Bacterial Coinfection in COVID-19

To the Editor—We read with interest the work of Rawson et al, “Bacterial and Fungal Coinfection in Individuals With Coronavirus: A Rapid Review to Support COVID-19 Antimicrobial Prescribing” [1]. In most of the cited studies there is no distinction made on the timing of acquisition of the infection relative to the patients’ coronavirus disease 2019 (COVID-19) diagnosis. This results in the inclusion of both coinfections: 2 separate infectious processes contemporaneously and secondary infections; and a second infective process developing as a result of the first. In fact, almost all studies considered by Rawson et al examine infections secondary to COVID-19.

The North Middlesex University Hospital (NMUH) was one of the most COVID-19 affected hospitals in the early stages of the pandemic in the United Kingdom [2]. At this time the prevalence of COVID-19 among the community served by NMUH was high: from 1 March 2020 to 30 April 2020, 728 of the 1944 (37%) patients tested by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) polymerase chain reaction (PCR) returned positive results. This study examined the incidence of diagnosis at presentation of both COVID-19 and a confirmed bacterial bloodstream coinfection.

From 1 March 2020 to 30 April 2020, 420 patients were identified as SARS-CoV-2 PCR positive on nasopharyngeal swab at the time of admission to NMUH. Eleven (3%) also had a significant positive blood culture (excluding the growth of skin flora organisms in a single set of blood cultures). These patients were older (median 83 years, interquartile range [IQR] 71–86) than the cohort of admitted COVID-19 patients as a whole (median 64, IQR 50–79). All had ≥1 comorbidity that has been identified as a risk factor for severe COVID-19 disease [3] (Table 1). The range of clinical presentations, organisms identified, and underlying causative pathologies were diverse (Table 1). Only 2/11 (18%) patients reported respiratory symptoms, and 4/11 (36%) reported fever. Although the prevalence of respiratory symptoms was low, 6/11 (55%) had a chest radiograph consistent with COVID-19 infection at the time of presentation. Despite the universal treatment of severely unwell emergency department patients with ceftriaxone at this time, the outcomes of patients with COVID-19 and bacteremia were poor: 7/11 (64%) patients died during their admission, and the remaining 4 (36%) had prolonged hospital admissions (8–17 days, median 14 days).

Table 1.

Summary of the Characteristics of Patients Presenting to North Middlesex University Hospital in March−April 2020 With Both PCR Confirmed Coronavirus Disease 2019 (COVID-19) and Bacteremia

Age	Presentation	Relevant Background and Comorbidities	Organism	Source of Bacteremia	Outcome
92	Collapse	CVD T2DM	S. aureus	Not identified	Died
71	Collapse	COPD CVD T2DM	S. aureus	Not identified	Died
86	Collapse	CVD	S. parasanguinis	Skin/soft tissue infection leg	Died
58	Fever, lethargy, cough	T2DM	E. faecalis and K. pneumoniae	Hepatic abscess and infected biliary stent	Discharged
70	Deranged blood sugars and vomiting	Pancreatic cancer T2DM	E. coli	Presumed hepatobiliary	Died
88	Fever and fall	Care home resident CVD	E. coli	Likely urinary	Discharged
86	Fever and dysuria	Care home resident CVD	K. pneumoniae and E. coli	Urinary	Died
85	Collapse and dysuria	Care home resident CKD CVD	P. mirabilis	Urinary	Discharged
81	Fever and fall	CVD T2DM	P. mirabilis	Urinary	Discharged
83	Diarrhea and lethargy	Care home resident CVD IBD Pressure sores	E. coli	Sacral osteomyelitis	Died
53	Shortness of breath	CKD CVD	S. epidermidis	Line infection	Died

Abbreviations: CVD, cardiovascular disease including hypertension; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; IBD, inflammatory bowel disease; PCR, polymerase chain reaction; T2DM, type 2 diabetes mellitus.

The majority, if not all, of these cases represented true bacterial coinfection of an etiology independent of COVID-19 (Table 1). This suggests that in times of high COVID-19 prevalence Hickam’s dictum, “a patient may have as many diseases as he pleases,” trumps Occam’s razor, the principle that a single explanation for the patient’s symptoms is most likely, particularly in older patients.

Rawson et al rightly identify the need for antibiotic stewardship in the era of COVID-19, especially given low rates of confirmed bacterial infection [1]. However, the nonspecific presentation of COVID-19 patients with bacterial coinfection makes them challenging to identify prospectively, and their outcomes are extremely poor. In the context of increasing availability of rapid SARS-CoV-2 testing, it is imperative that clinicians remain alert to the possibility of bacterial coinfection and that patients are not denied antibiotics based on a positive SARS-CoV-2 result in the emergency department.

Note

Potential conflicts of interest. The authors: No reported conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest.