Severe COVID-19 is associated with critical illness and immune dysregulation, both of which have been previously associated with increased risk of nosocomial infection. The care of COVID-19 patients has required dramatic changes to usual hospital practices and heightened concern for infection control practices. This is a brief summary of published evidence related to the effect of the COVID-19 pandemic on non-COVID infections. Included are manuscripts published in peer-reviewed journals or on pre-print servers through March 10, 2021. This summary does not consider the second order effects of increased antibiotics associated with COVID-19 on antimicrobial resistance. References summarized in this report were drawn from the COVID-19 Literature Report (Lit Rep) team database. References that appeared in the daily Lit Rep are marked with an asterisk*, and the summary is shown in the annotated bibliography below.
Executive Summary
- Contamination of improperly used personal protective equipment likely contributed to an outbreak of nosocomial infection with Candida auris in a COVID-19-specific healthcare setting.
- Evidence is currently inconclusive regarding the risk of healthcare facility acquired (nosocomial) infections in patients with COVID-19 and whether critical illness associated with COVID-19 creates a higher risk of nosocomial infections compared to other forms of critical illness.
- The literature does not support routine use of empiric antibiotics in the management of confirmed COVID-19 infection.
- Successfully implemented infection control practices in response to COVID-19 have been associated with a decrease in the incidence of hospital acquired Clostridiodes difficile.
- The COVID-19 pandemic has been associated with a remarkable decline in other respiratory viruses, most notably influenza.
Outbreaks of Other Infections in Clinical Units Caring for Patients with COVID-19
At this time, outbreaks of fungal and bacterial infections in clinical units caring for patients with COVID-19 do not appear to be widespread; however ongoing surveillance is warranted to prevent such outbreaks in these settings. A cluster of infections with the yeast Candida auris was reported in a specialty care ward treating COVID-19 patients (Prestel*). Multiple studies have characterized changes in the frequency of infections with the common hospital-acquired bacterium Clostridioides (formerly Clostridium) difficile in the setting of COVID-19 care and have indicated that the incidence C. difficile has decreased in many healthcare facilities during the COVID-19 pandemic, potentially due to the implementation of stronger infection control measures.
Candida auris
C. auris is an emerging multidrug-resistant yeast that can cause invasive infection. Prior to the COVID-19 pandemic, outbreaks of C. auris have occurred in healthcare settings due in part to asymptomatic carriage and its persistence on surfaces (CDC, Tsay). Prevention and control measures have included aggressive implementation of contact tracing and screening in response to the identification of new cases. During July-August 2020, a hospital in Florida experienced an outbreak of C. auris that involved three C. auris bloodstream infections and one urinary tract infection in patients with COVID-19 who were being treated in a dedicated COVID-19 ward (Prestel*). Upon identification of these cases, screening of 67 patients admitted to the COVID-19 unit found that 35 (52%) had positive test results. An investigation of the outbreak concluded that widespread C. auris transmission was likely aided by the use of multiple gown and glove layers by the healthcare providers in the COVID-19 unit, extended use of the underlayer of personal protective equipment (PPE), lapses in cleaning and disinfection of shared medical equipment, and lapses in adherence to hand hygiene (Prestel*). After the hospital put in place measures to address these concerns, no subsequent C. auris cases were detected in follow-up surveys. The authors state that “CDC does not recommend the use of more than one isolation gown or pair of gloves at a time when providing care to patients with suspected or confirmed SARS-CoV-2 infection. Such practices among HCP might be motivated by fear of becoming infected with SARS-CoV-2 but instead might increase risks for self-contamination when doffing and for transmission of other pathogens among patients and exacerbate PPE supply shortages.”
Clostridioides (previously Clostridium) difficile
C. difficile is a common hospital acquired bacterial gastrointestinal infection with spores that are resistant to the alcohol-based sanitizers commonly used for hand hygiene in COVID-19 care. Hand hygiene procedures effective against C. difficile transmission include washing hands with soap and water. Antibiotic exposure is a risk factor for C. difficile. Several studies have characterized changes in C. difficile infection patterns in settings of COVID-19 clinical care. Luo et al found no significant difference in incidence of hospital-associated C. difficile episodes in a New York hospital despite a trend toward increased high-risk antibiotic exposures. A study from a hospital in Spain observed a 70% decrease in the incidence of hospital associated C. difficile during the COVID-19 pandemic compared to a pre-pandemic control period, despite no change in antibiotic use, which the authors attribute to the implementation of control measures to prevent nosocomial transmission by healthcare workers and asymptomatic colonized patients (Ponce-Alonso). Similar declines in hospital acquired C. difficile infections during the COVID-19 pandemic have been observed elsewhere (Bentivegna, Hazel).
Nosocomial Infections in Patients with COVID-19
Bacterial co-infections have been reported in hospital settings caring for patients with COVID-19, but the frequency of bacterial infections has been low. The Infectious Disease Society of America has concluded the following:
- Current data indicate that bacterial coinfections with SARS-CoV-2 infection are relatively infrequent (likely occurring in <10% of hospitalized COVID-19 patients).
- The literature does not support routine use of empiric antibiotics in the management of confirmed COVID-19 infection.
- Objective findings that increase the concern for bacterial superinfection include rise in leukocyte counts, lobar consolidation or evidence of necrotizing infection on chest imaging and recrudescence of fever after initial defervescence.
- Fungal superinfection (with Aspergillus) is also a concern, but the true incidence has not been defined. Risk factors for fungal superinfection include steroid use, invasive catheters and prolonged mechanical ventilation.
- Antimicrobial stewardship programs can help optimize antimicrobial use during the pandemic. Continued investigation into optimal antimicrobial stewardship program interventions to limit antibiotic overuse during the COVID-19 pandemic is warranted.
A multicenter analysis of the clinical microbiology usage in hospitalized patients in the US found patients with COVID-19 had similar rates of non-SARS-CoV-2 co-infections compared to patients without COVID-19, but that COVID-19 disease was associated with higher rates of hospital-onset infections, greater antimicrobial usage, and extended hospital and ICU length of stay (Puzniak). In a case series of patients with COVID-19 in the UK, bacterial co-infections were identified in 3% of patients 0-5 days after admission and in 6% of patients throughout hospital admission (Hughes). Comparable rates of bacterial co-infection were identified in a control group of patients with confirmed influenza infection. In one study of samples with a positive co-infection culture from patients hospitalized in New York City during the first COVID-19 surge, the most commonly isolated organisms in respiratory infections were Staphylococcus aureus (44%), Pseudomonas aeruginosa (16%), Klebsiella spp (10%), Enterobacter spp (8%), and Escherichia coli (4%) (Nori). The most commonly isolated organisms in bloodstream infections were S. aureus (30%), Staphylococcus epidermidis (12%), Streptococcus spp (10%), Enterococcus spp (7%), E. coli (7%), P. aeruginosa (6%), Klebsiella spp (3%), and Enterobacter spp (3%).
Invasive Pulmonary Aspergillosis
Several studies have raised concern that COVID-19 patients may be more susceptible to invasive fungal infections, particularly invasive pulmonary aspergillosis (IPA), through multiple mechanisms, including declines in lymphocyte counts and viral damage to airway tissues (Koehler). Several case series and observational cohort studies, most of which have involved relatively small numbers of patients, have found that 20-30% of patients with severe COVID-19, particularly those receiving mechanical ventilation, have invasive pulmonary aspergillosis, with a high case fatality (~65-75%) among these patients. However, the Infectious Disease Society of America has urged caution regarding these findings, outlining the limitations in interpreting these observational studies, and has concluded that currently, the true incidence of COVID-19 associated pulmonary aspergillosis is not known.
A literature review on the topic of IPA in patients with COVID-19 with severe or critical illness concluded that the incidence of IPA in patients with COVID-19 ranged from 20% to 33% (Lai). Most studies included in the literature review involved a relatively small number of patients (~30), although one study from Jiangsu Province, China from January 22 to February 2, 2020 found IPA in 60 (23%) out of 243 patients. The case fatality among patients with COVID-19 who were co-infected with IPA was high (65% among a cohort of 34 patients). In an Italian cohort of 108 patients with confirmed COVID-19 (4 ICUs from 3 hospitals) who were receiving mechanical ventilation, 28% were found to have pulmonary aspergillosis and 18% were classified as having putative IPA (Bartoletti). The 30-day mortality was significantly higher among patients with pulmonary aspergillosis (44% vs 19%, P = 0.002) and IPA (74% vs 26%, P < .001) compared to patients without aspergillosis. A summary of reports of pulmonary aspergillosis and guidance on diagnosis is provided by Armstrong-James et al.
Other studies have not found pulmonary aspergillosis to be common in COVID-19 patients. An investigation in a single center in France of 54 patients with COVID-19 with moderate to severe acute respiratory distress syndrome (ARDS) found that two patients (3.7%) showed early putative IPA, neither of whom had prior immunosuppression or host risk factors (Versyck*). Immunosuppression was observed in two patients, neither of whom had IPA. The authors conclude that the frequency of IPA among patients with COVID-19 with ARDS was relatively low and was similar to what has been described in other populations with ARDS.
Nosocomial infections associated with immunomodulatory treatments
There are two primary immunomodulatory therapies currently in use for the treatment of severe COVID-19. These include corticosteroids (primarily dexamethasone) and IL-6 blockers (primarily tocilizumab). Since each of these medications target the immune system, it is biologically plausible that they would be associated with secondary infections.
Tocilizumab
Studies have reported mixed results regarding the risk of secondary infections associated with tocilizumab. A systematic review found five studies have reported an increased prevalence of infection whereas twelve studies have reported either no association or reduced infections in patients treated with tocilizumab (Khan). The differences between these studies that lead to such wide discrepancies in results are unclear. It is worth noting that almost all studies followed patients for only 28 days, whereas the half-life of tocilizumab is long with the doses typically used for treatment in COVID-19 (8mg/kg intravenous), leading to a half-life of 18 days (Abdallah). One hypothesis for why some studies have found an elevated risk of infection in the placebo arm is that tocilizumab is very effective at suppressing fever, suggesting that there may be fewer investigations for infections performed (Strohbehn).
A randomized, double blind trial of severe COVID-19 determined that there were fewer “serious infections” in tocilizumab patients. Details regarding the nature of these infections are unclear (Stone). Industry-sponsored randomized trials have not found a difference in the risk of “serious” secondary infections when compared to placebo. These include COVACTA and EMPACTA (Salama, Rosas). In the former study, the majority of participants received glucocorticoids (~80%) whereas in the latter only 20-30% of participants received glucocorticoids. The RECOVERY Trial (a large multicenter public sector trial) did not find differences in “excess deaths from other infections” among patients who received tocilizumab. Another randomized trial in patients with moderate-to-severe COVID who were not intubated did not find a difference in secondary infections (Hermine). In India, the COVINTOC trial (open label) recently reported no differences in infections between treatment arm and standard of care (Soin).
Some observational studies of tocilizumab under real-work conditions have also found no increased risk of secondary infections (Ignatius, Campochino et al., Eur J Int Med 2020), while others have found a substantially elevated risk of infection, including some autopsy proved bacterial infections (Kimmig, Guaraldi).
Corticosteroids
We identified few investigations of the potential risk of secondary infections associated with short term corticosteroid use for the treatment COVID-19. The median duration of dexamethasone in the RECOVERY trial was seven days and no increases in other infections were reported (Horby*).
Indirect Effects of the COVID-19 Pandemic on Vaccination
Stay-at-home orders intended to reduce the transmission of SARS-CoV-2 and concerns about acquiring SARS-CoV-2 in health care settings had the indirect of effect of reducing the rate of administration of routine vaccinations, at least in the early phase of the pandemic (Lassi*). Vaccination rates for bacterial pneumonia, tetanus-diphtheria-pertussis, and shingles among Medicare beneficiaries over the age of 65 declined by 25% to 62% compared with the corresponding period in 2019, reaching a low point in mid-April 2020 and recovering slowly in between May and July 2020 (Hong*). While national vaccination coverage among US kindergarteners was high for the 2019-20 school year, despite most schools shifting to virtual learning in the spring due to the COVID-19 pandemic, the CDC authors caution that disruptions caused by the pandemic are likely to reduce vaccination coverage for the 2020-21 school year (Seither*).
Indirect Effects of the COVID-19 Pandemic on Other Respiratory Viruses
Widespread implementation of measures to prevent community transmission of SARS-CoV-2, including stay-at-home orders, physical distancing, use of facemasks, and other non-pharmaceutical interventions (NPI) appears to have had broad indirect effects on the transmission of other respiratory infections. Evidence from many settings has shown dramatically lower incidence of influenza, rhinovirus, enterovirus, and respiratory syncytial virus (RSV) infections following the implementation of NPIs to prevent SARS-CoV-2 compared to the same time period in years prior to the COVID-19 pandemic (Freeman*, Partridge*, Sherman*). An examination of temporal trends in the incidence of these infections indicates that they occurred at rates similar to previous years during the period immediately prior to the COVID-19 pandemic and declined rapidly following implementation of NPIs (Partridge*, Zhang*). Evidence from children’s hospitals indicates that declines in influenza, RSV, and hospitalization for lower respiratory infections have also declined substantially in children (Trenholme*).