Autopsy Findings of Severe COVID-19 Pneumonia Combined with Pulmonary Aspergillosis, Pneumothorax, and Pulmonary Thromboembolisms

Background

Coronavirus disease 2019 (COVID-19) is accompanied by many complications. Pneumonia is most frequently observed and can cause acute respiratory distress syndrome (ARDS). Pulmonary thromboembolism (PTE) is also a well-known complication. A pooled analysis of 3342 COVID-19 patients revealed that the incidence of PTE was 16.5% in total, and reached 24.7% in patients admitted to the Intensive Care Unit (ICU) [1]. In a study of mechanically ventilated COVID-19 patients in an ICU, 20 of 168 patients (11.9%) developed COVID-19-associated pulmonary aspergillosis (CAPA), 13 of whom (65%) died [2]. A review of 34 CAPA cases also reported high mortality (64.7%) [3]. However, autopsy cases with multiple complications associated with COVID-19 have not been well reported. Here, we describe an autopsy case with severe COVID-19 pneumonia, PTE, CAPA, and CAPA-induced pneumothorax to help in the understanding of severe COVID-19 and the future development of its treatment strategy.

Case Report

A 64-year-old Japanese man presented with fever and a cough (day 0) and was diagnosed with COVID-19 by positivity for SARS-CoV-2 PCR in the delta variant spread period in Japan (July to September, 2021). Six days later (day 6), COVID-19 pneumonia was recognized by computed tomography showing a wide area of ground-glass opacities (Figure 1A). He was admitted to a hospital because of hypoxemia and started to receive oxygen inhalation, remdesivir (200 mg on the first day and then 100 mg per day), dexamethasone (6.6 mg per day), baricitinib (4 mg per day), and edoxaban (60 mg per day). However, his respiratory condition worsened, noninvasive positive-pressure ventilation (NPPV) management was initiated on day 15, and he was transferred to our hospital on day 17. He had never received SARS-CoV-2 vaccination. Although he had type 2 diabetes mellitus, he did not receive any treatment. He had no other comorbidities. SARS-CoV-2 was again positive by PCR. His height was 163 cm, and his body weight was 73 kg (body mass index 27.5). The respiratory rate was 28 per minute. Fine crackles were auscultated. SpO2 was 90% under NPPV with FiO2 100%. Blood tests revealed WBC 24 740/mL (neutrophils 95.9%), hemoglobin 13.6 g/dL, platelets 59 000/mL, C-reactive protein 2.94 mg/dL, lactate dehydrogenase 985 U/L, glucose 318 mg/dL, HbA1c 8.5%, KL-6 5671 U/mL, ferritin 1318 ng/mL, D-dimer 12.2 mg/mL, and b-D-glucan 9.1 pg/mL. Chest CT showed extremely extended ground-glass opacities in both lung fields with partial consolidation posteriorly (Figure 1B).

Arterial blood gas analysis (NPPV, FiO2 100%) showed pH 7.359, PaCO2 42.9 Torr, and PaO2 58.7 Torr. Invasive mechanical ventilation was started. Management by prone positioning could be implemented for only 1 day due to manpower issues. Methylprednisolone (1000 mg per day) instead of dexamethasone was administered because of progressive respiratory failure during dexamethasone and doses were reduced by half every 3 days. Intravenous unfractionated heparin was started continuously based on its short half-time and adjusted according to activated partial thromboplastin time. A foot pump was also used to combat risk of thromboembolism. Because Aspergillus fumigatus was detected from intratracheal samples when the patient began invasive mechanical ventilation management prior to the start of high-dose methylprednisolone, voriconazole was started (800 mg on the first day and then 400 mg per day). Despite these treatments, the platelet count further decreased to 13 000/mL, and platelet transfusions were needed. Hypoxemia deteriorated and bilateral consolidations on chest X-ray increased on day 28 (Figure 1C). At this time, SARS-CoV-2 PCR was negative. On day 31, right pneumothorax occurred, and a drainage tube was inserted. On day 32, the patient died of respiratory failure.

Autopsy was performed. Macroscopically, there was a hole in the right upper lobe (Figure 2A, arrow). Microscopically, there were several cavities accompanied by hemorrhage and necrosis dominantly in the right upper lobe (Figure 2B). One cavity formation adjacent to the pleural membrane was broken, which was the cause of the pneumothorax (Figure 2B, arrow). In and out of cavities, agglomeration of aspergillus was recognized, consistent with invasive pulmonary aspergillosis (Figure 2C, 2D). Inflammatory cells infiltrated around aspergillus were predominantly neutrophils, with some lymphocytes and histiocytes, accompanied by nuclear waste materials. Aspergillus was also detected in the left upper lobe. In a wider area of both lungs than pulmonary aspergillosis, alveolar structures were destroyed with fibrotic changes and infiltration of inflammatory cells and fibroblasts (Figure 3A), suggesting diffuse alveolar damage (DAD) had occurred. The formation of hyaline membranes was also observed (Figure 3B), suggesting a relatively new DAD. Pulmonary edema was shown in some parts (not shown). Many thromboembolisms were recognized in peripheral pulmonary arteries (Figure 3C). Organized thromboses were also observed (Figure 3D). There were no obvious thromboses in organs other than the lungs.

Discussion

Pathological findings of autopsy are summarized as follows: (1) old and newly developed DAD in a wide area of the lungs, which is consistent with ARDS due to COVID-19 pneumonia; (2) PTEs in peripheral pulmonary arteries; (3) CAPA; and (4) pneumothorax by CAPA. The autopsy clarified that these pathological changes led to respiratory failure.

DAD is the most common pathological finding of ARDS due to COVID-19 pneumonia [4]. DAD was recognized in 127 (80.9%) of 157 COVID-19 patients in whom postmortem autopsy was performed [5]. In the present case, various time phases of DAD were recognized: the exudative phase, represented by the formation of hyaline membranes; the proliferative phase, in which many inflammatory cells and fibroblasts are present; and the fibrotic phase, in which fibrosis is dominant. Although over 1 month had passed from COVID-19 onset, and SARS-CoV-2 PCR had already turned negative, findings of not only the late phase but also the early phase of ARDS were observed. This might be suggestive of insufficiency of the dose or period of corticosteroids or non-response to corticosteroids.

Corticosteroids are now recommended for severe COVID-19 pneumonia [6]. In the RECOVERY trial, dexamethasone (6 mg/day for 10 days) decreased mortality for COVID-19 patients receiving either oxygen or invasive mechanical ventilation [7]. A recent prospective randomized trial showed that the methylprednisolone (2 mg/kg/day) group (n=47) demonstrated better clinical status, including a lower admission rate, improvement in the WHO ordinal scale, and a lower rate of mechanical ventilation use, than the dexamethasone (6 mg/day) group (n=46) [8]. There is no evidence of efficacy of high-dose corticosteroids such as methylprednisolone pulse therapy for COVID-19 pneumonia. In the present case, ARDS progressed even after treatment with a high dose of methylprednisolone. On the other hand, there are COVID-19 cases that improve without corticosteroid. There is no uniform dose of corticosteroid suitable for all cases. Thus, the quantity of necessary corticosteroids should be decided for each case depending on the clinical situation, such as infection.

Venous thromboembolism, including PTE, is well-recognized as a complication of COVID-19. Intensified doses of heparin are recommended for antithrombotic prophylaxis, depending on the clinical or biological severity of COVID-19 [9]. Although low-molecular-weight heparin, such as enoxaparin, would also be effective [10], standard drugs and their doses have not yet been determined for COVID-19. In the present case, an intensified dose of unfractionated heparin was intravenously administered. However, thrombocytopenia continued throughout the course, and autopsy revealed many newly formed PTEs. This suggests that cases exist in which even continuous heparin is insufficient for antithrombotic prophylaxis.

Several autopsies of CAPA cases have been reported [2,11,12]. Several patterns have been reported, including airway colonization [2], focal pulmonary aspergillosis [12], and invasive pulmonary aspergillosis (IPA) [2,11]. The current case was classified as IPA, and the existence of ARDS would be a characteristic finding of CAPA. A study reported that the incidence of CAPA increased in the second wave of the pandemic, in which corticosteroids were regularly used for moderate-to-severe COVID-19 pneumonia, compared with the first wave [11]. As characteristics of patient with CAPA, hypertension, diabetes mellitus, obese, use of mechanical ventilation, and use of corticosteroids have been reported [3]. Interleukin-6 is secreted in severe COVID-19 patients and can play an important role in protective immunity against Aspergillus [13]. In this regard, blockade of interleukin-6 may potentially progress Aspergillus infection. Although the efficacy of corticosteroids in severe COVID-19 pneumonia is established, the pros and cons of corticosteroids in the development of CAPA are not clear. In the present case, risk factors for CAPA development could be uncontrolled diabetes mellitus, dexamethasone (6.6 mg per day for 10 days), baricitinib, management with mechanical ventilation, and no SARS-CoV-2 vaccination history. However, the reason for such medication and respiratory management is severe COVID-19, and we already know some ways to prevent COVID-19 from becoming severe, including vaccination and appropriate blood glucose control.

Pneumothorax is a complication of severe COVID-19 and occurs in approximately 10% of critically ill patients [14,15]. Several causes of pneumothorax in ARDS have been illustrated, including increased alveolar pressure caused by mechanical ventilation, changes in alveolar structure and function, increased negative pressure of the pleural cavity by severe cough or forced inhalation, and shear stress [15]. The present case revealed that CAPA can cause pneumothorax, which is the first report of its kind.

The autopsy of the present case showed that the findings observed in the severe COVID-19 case were gathered in 1 patient. In addition, ARDS, PTE, and CAPA were all still active states; in other words, their treatments were insufficient. The treatment strategy, including the dose of corticosteroids for such severe COVID-19 cases, is an unresolved problem.

Conclusions

Active states of combined ARDS, PTEs, CAPA, and CAPA-induced pneumothorax were observed in 1 patient with severe COVID-19. It is difficult to improve these conditions simultaneously because their treatments can induce antagonizing biological actions. Severe COVID-19 can be prevented by the use of vaccinations and reduction of risk factors.