key: cord-0683561-bwc6pdl5
authors: Kalpaxi, Angeliki; Kalokairinou, Mariana; Katseli, Paraskevi; Savvopoulou, Vasiliki; Ioannidi, Pinelopi; Triantafyllou, Evangelia; Flokatoula, Maria; Pythara, Chrystalla; Papaevangelou, Angeliki
title: Spontaneous pneumomediastinum and COVID-19 pneumonia: report of three cases with emphasis on CT imaging
date: 2021-06-23
journal: Radiol Case Rep
DOI: 10.1016/j.radcr.2021.06.040
sha: 9b3d4175a19fc70e922931899d7a7839e61b32d8
doc_id: 683561
cord_uid: bwc6pdl5

Spontaneous pneumomediastinum (SPM) is a rare complication of coronavirus disease 2019 (COVID-19). The published literature consists mainly of case reports and small case series. There are still many questions regarding the pathogenesis, the prognostic significance and the implications on patient management. In our hospital three COVID-19 patients developed SPM: one on admission at the emergency department and the other two during hospitalization. In this study we describe their clinical course and computed tomography (CT) findings. All of them had severe disease according to the total severity score on admission CT. The management of pneumomediastinum was conservative and follow-up CT showed resolution in all patients. As the correlation between extension of parenchymal lung lesions and development of pneumomediastinum is still under investigation, we highlight the importance of reporting the severity score on chest CT in order to obtain more comparable results between different studies. Furthermore, in this tragic circumstance we also had the opportunity to familiarize ourselves with the otherwise uncommon occurrence of air along the bronchovascular sheaths (Macklin effect) and evaluate the ability of CT to detect it.

Since the onset of the worldwide coronavirus pandemic new data that enrich our knowledge and comprehension of the disease appear daily in international literature. Radiological imaging plays a crucial role in evaluating the course of COVID -19 infection and in choosing proper management of infected patients. The imaging characteristics of COVID-19 pneumonia depend on the stage of the disease. The most frequent and early presentation consists of bilateral multilobar ground glass opacities (GGO) with a peripheral and posterior distribution [1] . Septal thickening, consolidation superimposed on GGO, bronchiectasis and fibrotic-like lesions are some of the imaging findings that occur over time. A comparatively unusual complication that has been noted in recent reports [2, 3, 4] is the development of pneumomediastinum in non-intubated COVID -19 patients.

Spontaneous pneumomediastinum is defined as the presence of free air in the mediastinal cavity that is not associated to trauma or iatrogenic causes, such as intubation, endoscopic procedures or central line placement [5] . Predisposing risk factors of SPM are a positive history of smoking, asthma, interstitial lung disease and a recent history of respiratory infection. SPM in COVID-19 patients is isolated in some cases, while in others it is combined with pneumothorax, pneumopericardium or subcutaneous emphysema. Given the infrequent occurrence of the complication, the published literature consists mainly of case reports and case series.

The pathophysiological mechanism, which was first described by Macklin et al. [6] and is, therefore known as Macklin effect, refers to the presence of a pressure gradient between the alveoli and the lung interstitium resulting in alveolar rupture, followed by air dissection centripetally through the pulmonary interstitium along the bronchovascular sheaths toward the pulmonary hila and the mediastinum. Although the exact mechanism of alveolar rupture in COVID-19 pneumonia is unknown, it has been suggested that it may be linked to a combination of inflammation and diffuse alveolar damage caused by the SARS-COV2 virus and increased alveolar pressure due to pronounced cough [7] .

The clinical significance of the radiologic finding in terms of prognosis and patient management is still under investigation. There are many questions yet to be answered; thereby it is important that every institution reports back all experiences of this rare complication and in as much detail as possible.

Since the outbreak of COVID-19 three cases of SPM have been diagnosed in our radiology department. We subsequently followed up these patients and collected all clinical and imaging data relevant to the condition.

A 49-year-old man with a medical history of obesity was referred to the emergency department complaining of fever associated with chest pain and dry cough. The patient was diagnosed with COVID-19 on the basis of RT-PCR analysis performed 3 days before. No medical history of smoking, emphysema or chronic lung disease was reported. His vital signs were as follows: temperature of 37.8°C, blood pressure of 123/85 mmHg and oxygen saturation of 79% on room air. The patient was immediately placed on supplemental oxygen via Venturi mask (maximum oxygen concentration). Laboratory studies were remarkable for normal white blood cells count of 6.900 /μL with a neutrophil count of 85.7%, elevated C-reactive protein (CRP) at 255mg/L and elevated soluble urokinase plasminogen activator receptor biomarker (suPAR) at 16ng/mL. His D-dimer was 1,51mg/L. Chest CT ( Figure 1A -B) revealed extensive bilateral ground glass lesions. According to the method proposed by Li et al. [8] , the total severity score (TSS) was 19/20. Unexpectedly, extraluminal gas was observed in the mediastinum anterior to the pulmonary artery and aortic arch and in the left hila (perihilar Macklin effect).

During the hospital course his hypoxia worsened requiring transition to increasing levels of supplemental oxygen to reach the level of 60L/min by high flow nasal cannula (HFNC) to maintain his oxygenation.

Two more chest CT examinations were requested on hospital days 6 and 13, due to episodes of fever and raised inflammatory markers. They showed resolution of previous pneumomediastinum (figure 1C, 1D) and gradual progression of ground glass opacities into denser consolidations and more fibrotic lesions.

His antibiotic treatment was adjusted and a gradual clinical improvement with reducing oxygen requirements was observed. Eventually, 22 days after admission, in stable condition, he was discharged home.

A 66-year-old man with a history of hypertension, dyslipidemia and obesity presented to the emergency department with a three-day history of fever, dyspnea, dry cough and a positive rapid test for COVID-19. Smoking history was also reported. Vital signs were temperature of 36.8°C, blood pressure of 110/60 mmHg and oxygen saturation of 90% on room air. Laboratory tests revealed normal white blood cells count (4.900 /μL) with an elevated neutrophilic count (85,4%), elevated CRP (144mg/L) and suPAR biomarker (6,20ng/mL). The patient was placed on supplemental oxygen via Venturi mask (oxygen concentration 50%). Chest CT on admission revealed extensive ground glass opacities involving all lobes, with a TSS of 9.

The patient was placed on antibacterial, antiviral, corticosteroid treatment and thromboprophylaxis with LMWH. During hospitalization the patient started to decline clinically, presenting increasing oxygen requirements. He was eventually placed on a high flow nasal cannula (HFNC, FiO2: 91%, 60L/min). A modification of his antibiotic treatment was also necessary due to raised inflammatory markers.

On hospital day 10 he had a CT pulmonary angiography (CTPA) because of clinical deterioration and suspicion of pulmonary embolism ( Figure 2 ). Small filling defects in right upper lobe subsegmental arteries were noted. Surprisingly, gas bubbles were revealed in the anterior compartment of the mediastinum and in the left hila (perihilar Macklin effect). A more careful examination of the left lung parenchyma permitted the identification of linear air collections along branches of the left upper pulmonary vein (peripheral Macklin effect). Previous ground glass lesions were accompanied by thickened interlobular septa and were more extensive in comparison to the admission CT.

The patient received Fondaparinoux for the treatment of pulmonary embolism. Conservative treatment was advised for pneumomediastinum.

His respiratory status continued to worsen with increasing oxygen requirements and ultimately necessitating endotracheal intubation. Transfer to the intensive care unit (ICU) was decided. We would like to highlight the fact that his complicated clinical course was mainly due to extensive lung disease, septic episodes and bleeding diathesis and was unrelated to the pneumomediastinum. In fact, on hospital day 30 a chest CT was obtained that showed its resolution. At the moment of writing the patient remains in the ICU.

A 44-year-old man was admitted to the emergency department complaining of a five-day history of fever, dyspnea and dry cough. He had tested positive for COVID-19 eight days before presentation via a rapid test. No medical history was reported. His vital signs were temperature of 36.2°C, blood pressure of 163/102 mmHg, and oxygen saturation of 84% on room air. The patient was initially placed on supplemental oxygen via nasal cannula (3 L/min) and then on a Venturi mask (40%). His laboratoty tests were remarkable for slightly elevated white blood cell count of 11.900 /μL with a neutrophil count of 90.6%, elevated CPR at 181mg/L and elevated suPAR at 7.6ng/mL His Ddimer was 0.86mg/L.

The patient was placed on antibacterial, antiviral, corticosteroid treatment and thromboprophylaxis with LMWH. During hospitalization his oxygen requirements increased and he was placed on high flow nasal cannula (HFNC, FiO2: 89%, 60L/min) to maintain his oxygenation. A modification of his antibiotic treatment was also decided.

On hospital day 8 a CTPA was performed because of a clinical suspicion of pulmonary embolism. In fact, CT confirmed the presence of small bilateral contrast filling defects of segmental and subsegmental arteries. Additionally, it revealed the presence of pneumomediastinum, pneumopericardium and subcutaneous emphysema. Perihilar and peripheral Macklin effect was evident ( Figure 3D ).

The patient received Fondaparinoux for the treatment of pulmonary embolism. In the following days there was an improvement in the respiratory condition and general status of the patient. Chest CT on hospital day 13 showed minimal gas persistence in the anterior mediastinum. Peripheral Macklin effect was not evident anymore ( Figure 3E ).

The patient was successfully discharged home 18 days after admission, in good general condition.

Before the COVID-19 outbreak SPM was an extremely rare condition, diagnosed in 1/44,500 of accident and emergency attendances [9] . In the context of viral infections, some sporadic cases had been published, mainly associated with influenza virus [10] . The SARS infection is a remarkable exception. Indeed, the reported frequency of SPM development in two different SARS studies was 12 % [11, 12] . In regard to COVID-19, the frequency seems to be much lower, but still notable. 3 cases of SPM out of 720 patients (frequency 0.42%) have been identified in our radiological department since November 2020, but no cases were noted earlier, during the first wave of the pandemic. These observations are in accordance with the results of larger studies [13, 14, 15] . In fact, in their study of 4081 hospitalized COVID-19 patients Brito et al. [13] report a frequency of 0.5 % and in their study of 1648 patients Cut et al. [14] report a frequency of 0.66%. Moreover, Palumbo et al. [15] point out the significantly higher rate of the complication during the second wave of the pandemic.

The risk of development of SPM in COVID-19 patients has been associated with mechanical and non-invasive ventilation (16, 17, 18 ). In addition, similarly to our patients 2 and 3, there are some reported cases of SPM in COVID-19 patients managed with HFNC [3, 19, 20, 21, 22, 23, 24, 25] , whose contribution to the occurrence of SPM is still unclear. The aforesaid observations may confound the role that COVID-19 plays in the development of SPM. Patient 1 represents one of the few cases in the published literature of SPM diagnosed on presentation at the emergency department, at the beginning of oxygen support therapy [7, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35] . These cases are of great importance because they suggest that there must be an inherent component in the COVID-19 infection that predisposes to air leaks, which can precipitate with anything that increases the alveolar pressure [24] .

Due to the vague nature of presenting symptoms [5] , imaging is essential to make a diagnosis of pneumomediastinum and should be considered in all COVID-19 patients with unexplained clinical deterioration (36, 37) . Chest X-ray is the first imaging modality employed and often suggests the diagnosis by showing linear or curvilinear lucencies outlining mediastinal contours. CT is more sensitive in the detection of air in the mediastinum and is very helpful in establishing or confirming the diagnosis. It may also provide extra diagnostic information regarding the exclusion of coexisting illness, such as perforated esophagus. All that certainly comes at the cost of increased radiation dose.

An additional advantage that CT offers is that it provides the mean to reveal the Macklin effect. It appears on thoracic CT as linear collections of air contiguous to the bronchovascular sheaths [38] . The main reason to communicate its occurrence is that it offers a confirmation of the proposed mechanism of SPM formation in COVID-19 patients and increases confidence that SPM is due to respiratory and not other causes of pneumomediastinum. We detected the CT Macklin effect in the perihilar area of all our patients and in the peripheral area of 2 out of 3 patients. A finding that caught our attention was that the peripheral Macklin effect was located on the left in all of our patients. The CT Macklin effect has been mentioned in few other reports on COVID-19 patients [13, 29, 39] .

The prognostic significance of the development of SPM is still under investigation. A hypothesis is that, if indeed there is a correlation between the degree of alveolar damage and the presence of pneumomediastinum, then it would be more likely to occur in patients with an advanced stage of the disease with extensive pulmonary lesions and correlate with a worse outcome [28, 40] . Actually, in the larger published case series [13, 14, 41, 42] most patients with SPM had extensive lung lesions. This also applies to our study, in which all of our 3 patients that developed SPM had extensive disease with high TSS of the admission CT. However, SPM has also been described to occur in the absence of extensive lung involvement, with a favorable clinical course and recovery of the patients in the majority of these cases [28] . In addition, we would like to highlight the importance of using a CXR or CT scoring system to assess the severity of lung involvement in COVID-19 patients in order to obtain more comparable results between different studies.

Once the diagnosis of pneumomediastinum has been made, the radiologist should communicate the finding to the clinical doctor as soon as possible. Although spontaneous pneumomediastinum is usually a self-limiting condition, it may require some changes in the management of the respiratory support in order to maintain the balance of adequate oxygenation while preventing pneumomediastinum expansion and pneumomediastinum-related cardiovascular and respiratory complications [2, 20] . Serial imaging is usually required.

It is important to be alert to the diagnosis of pneumomediastinum in COVID-19 patients even in the absence of mechanical ventilation. Apart from permitting early recognition, CT imaging also provides significant information, such as the severity score and the revelation of the Macklin effect. It is suggested that this information is incorporated in published studies on this topic. 

The authors declare that written informed consent has been obtained from all the patients of this study.

Coronavirus Disease 2019 (COVID-19): A Systematic Review of Imaging Findings

COVID-19 with spontaneous pneumomediastinum

Mediastinal Emphysema, Giant Bulla, and Pneumothorax Developed during the Course of COVID-19 Pneumonia

Spontaneous Pneumomediastinum: A Probable Unusual Complication of Coronavirus Disease

Spontaneous pneumomediastinum: time for consensus

Malignant interstitial emphysema of the lungs and mediastinum as an important occult complication in many respiratory diseases and other conditions: interpretation of the clinical literature in the light of laboratory experiment

Spontaneous pneumomediastinum in COVID-19

CT image visual quantitative evaluation and clinical classification of coronavirus disease (COVID-19)

Spontaneous pneumomediastinum in H1N1 infection: uncommon complication of a common infection

Spontaneous pneumomediastinum in patients with severe acute respiratory syndrome

Clinical progression and viral load in a community outbreak of coronavirus-associated SARS pneumonia: a prospective study

Pneumomediastinum in COVID-19 disease: Outcomes and relation to the Macklin effect

Spontaneous Pneumomediastinum, Pneumothorax, Pneumopericardium and Subcutaneous Emphysema-Not So Uncommon Complications in Patients with COVID-19 Pulmonary Infection-A Series of Cases

Pneumothorax/pneumomediastinum in non-intubated COVID-19 patients: Differences between first and second Italian pandemic wave

Pulmonary Barotrauma in COVID-19 Patients: Invasive versus Noninvasive Positive Pressure Ventilation

Pulmonary Barotrauma in COVID-19 Patients With ARDS on Invasive and Non-Invasive Positive Pressure Ventilation

Massive emphysema subcutis, pneumothorax, pneumomediastinum and pneumoperitoneum as uncommon complication of covid-19 pneumonia, a rare case

Coronavirus disease 2019 (COVID-19) complicated by Spontaneous Pneumomediastinum and Pneumothorax

Spontaneous pneumomediastinum, pneumopericardium, pneumothorax and subcutaneous emphysema in patients with COVID-19 pneumonia, a case report

Two cases of spontaneous pneumomediastinum with pneumothorax in patients with COVID-19 associated pneumonia

Pneumothorax and Pneumomediastinum Secondary to COVID-19 Disease Unrelated to Mechanical Ventilation. Case Rep Crit Care

COVID-19 and Spontaneous Pneumomediastinum: A case series

Spontaneous pneumomediastinum, pneumothorax and subcutaneous emphysema in COVID-19 pneumonia: a rare case and literature review

Spontaneous subcutaneous emphysema and pneumomediastinum in non-intubated patients with COVID-19

COVID-19-associated pneumomediastinum

Pneumomediastinum Related to Covid-19 Pneumonia

Spontaneous pneumomediastinum occurring in the SARS-COV-2 infection

Unusual presentations of COVID-19 pneumonia on CT scans with spontaneous pneumomediastinum and loculated pneumothorax: A report of two cases and a review of the literature

Case Report: Pneumothorax and Pneumomediastinum as Uncommon Complications of COVID-19 Pneumonia-Literature Review

Spontaneous pneumomediastinum as the only CT finding in an asymptomatic adolescent positive for COVID-19. BJR Case Rep

COVID-19 and spontaneous pneumomediastinum: a rare complication

Spontaneous pneumomediastinum, pneumopericardium, pneumothorax, and subcutaneous emphysema in a patient with COVID-19

Pneumomediastinum as a complication of COVID-19 disease: A case report. COVID-19 hastalığının bir komplikasyonu olarak pnömomediastinum: Bir olgu sunumu

Typical Covid-19 case with primary pneumomediastinum in a 37 year old male. Radiol Case Rep

Broadening the differential: pneumomediastinum and COVID-19 infection

Spontaneous pneumothorax and pneumomediastinum as a rare complication of COVID-19 pneumonia: Report of 6 cases

Spontaneous pneumomediastinum and Macklin effect: Overview and appearance on computed tomography

Spontaneous Pneumomediastinum and Macklin Effect in COVID-19 Patients

Spontaneous pneumomediastinum as a complication of a COVID-19 related pneumonia: case report and review of literature

Is Spontaneous Pneumomediastinum a Poor Prognostic Factor in Covid-19?

Spontaneous Pneumomediastinum/Pneumothorax in Patients With COVID-19

Declaration of interests

☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work