kimmig LM, Wu D, Gold M, Pettit NN, Pitrak D, Mueller J, et al. IL6 inhibition in critically ill COVID-19 patients is associated with increased secondary infections. Front Med [online ahead of print] 28 Oct 2020; DOI: 10.3389/fmed.2020.583897 [DOI] [PMC free article] [PubMed]
- 5. Goyal P, Choi JJ, Pinheiro LC, Schenck EJ, Chen R, Jabri A, et al. Clinical characteristics of COVID-19 in New York City. N Engl J Med. 2020;382:2372–2374. doi: 10.1056/NEJMc2010419. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Sepulveda J, Westblade LF, Whittier S, Satlin MJ, Greendyke WG, Aaron JG, et al. Bacteremia and blood culture utilization during COVID-19 surge in New York City. J Clin Microbiol. 2020;58:e00875-20. doi: 10.1128/JCM.00875-20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Dhesi Z, Enne VI, Brealey D, Livermore DM, High J, Russell C, et al. Organisms causing secondary pneumonias in COVID-19 patients at 5 UK ICUs as detected with the FilmArray test [preprint]. medrXiv; 2020 [accessed 2021 Jul 13]. Available from https://www.medrxiv.org/content/10.1101/2020.06.22.20131573v1.
- 8. Lansbury L, Lim B, Baskaran V, Lim WS. Co-infections in people with COVID-19: a systematic review and meta-analysis. J Infect. 2020;81:266–275. doi: 10.1016/j.jinf.2020.05.046. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Magill SS, Klompas M, Balk R, Burns SM, Deutschman CS, Diekema D, et al. Developing a new, national approach to surveillance for ventilator-associated events: executive summary. Clin Infect Dis. 2013;57:1742–1746. doi: 10.1093/cid/cit577. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Klein EY, Monteforte B, Gupta A, Jiang W, May L, Hsieh YH, et al. The frequency of influenza and bacterial coinfection: a systematic review and meta-analysis. Influenza Other Respir Viruses. 2016;10:394–403. doi: 10.1111/irv.12398. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Yap FH, Gomersall CD, Fung KS, Ho PL, Ho OM, Lam PK, et al. Increase in methicillin-resistant Staphylococcus aureus acquisition rate and change in pathogen pattern associated with an outbreak of severe acute respiratory syndrome. Clin Infect Dis. 2004;39:511–516. doi: 10.1086/422641. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Magill SS, O’Leary E, Janelle SJ, Thompson DL, Dumyati G, Nadle J, et al. Emerging Infections Program Hospital Prevalence Survey Team. Changes in prevalence of health care-associated infections in U.S. hospitals. N Engl J Med. 2018;379:1732–1744. doi: 10.1056/NEJMoa1801550. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Horby P, Lim WS, Emberson JR, Mafham M, Bell JL, Linsell L, et al. RECOVERY Collaborative Group. Dexamethasone in hospitalized patients with COVID-19. N Engl J Med. 2021;384:693–704. doi: 10.1056/NEJMoa2021436. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Langford BJ, So M, Raybardhan S, Leung V, Westwood D, MacFadden DR, et al. Bacterial co-infection and secondary infection in patients with COVID-19: a living rapid review and meta-analysis. Clin Microbiol Infect. 2020;26:1622–1629. doi: 10.1016/j.cmi.2020.07.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Kourtis AP, Hatfield K, Baggs J, Mu Y, See I, Epson E, et al. Emerging Infections Program MRSA author group. Vital signs: epidemiology and recent trends in methicillin-resistant and in methicillin-susceptible Staphylococcus aureus bloodstream infections - United States. MMWR Morb Mortal Wkly Rep. 2019;68:214–219. doi: 10.15585/mmwr.mm6809e1. [DOI] [PMC free article] [PubMed] [Google Scholar]