Incidence and Outcome of Spontaneous Alveolar Air Leak Events in COVID-19 Pneumonia: A Prospective Cohort Study

1. DM Fellow Department of Pulmonary and Critical Care Medicine, PGIMS, Rohtak, Haryana, India. 2. Associate Professor, Department of Respiratory Medicine, Bharati Vidyapeeth (DTU) Medical College, Pune, Maharashtra, India. 3. Senior Resident, Department of Respiratory Medicine, Bharati Vidyapeeth (DTU) Medical College, Pune, Maharashtra, India. 4. DM Fellow, Department of Critical Care Medicine, Bharati Vidyapeeth (DTU) Medical College, Pune, Maharashtra, India. 5. Consultant, Department of Respiratory Medicine, Ravi Hospital, Hyderabad, Andhra Pradesh, India. 6. Consultant, Department of Respiratory Medicine, Nobel Hospital, Pune, Maharashtra, India. 7. DM Fellow, Department of Critical Care Medicine, Bharati Vidyapeeth (DTU) Medical College, Pune, Maharashtra, India; Assistant Professor Medicine, Rangraya.

Abstract

Introduction: Coronavirus Disease 2019 (COVID-19) has increased the burden of hospitalised pneumonia cases and related complications. Spontaneous Pneumothorax (PT) and Pneumomediastinum (PM) have been reported in both spontaneously breathing and ventilated patients with COVID-19 pneumonia.

Aim: To determine the incidence and outcomes of spontaneous alveolar air leak events in COVID-19 pneumonia.

Materials and Methods: This prospective cohort study was carried out from June 2020 to June 2021 at a tertiary care centre in Western India. All incident cases of alveolar air leaks in COVID-19 pneumonia were included. Clinical and demographic data were collected, and statistical analysis was performed. The Chi-square test or Fisher’s exact test were used to assess the differences in subgroup proportions.

Results: A total of 79 patients (63 males and 16 females) experienced spontaneous alveolar air leaks in the form of PT, PM (mediastinal emphysema), or Subcutaneous Emphysema (SE), either isolated or in combination. A total of 58 patients (73.41%) had PT, while 8 patients (10.12%) had isolated PM and 2 patients (2.53%) had isolated SE. Of the total events, 35 (44.30%) occurred in spontaneously breathing patients, among them vigorous coughing was an important precipitating factor. At the time of the incident, 1.27%, 21.52%, and 77.21% of the affected cases belonged to mild, moderate, and severe COVID-19 categories, respectively. Male patients (n-63, 79.74%) in the age group of 30-60 years were predominantly affected. A total of 38 events (48.10%) occurred within two weeks (early) of symptom onset. The PaO2:FiO2 ratio at the time of the alveolar leak showed a significant association with the outcome. Patients with PT had a poorer outcome compared to those with other types of alveolar leaks (p-value<0.005). Major bleeding occurred in 2 (3.33%) of the total 60 Intercostal Drainage (ICD) procedures. Prolonged alveolo-pleural fistula healed spontaneously in four out of five cases. The cumulative incidence for air leak events was 1.55%, and for barotrauma, it was 6.47%. The overall mortality in this cohort was 74.68% (n=59), while it was 29.41% (5 out of 17) in the moderate severity group. Patients with late-onset events had a better outcome (p-value<0.005).

Conclusion: In this cohort of COVID-19 pneumonia from Western India, the cumulative incidence of spontaneous alveolar air leaks was 1.55%, predominantly affecting males. The early occurrence of PT in severely hypoxic patients on mechanical ventilator was associated with higher mortality.

Barotrauma, Mediastinal emphysema, Pneumothorax, Severe acute respiratory syndrome coronavirus 2

The COVID-19 pandemic, caused by Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2), posed multiple challenges to the healthcare system. Severe cases of COVID-19 were complicated by Acute Respiratory Distress Syndrome (ARDS), Multiorgan Dysfunction Syndrome (MODS), pulmonary embolism, myocarditis, secondary bacterial or fungal infections, and post-COVID-19 lung sequelae. Complications related to spontaneous alveolar air leaks, in the form of PT and PM, were also reported in patients with SARS-CoV-2 pneumonia (1),(2),(3). The term spontaneous PT refers to the presence of air in the pleural space that is not caused by trauma or another obvious precipitating factor (4). While primary spontaneous PT occurs without a clinically apparent lung condition, secondary spontaneous PT is a complication of pre-existing lung diseases, such as bronchial asthma, Chronic Obstructive Pulmonary Disease (COPD), cystic lung diseases, and some lung infections like tuberculosis or Pneumocystis jirovecii pneumonia (4). PM and SE refer to the presence of air in the mediastinum and subcutaneous tissue, respectively (5).

In COVID-19 patients, spontaneous air leak events have been reported with or without Invasive Positive Pressure Ventilation (IPPV) (1),(2),(3). There is limited Indian data on the incidence of spontaneous alveolar air leak events in COVID-19 pneumonia and the factors affecting their outcomes [6-8]. Thus, this study was conducted to estimate the incidence of PT, PM, and SE in COVID-19 pneumonia and their outcomes in a tertiary care centre in western India, which was highly affected during the COVID-19 pandemic.

Material and Methods

This prospective cohort study was carried out at a tertiary care hospital in Pune, Maharashtra, India, from June 2020 to June 2021, covering two major COVID-19 waves in western India. Institutional Ethical Committee (IEC) approval was obtained (IEC number-BVDUMC/IEC/23).
Inclusion criteria: All consecutive COVID-19 positive cases (as confirmed by Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR) or Rapid Antigen Testing (RAT)) with a clinical and/or radiological diagnosis of spontaneous PT, PM, or SE, either isolated or in combination, were included. These cases were grouped together as spontaneous alveolar air leak events in COVID-19 pneumonia.

Exclusion criteria: PT secondary to iatrogenic injuries or trauma was excluded from the study.

Sample size: This was an exploratory study on COVID-19, and all incident cases over the study period were included. Informed consent was obtained from the patient or their immediate relative.

Demographic data, including age, gender, co-morbidities, and clinicoradiological characteristics, were documented. The time of occurrence of the alveolar air leak event, clinical symptoms, ventilator parameters, requirement for Intercostal Drain (ICD), and associated complications were recorded.

Case definitions: COVID-19 positive: A nasopharyngeal swab for RT-PCR was used to establish the diagnosis of COVID-19.

Severity grading of COVID-19 disease (9):

- Asymptomatic/Presymptomatic infection: Individuals who test positive for SARS-CoV-2 but show no symptoms consistent with COVID-19.
- Mild illness: Individuals who exhibit any signs and symptoms of COVID-19 but do not have shortness of breath or have abnormal chest imaging.
- Moderate illness: Individuals who show evidence of lower respiratory disease during clinical assessment or imaging and have an oxygen saturation (SpO₂) ≥94% on room air.
- Severe illness: Individuals with SpO2 <94% on room air, a ratio of arterial partial pressure of oxygen to fraction of inspired oxygen (PaO₂/FiO₂) <300 mm Hg, a respiratory rate >30 breaths/min, or lung infiltrates >50%.
- Critical illness: Individuals who have respiratory failure, septic shock, and/or multiple organ dysfunction.

Cytokine storm: The working diagnosis of cytokine storm in COVID-19 was based on supporting laboratory parameters, including C-Reactive Protein (CRP) >50 mg/L, Ferritin >700 μg/L, D-Dimer >0.5 mg/L, and Interleukin-6 >7 pg/mL (10).

Place of event: Pneumonia cases requiring less than 15 litres of oxygen on non rebreather masks with stable haemodynamics were managed in COVID wards, while critical cases were treated in the COVID-19 Intensive Care Unit (ICU).

Statistical Analysis

The collected data was coded and entered into a Microsoft Excel sheet. The data were analysed using Statistical Package for the Social Sciences (SPSS) version 20.0 software. The results were presented in tabular and graphical formats. For qualitative data, various rates, ratios, and percentages (%) were calculated. For quantitative data, the mean was calculated. The Chi-square test or Fisher’s exact test was used to assess the differences in subgroup proportions. A p-value of <0.05 was considered statistically significant.

Results

Total 5,101 SARS-CoV-2 positive patients were admitted during the study period. The study protocol for case inclusion is shown in (Table/Fig 1). A total of 79 patients were included in the final analysis. A total of 58 (73.42%) patients suffered from spontaneous PT, with 39 (49.37%) had isolated PT, while others presented with either PM or SE.

Incidence of Alveolar Air Leak: In this study, the cumulative incidence of alveolar air leaks in COVID-19 positive cases was 1.55% (79 out of a total of 5,101). Barotrauma occurred in 6.47% (44 out of 680 patients on invasive ventilation). The incidence of PT was 1.14% (58 out of 5,101) among all hospitalised cases of COVID-19 pneumonia.

A total of 63 (79.75%) male and 16 (20.25%) female COVID-19 positive cases with alveolar air leaks were included in this cohort. Age groups from 41 to 60 years were more affected. Diabetes and hypertension were common co-morbidities in the study population. A total of 61 (77.21%) cases had severe COVID-19 pneumonia at the time of the alveolar leak (Table/Fig 2). Total 34 (58.62%) cases had right-sided PT, while 8 (13.79%) had bilateral PT (Table/Fig 3). Retrosternal chest pain was a predominant symptom in 14 (46.67%) of the total 30 patients with PM (mediastinal emphysema). Clinical examination revealed SE in the form of palpable crepitus and swelling of the neck and chest wall.

(Table/Fig 4) gives distribution of cases according to the place of event and the pattern of alveolar leaks. Five patients had PT or PM at the time of hospitalisation. The other cases are divided (inpatient ward/ICU) as per COVID-19 severity. Out of a total of 79 patients, 35 (44.30%) were spontaneously breathing at the time of the air leak event, while 44 (55.70%) had barotrauma. All patients with severe pneumonia who were on a mechanical ventilator (IPPV) died, with the exception of two -one with PM and the other with PM+SE.

Early versus late events: In 38 (48.10%) cases, alveolar air leaks occurred within two weeks of symptom onset. Radiologically, these patients had severe disease with bilateral patchy Ground-Glass Opacities (GGO) and extensive consolidation. Interstitial emphysema was noted on HRCT in 2 (2.53%) cases (Table/Fig 5)a. Patients with late-onset alveolar air leak events had subpleural blebs, parenchymal cysts, or cavitations contributing to the risk of alveolar rupture (Table/Fig 5)b-d.

Radiological findings: All patients had features of COVID-19 pneumonia with varying degrees of bilateral GGO and consolidation, as shown in (Table/Fig 5),(Table/Fig 6). Chest X-rays (CXR) showed lung cysts in 4 cases (5.06%) (Table/Fig 6)a and cavities in 2 cases (2.53%). Thirty-eight patients (48.10%) who experienced air leak events underwent High-Resolution Computed Tomography (HRCT) scans during hospitalisation. On HRCT, 19 cases (24.09%) presented with discrete lung cysts (Table/Fig 5)b, and subpleural multiple bullae were observed in two cases (2.53%) (Table/Fig 5)c. Additionally, one case (1.26%) demonstrated a cavity with hydro-pneumothorax (Table/Fig 5)d. No cysts were detected in the other cases.

Outcome: (Table/Fig 4),(Table/Fig 7),(Table/Fig 8) summarises the outcomes in this cohort of COVID-19 pneumonia patients. A total of 20 patients (25.32%) survived, 18 of them had late-onset alveolar air leak events. Overall mortality in this group was 74.68% (n=59). Mortality among females was 81.3%, while it was 73% among males. The gender difference in mortality was not statistically significant (p-value=0.749). In this study, mortality in patients with moderate COVID-19 was 29.41%, primarily due to progressive worsening of hypoxia or sepsis, whereas it was 88.52% in ICU patients with severe COVID-19. In subgroup analysis, invasively ventilated patients with early occurrences of air leak, a PaO₂:FiO₂ ratio of less than 100 at the time of air leak, and acidosis on arterial blood gas analysis had higher mortality (p-value <0.001). Patients with pneumothorax had poorer outcomes compared to those with isolated subcutaneous emphysema or pneumomediastinum (Table/Fig 7). In the subgroup analysis, PaO₂:FiO₂ ratio of less than 100 at the time of air leak had higher mortality. The median age of those who died was higher than that of those who survived, as illustrated in the boxplot (Table/Fig 8).

Eight patients (28.57%) out of a total of 28 with severe SE required decompressing incisions. There were 60 Intercostal Drainage (ICD) procedures performed on 52 patients, 8 (15.38%) of whom had bilateral pneumothorax. Non trocar ICD tubes, ranging in size from 20 to 24, were used in the majority of cases, while a pigtail catheter was inserted in two cases. There was a persistent air leak due to Bronchopleural Fistula (BPF) beyond 4-6 weeks in 5 (9.61%) patients. One case with pyopneumothorax and BPF required lobectomy, while the others were managed conservatively. Histopathological examination of the resected lung segments showed micro-abscesses with vasculitis.

Complications related to ICD (Table/Fig 3):

Bleeding: Out of the 60 ICD procedures, self-limiting bleeding occurred after emergency ICD in 6 (10%) cases of tension PT. These patients had severe COVID-19 pneumonia and were on therapeutic anticoagulation. Major bleeding complicated two cases (3.33%), around 400 mL-600 mL lost within 4-6 hours post-ICD insertion. One patient on Non Invasive Ventilation (NIV) survived with spontaneous cessation of bleeding but later developed segmental pulmonary artery embolism. The other case, complicated by sepsis and coagulopathy, died due to simultaneous multisite bleeding, shock, and severe hypoxia.

Empyema: Post-ICD empyema secondary to nosocomial pathogens developed after the second week of ICD in 6 (10%) cases. All recovered with antibiotics and ICD drainage.

Discussion

Alveolar air leak events PT, PM in COVID-19 pneumonia have been reported in both spontaneously breathing and ventilated patients. Secondary spontaneous PT has also been reported in cases of H1N1, SARS, and Middle East Respiratory Syndrome (MERS) viral pneumonia-associated Acute Respiratory Distress Syndrome (ARDS) (11),(12),(13).

The present study provides a large single-institution data from western India on spontaneous alveolar air leak events in COVID-19 pneumonia. The proportion of alveolar air leaks occurring in spontaneously breathing patients was higher (44.30%) in this study group compared to other studies (20-33%) (6),(14). The male population was affected more than the female population (4:1), which was consistent with other studies (1),(14). In a study by Martinelli AW et al., 90% of the patients were aged between 30 and 70 years, similar to present study cohort (1). Co-morbid conditions such as diabetes and hypertension were associated with increased severity of SARS-CoV-2 pneumonia (15). Asthma and COPD are known risk factors for secondary spontaneous pneumothorax (4). However, this study cohort had a very low prevalence of underlying respiratory conditions that predispose individuals to spontaneous PT. Right-sided PT was more common than left-sided (58.62% vs. 27.59%). In a study by Geraci TC et al., 50% of cases had right-sided PT (14). The average median age of those who died was 57 years. In a multicentric study from Mumbai, the mean population age was 60 years, with a mortality rate of 74% (6). The severity of COVID-19 pneumonia was associated with an increased risk of spontaneous air leak events (p-value<0.001), which may be related to severity of alveolar damage. Cases of pneumothorax related to SARS were also associated with severe diffuse alveolar damage (16).

This study highlights the outcomes of cases in relation to the timing of the occurrence of alveolar air leaks. Early occurrence of air leaks within two weeks of symptom onset was associated with poor outcomes (p-value<0.001). Multiple mechanisms play a role in spontaneous alveolar air leaks in COVID-19 pneumonia. One such mechanism is extensive alveolar injury caused by the SARS-CoV-2 virus, leading to alveolar rupture secondary to damage of type-1 and type-2 alveolar epithelial cells (17). This can explain the early events occurring during the acute phase within two weeks of illness. Another mechanism could involve the formation of cystic spaces in the lungs, which are prone to rupture (18),(19). The cause may be ischaemic injury leading to decreased lung compliance and cyst formation or could be secondary to resolving consolidation (20). This mechanism will be responsible for late events that occur after two weeks of illness.

In the present study, 51.90% of air leaks occurred after two weeks (late events) of symptom onset. In a large multicentric case series, pneumothorax events were reported after a median of 14 days of hospitalisation (11). The present study can relate these early and late radiological findings, ranging from bilateral subpleural patchy GGO with extensive consolidation and interstitial emphysema in early cases to cyst or bullae formation with interstitial thickening in late events. Another added risk factor was cavitating pneumonia and Pyo-pneumothorax secondary to Klebsiella pneumoniae. There is a higher risk of these infections in critically ill COVID-19 patients (21). Secondary bacterial infections were lower in those who survived, but the difference was not statistically significant.

In the present study, two patients had interstitial emphysema and PM (Table/Fig 5)a without PT on HRCT. Most of the awake patients had severe bouts of dry cough as a predominant symptom. This leads to an increase in alveolar pressure and contributes to alveolar rupture in already damaged parenchyma (22),(23). Sudden alveolar rupture resulting in interstitial emphysema and PM can be explained by the Macklin effect (23).

Invasive Positive Pressure Ventilation (IPPV) may also contribute to barotrauma in addition to the mechanisms mentioned above. According to an update by Diaz R and Heller D every mechanically ventilated patient is at risk for barotrauma (24). This risk significantly increases with ventilator settings such as large tidal volumes or high Positive End-Expiratory Pressure (PEEP), which manifest as high plateau pressures (25). The COVID-ARDS ventilation in the study population was in accordance to the ARDSnet protocol of low tidal volume, i.e., 4-6 ml/kg of ideal body weight, along with prone positioning (25). Four patients had plateau pressures between 30 and 40 cm H₂O just before the event. Plateau pressures more than 35 cm H₂O are associated with an increased risk of barotrauma (26).

In the present study, the incidence of PT was 1.14% among hospitalised cases of COVID-19 pneumonia, which was in accordance with recent studies (1),(6). Retrospective studies have shown that PT in COVID-19 might occur in 1% of the total hospitalised patients and in 2% of ICU admissions, contributing to 1% of deaths from COVID-19 infection (27),(28),(29).

In a similar study by Tetaj N et al., the incidence of barotrauma was 5.8% in COVID-19 patients on invasive ventilation, compared to 6.47% in this cohort (30). This incidence of barotrauma in COVID-19 was quite low compared to MERS-CoV infected patients (30%) and SARS infected patients (12-34%) (11),(12),(31).

In the present study, patients with a late occurrence of alveolar air leak had better survival outcomes, even though prolonged ICD was required in few of them. This point towards the primary disease processes, such as ARDS or the cytokine storm, may contribute to increased mortality in the early phase. Other studies in COVID-19 support these findings (15),(29).

In severe COVID-19 cases requiring mechanical ventilation, reported mortality rates range from 57% to 94% (6),(15),(32),(33). In the present study, among a cohort of COVID-19 patients with air leak events, the mortality rate in those with moderate COVID-19 pneumonia (admitted to the ward) was 29.41% (5/17). The overall mortality rate in moderate to severe COVID-19 cases with alveolar air leaks was 74.68%, which was similar to the findings of a multicentric study from western India, where the mortality rate was 74% (6). In a study by Geraci TC et al., the in-hospital mortality rate was 58% for COVID-19 patients who developed PT, with most events occurred beyond the second week of hospitalisation (14). This aligns with the findings of the present study, where the mortality rate for late-onset air leak events was 56%. Mortality among the female subgroup was slightly higher than that of males, as most of the cases belonged to age group of 61-70 year age group. Patients with acidosis (pH < 7.35) had significantly increased mortality. Similar findings were noted by Martinelli AW et al., (1).

In the study by Martinelli AW et al., the 28-day mortality rate did not significantly differ between those who developed PT and those with isolated PM (1). In the present study, there was a statistically significant difference in mortality between the PT group and other cases of alveolar air leaks without PT. These findings correlate with the systematic review by Zhong Z et al., (34).

In the present study, the incidence of major bleeding complications from the insertion of an ICD was very low at 3.33%, similar to the study by Geraci TC et al., (14). Bleeding complications are known to occur in pleural procedures (4). Severe cases of COVID-19 pneumonia were on therapeutic anticoagulation, and spontaneous cessation of bleeding occurred after anticoagulation was withheld (8). In non life-threatening situations, adherence to the protocol for periprocedural management of anticoagulation is necessary. Surgical intervention was required in only one case of persistent Bronchopleural Fistula (BPF). In the study by Geraci TC et al., 5% of cases underwent surgical intervention (14). (Table/Fig 9) highlights the salient review of relevant studies on COVID-19-related PT/barotrauma (1),(6),(14),(30).

Limitation(s)

As this was a single-centre study, the results cannot be generalised. Many asymptomatic patients with subtle PT, PM, and SE may have been missed, as serial HRCT scans were not performed in all hospitalised patients. This study does not comment on the incidence of PT in the male and female populations separately, the incidence in COVID-19 patients on non invasive ventilation (NIV) or high-flow nasal cannula (HFNC), or the comparative outcomes of patients without air leak events.

Conclusion

In this cohort of COVID-19 cases from western India, the cumulative incidence of spontaneous alveolar air leak was 1.55%. The occurrence of PT within the first two weeks of illness in a severely hypoxemic patients on invasive mechanical ventilator carries poor prognosis. There is a need for multicentric prospective case-control studies to determine the incidence of spontaneous secondary PT and the factors affecting its outcome.

DOI and Others

DOI: 10.7860/JCDR/2024/69709.20267

Date of Submission: Feb 01, 2024
Date of Peer Review: Mar 08, 2024
Date of Acceptance: Oct 10, 2024
Date of Publishing: Nov 01, 2024

AUTHOR DECLARATION:
• Financial or Other Competing Interests: None
• Was Ethics Committee Approval obtained for this study? Yes
• Was informed consent obtained from the subjects involved in the study? Yes
• For any images presented appropriate consent has been obtained from the subjects. Yes

PLAGIARISM CHECKING METHODS:
• Plagiarism X-checker: Feb 02, 2024
• Manual Googling: Mar 13, 2024
• iThenticate Software: Oct 09, 2024 (12%)

ETYMOLOGY: Author Origin

EMENDATIONS: 8