Legionella co-infection in a patient with COVID-19

Abstract

We describe a case of an 80-year-old man with COVID-19 and Legionella bacterial co-infection who initially presented to hospital with fever, respiratory symptoms, and diarrhea with radiographic evidence of atypical infection. His initial nasopharyngeal swab was negative; however, a subsequent swab was positive. A Legionella urinary antigen test was positive for Legionella pneumophilia serogroup 1 antigen. Despite a low prevalence of bacterial co-infection in patients with COVID-19, a large number of patients receive antimicrobial therapy. Based on clinical context, a high index of suspicion is warranted for both bacterial and viral infectious processes during the COVID-19 pandemic; this will help to ensure that appropriate antimicrobial therapy is used.

Case Presentation

An 80-year-old man presented to hospital in March 2020 with a 4-week history of non-productive cough, fever, shortness of breath, and diarrhea. His medical history was significant for chronic obstructive pulmonary disease (COPD), coronary artery disease, chronic kidney disease, and significant alcohol use.

He was initially treated as an outpatient for a community-acquired pneumonia with a 10-day course of amoxicillin and 5-day course of prednisone for a presumed COPD exacerbation. Despite therapy, he had progressive respiratory symptoms which prompted a visit to the emergency department 2 weeks after his antibiotic course ended.

In terms of risk factors for COVID-19, he reported no travel history but had been visited by relatives with upper respiratory tract illness symptoms 2 weeks prior to hospitalization. He was also known to frequently leave the house to visit his friends in the community. He did not have any recent exposure to contaminated water or soil but lived in an apartment building. On further review, he also reported loose watery diarrhea, with up to 10 bowel movements per day. This started after the amoxicillin treatment.

On presentation, he was febrile with a temperature of 40.3°C, tachypneic with a respiratory rate of 23 breaths per minute, and an oxygen saturation of 97% was measured on room air. He did not appear to be in distress. Abdomen was soft, non-tender, and nondistended. Respiratory examination did not reveal any adventitious sounds, but a chest X-ray demonstrated increased interstitial markings in the left and right upper lung fields.

Bloodwork demonstrated lymphopenia (0.3 × 10⁹/L [normal 1.0 × 10⁹/L–4.0 × 10⁹/L]), elevated serum concentrations of high-sensitivity troponin-T (117 ng/L [normal ≤14 ng/L]), lactate dehydrogenase (395 U/L [normal ≤225 U/L]), and C-reactive protein (109.8 mg/L, [normal ≤5.0 mg/L]). Electrolytes showed a decreased serum sodium concentration (134 mmol/L [normal 135 mmol/L]) but were otherwise within normal limits.

A SARS-CoV-2 nasopharyngeal swab (NPS) was sent at admission due to his history of recent sick contacts and a high index of suspicion given his lymphopenia, as well as elevated troponin, lactate dehydrogenase and C-reactive protein. C. difficile test was performed due to the history of frequent diarrhea in the setting of recent antibiotics. A Legionella urinary antigen test was sent due to the history of diarrhea with no lobar consolidation on chest imaging and mild hyponatremia.

He was admitted to hospital and initially treated with ceftriaxone and azithromycin therapy for community-acquired pneumonia.

His initial SARS-CoV-2 swab was negative for the E-gene via reverse transcription–polymerase chain reaction (RT-PCR). The Legionella urinary antigen test was positive for Legionella pneumophilia serogroup 1 antigen. Clostridioides difficile toxin gene was positive by molecular method, but toxin was negative by enzyme immunoassay.

For his Legionella infection, he was treated with azithromycin while ceftriaxone was discontinued. He was treated with oral vancomycin therapy for a presumed C. difficile infection.

Four days into admission, he became acutely hypoxic requiring 5 L/min of oxygen therapy via nasal cannula. His therapy was broadened to piperacillin/tazobactam with continued azithromycin intravenous therapy. Due to his ongoing clinical deterioration, a repeat NPS swab for SARS-CoV-2 was sent on day 4 of admission and was positive for the E and S genes.

Following this, he was subsequently switched to 750 mg IV q24h levofloxacin therapy, and the piperacillin/tazobactam was discontinued.

A CT thorax showed ground-glass attenuation in the mid to upper lungs bilaterally consistent with atypical infection on background of moderate to severe emphysema (Figure 1).

Figure 1:

Chest computed tomography scan of an 80-year-old man with peripheral areas of ground-glass predominantly affecting mid to upper lungs bilaterally on the background of emphysema

He received a 10-day course of levofloxacin therapy for his Legionella infection. While his diarrhea was likely multi-factorial secondary to concomitant Legionella, COVID-19, C. difficile infection, and concurrent use of antibiotics, it did initially improve with a 10-day course of oral vancomycin therapy. Our patient ultimately decided to pursue palliative care and died in hospital.

Discussion

We describe a case of a patient presenting with Legionnaires’ disease, COVID-19, and presumed C. difficile infection.

Our case reflects several key points in caring for patients with COVID-19.

First, if there is a high index of suspicion, repeat swabs for SARS-CoV-2 must be performed as PCR assays for SARS-CoV-2 have limitations in their diagnostic accuracy (1). In our patient, even when we had evidence of an atypical infection confirmed by the BinaxNOW Legionella test, which has a specificity of >99%, the progression of respiratory symptoms despite appropriate therapy warranted repeat testing for COVID-19 (2). While nosocomial transmission of COVID-19 was possible in this patient, we think it was unlikely in this case. Our patient remained in droplet and contact precautions throughout the hospitalization. Our suspicion for community transmission of COVID-19 was higher in our patient due to his lack of physical distancing in the community and numerous sick contacts.

In this case, a diagnosis of COVID-19 allowed us to find us an alternate explanation of his respiratory decline, helping us to de-escalate from broad spectrum antibiotic therapy especially important in a patient who was toxin gene positive for C. difficile. As clinicians, these decisions are important in reducing side effects from antibiotics and minimizing development of antimicrobial resistance.

Second, our case reflects that careful consideration must be given in selecting patients who receive empiric antimicrobial therapy in patients with COVID-19. A recent systematic review reported that 8% of patients with COVID-19 were identified as having a bacterial co-infection, while 72% of patients with the virus received antimicrobial therapy (3). As reported previously, COVID-19 infections can present as fever, myalgias, respiratory symptoms, but also with gastrointestinal manifestations such as diarrhea, nausea, and vomiting (4). The combination of respiratory symptoms and gastrointestinal symptoms can mirror Legionnaires’ disease, especially if hyponatremia is present. In our patient, failure to identify a bacterial pathogen that was contributing to his symptoms may have led to treatment escalation in the form of ICU-level care and intubation. Our findings mirror those of a reported case in Japan with COVID-19 and Legionella co-infection, although that patient had travelled to Egypt for a river cruise (5).

Our case demonstrates that bacterial co-infections should be considered in patients with COVID-19 depending on clinical context.

Funding:

No funding was received for this work.

Disclosures:

The authors have nothing to disclose.

Consent:

The authors obtained patient consent for the publication of this article.

Peer Review:

This manuscript has been peer reviewed.