SARS-CoV-2 Infection is a frequent cause of pulmonary failure in 2020, but not the only one – a case report

Carlos Metz Department of Pneumology, Allergology and Critical Care Medicine, Saarland University, Homburg/Saar, Germany Torben Rixecker Department of Pneumology, Allergology and Critical Care Medicine, Saarland University Sebastian Mang Department of Pneumology, Allergology and Critical Care Medicine, Saarland University André Becker Department of Pneumology, Allergology and Critical Care Medicine, Saarland University Alexander Maßmann Department for Diagnostic and Interventional Radiology, Saarland University Medical Center Sören L. Becker Center for Infectious Diseases, Institute of Medical Microbiology and Hygiene, Saarland University Cihan Papan Center for Infectious Diseases, Institute of Medical Microbiology and Hygiene, Saarland University Barbara Gärtner Center for Infectious Diseases, Institute of Medical Microbiology and Hygiene, Saarland University Frederik Seiler Department of Pneumology, Allergology and Critical Care Medicine, Saarland University Guy Danziger Department of Pneumology, Allergology and Critical Care Medicine, Saarland University Robert Bals Department of Pneumology, Allergology and Critical Care Medicine, Saarland University Philipp M. Lepper (  philipp.lepper@uks.eu ) Department of Pneumology, Allergology and Critical Care Medicine, Saarland University


Abstract
Background In 2020, a novel coronavirus caused a global pandemic with a clinical picture termed COVID-19 accounting for numerous cases of ARDS. However, there are still other infectious causes of ARDS that should be considered, especially, as the majority of these pathogens are speci cally treatable.
Case Presentation We present the case of a 36-year-old gentleman who was admitted to the hospital with u-like symptoms after completing a half-marathon one week before admission. As infection with SARS-CoV2 was suspected based on radiologic imaging, the hypoxemic patient was immediately transferred to the ICU, where he developed ARDS. Empiric antimicrobial chemotherapy was initiated, the patient deteriorated further, therapy was changed and the patient was transferred to a tertiary care ARDS center.
As cold agglutinins were present, the hypothesis of an infection with SARS-CoV-2 was then questioned.
Bronchoscopic sampling revealed Mycoplasma (M.) pneumoniae. When antimicrobial chemotherapy was adjusted, the patient recovered quickly.

Case Presentation
On the 14th of March 2020, a 36-year-old gentleman presented at the emergency unit of a nearby hospital with fever, dry cough, head and limb aches that started three days earlier. In good physical condition and otherwise healthy, he had completed a half-marathon the week before, but on admission to another hospital, he presented with reduced general condition and shortness of breath. Physical examination revealed crackles in the right upper lobe upon auscultation. Medical, family, and psychosocial history was completely unremarkable.
As SARS-CoV-2 was suspected, he was immediately admitted to the ICU. He received microbiological and virological sampling as well as chest X-ray and thoracic CT. A calculated antimicrobial chemotherapy with piperacillin/tazobactam (3 x 4.5 g/d i.v.) and clarithromycin (2 x 500 mg/d i.v.) was started.
Polymerase chain reaction (PCR) for SARS-CoV-2 was negative, as were the results for in uenza, Mycobacterium tuberculosis, and Legionella spp. After 2 days, the antimicrobial regimen was changed to meropenem (3 x 1.0 g/d i.v. ), linezolid (2 x 600 mg/d i.v.) and fosfomycin (3 g/d i.v.) due to persistently elevated in ammatory parameters and further clinical deterioration. However, the assumption of the patient having COVID-19 was maintained. Imaging of the lungs revealed a diffuse interstitial reticular pattern, multilobular patchy ground-glass opaci cation and consolidation of the right upper lobe.
After a week on high-ow oxygen, he deteriorated and was intubated. Despite proning, he deteriorated further, needing higher doses of vasopressors (0.2 µg/kg/min), and our hospital's extracorporeal membrane oxygenation (ECMO) team was called on the 22 nd of March 2020 to transfer the patient. The patient was reported as having COVID-19 and moderate CARDS with respiratory acidosis.
The ECMO team set out to transfer the patient under COVID-19 personal protective equipment (PPE). In the external hospital, after reviewing the laboratory and imaging ndings, as well as the ventilator settings, the team leader decided against implanting an ECMO on site, and the patient was transferred under COVID-19 precautionary measures. After an otherwise unremarkable transport, the patient had to be re-intubated, as the cuff was leaking. On that occasion, bronchoscopy with bronchoalveolar lavage (BAL) was performed for microbiological sampling and to further elucidate the hypothesis of an infection with SARS-CoV-2. Additionally, a naso-and oropharyngeal swab for SARS-CoV-2 was obtained.
As the lung's compliance was preserved, we accepted a P insp of 30 cmH 2 O at a PEEP of 10 cmH 2 O, with a respiratory rate of 22/min. The patient was slightly hypercapnic at a pH of 7.39. The oxygenation index on admission at our unit was 115.
Chest radiographs were reviewed, and a new chest X-ray was made ( Figure 1). Imaging showed mainly right-sided pneumonia, not typical for COVID-19, and the hypothesis of COVID-19 was abandoned.
Antimicrobial therapy was changed, as atypical pneumonia was suspected. Meropenem (then given continuously i.v., monitored by determination of serum levels) was continued for a total of 7 days, and clarithromycin (2 x 500 mg/d) was added to the antimicrobial regime again (Table 1). Remarkably, for a young and otherwise healthy individual, the patient had elevated bilirubin and LDH levels, with diminished haptoglobin and macrocytic hyperregenerative anemia (hemoglobin 5.3 g/dL at a s cv O 2 of 67%) on admission, demonstrating hemolysis. Coombs' test revealed cold agglutinins the same day.
Microbiological results from a BAL on the 22 nd of March were negative for SARS-CoV-2, Pneumocystis jirovecii, Bordetella pertussis and B. parapertussis, Chlamydophila pneumoniae, Haemophilus in uenzae, Legionella pneumophila, Moraxella catarrhalis and Streptococcus pneumoniae; however, the patient tested positive for Mycoplasma (M.) pneumoniae. Thus, a diagnosis of mycoplasma-related ARDS with cold agglutinin disease was made. The patient was in total substituted with 4 units of packed red blood cells and received only warm infusions using a Level One infusion system (Smiths Medical, Minneapolis, Minnesota, USA). Ventilator support was deescalated soon after change of the antimicrobial regimen, and the patient was extubated on the 25 th of March and received noninvasive ventilator support for 4 days. Meropenem (then given continuously i.v., monitored by determination of serum levels) was continued for a total of 7 days, and clarithromycin (2 x 500 mg/d) was added to the antimicrobial regime again (Table 1). Hemolysis improved quickly under the antibiotic regime. The patient was discharged from the ICU on March 30 and was discharged home on April 6. He had no health-related complaints in a telephone interview conducted on June 18 th .

Discussion And Conclusion
The case is remarkable for several reasons. Under the impression of a seemingly predominant microorganism at the time of admission, a clinical diagnosis was established that was questionable according to radiologic evidence. Additionally, inappropriate management of the pulmonary infection might have favored clinical deterioration. Initial antimicrobial therapy was changed early empirically to a regimen that was much less effective for the causative microorganism, despite a microbiological workup.
The initial therapy was performed according to current guidelines 4 . The change was meant to extend the spectrum; instead, it missed the causative organism. Furthermore, at the time antimicrobial therapy was changed, treatment failure was not proven.
M. pneumoniae is a common cause of community acquired pneumonia, particularly in children and young adults 5 . M. pneumoniae is a very small bacterium without a peptido-glycan cell wall. It is a common cause of tracheobronchitis and atypical pneumonia mainly because of its adherence to respiratory cells. Infection of host cells occurs through special adhesins and an elongated polar attachment organelle 6 . Usually, the pneumonia caused by M. pneumoniae is mild and characterized by a dry cough or self-limiting pneumonia 7 . The rates of ICU admission ranged between 10% and 16%. With an intense epidemiological background of COVID-19 as a cause of respiratory disease leading to a surge of ICU admissions in many countries, the proper diagnosis of treatable causes for ARDS is highly important. Severe ARDS and fatal outcomes due to M. pneumoniae are rare and may be the result of unclear clinical features, delayed diagnosis, inappropriate respiratory support and/or insu cient initial treatment. If additional diagnostic measures are not con rming the suspected pathogen, alternative explanations need to be evaluated. This is especially important if a treatable cause is present, as in the case presented here. However, concomitant cold agglutinin disease (CAD) is frequently described in the context of M. pneumoniae and usually develops upon generation of polyclonal IgM antibodies directed against I antigens on RBCs. Hemolysis can be severe but is usually self-limited, while corticosteroids are reported to be barely effective 8 .
Antibiotic therapy of M. pneumoniae requires agents such as macrolides or uoroquinolones that do not target the bacterial cell wall and have good intracellular penetration. In our patient, macrolide therapy was started according to guidelines for severe pneumonia but was stopped after 2 days, and the regimen was unintentionally changed to a less effective one. These decisions might have been driven by the assumption the patient might have an infection with SARS-CoV-2.
The gold standard for the detection of COVID-19 in symptomatic individuals is the detection of viral RNA in naso-or oropharyngeal swabs by reverse-transcriptase polymerase chain reaction (rtPCR). It is possible, however, that RT-PCR following a naso-or oropharyngeal specimen fails to identify infected persons 9 . This is mainly due to the quality of the swabs and to the fact that the virus does not replicate in the upper respiratory tract in all patients at all time points. Especially during the later course, it often replicates in the lower respiratory tract. Thus, samples from the lower respiratory tract should be analyzed as well.
Radiologic ndings alone are often not reliable for differentiating pneumonia. Moreover, coinfection with other bronchopulmonary pathogens is not uncommon. In general, radiographic ndings should be used along with clinical and laboratory data to narrow the differential diagnosis. Currently, there is a threat of misinterpreting clinical pictures and lung imaging as SARS-CoV2-induced disease. It has been suggested only recently that low-dose CT might be of equal sensitivity and speci city as rtPCR testing of nasopharyngeal swabs 10 . However, multiple infectious diseases might produce similar pictures in pulmonary imaging modalities. Bronchopulmonary manifestations of Mycoplasma pneumoniae may vary with uni-, multi-, or bilateral in ltration. Some patients may present with additional pleural effusion and hilar lymphadenopathy.
At the time, the patient in the present case acquired the infection leading to ARDS, and the prevalence of COVID-19 in Germany, especially in the Federal State of Saarland, was low (2,078 con rmed cases in Germany on March 12 ). In contrast, awareness of COVID-19 was high-ying.
If the prevalence of a speci c infectious agent is predominant, it is very likely that the radiologic picture in fact happens to be the at the time frequently encountered infectious agent. However, the effective performance of CT for COVID-19 detection critically depends on the pretest probability for the occurrence of a disease, which in turn in uences positive and negative predictive values (PPV and NPV). If the prevalence of a disease is truly low, the PPV for the disease will be low. If caregivers overestimate PPV, they might come to the wrong conclusion if no gold standard for the diagnosis of a disease exists or is accepted.
In the present case, treating physicians unintentionally created an unfavorable situation for the patient urged by erroneous assumptions. Unfortunately, the well-intentioned putative escalation of antimicrobial therapy was less effective for the causative organism.
The unfortunate initially inappropriate management has led to a spiral of further unfortunate decisions. The case presented here thus underlines the importance of adhering to established treatment guidelines, scrutinizing treatment modalities and not forgetting other potential causes of severe pneumonia or ARDS to ensure that critically ill patients are safeguarded from common infections even in times ruled by a predominant pathogen.

Consent for Publication
Written informed consent was obtained from the patient for publication of this case report and any accompanying images. A copy of the written consent is available for review by the Editor-in-Chief of this journal.

Availability of data and materials
Data can be provided on request addressed to the corresponding author. All data sharing statements are subject to conformity with German data protection legislation and rules (Datenschutzgrundverordnung -DGSVO).  1A: Chest radiograph on the day of admission (March 14th, 2020) to an external hospital shows consolidation predominantly in the right upper lobe, vague ill-de ned opacities in the right lower lobe and left hilar region and a diffuse interstitial pattern combined with bronchial wall thickening. 1B: Chest X-ray on the day of admission to a tertiary care hospital depicts progressive pneumonia characterized by diffuse reticular and nodular patterns (March 22). 1C: Chest X-ray shortly after discharge from the ICU (April 2) shows almost complete regression of previous in ltrations. The patient did not need supplementary oxygen at that time. 1D&E: Computed tomography of the chest on March 16th con rms consolidation of the right upper lobe and reveals multifocal, patchy consolidations, ill-de ned airspace in ltrates and ground-glass opaci cations. Additional centrilobular nodular appearance and thickening of the bronchovascular structures is present.