Pulmonary Findings of [18F]FDG PET/CT Images on Asymptomatic COVID-19 Patients

During the coronavirus disease 2019 (COVID-19) pandemic, several case studies demonstrated that many asymptomatic patients with COVID-19 underwent fluorine-18 fluorodeoxyglucose ([18F]FDG) positron emission tomography/computed tomography (PET/CT) examination for various indications. However, there is a lack of literature to characterize the pattern of [18F]FDG PET/CT imaging on asymptomatic COVID-19 patients. Therefore, a systematic review to analyze the pulmonary findings of [18F]FDG PET/CT on asymptomatic COVID-19 patients was conducted. This systematic review was performed under the guidelines of PRISMA. PubMed, Medline, and Web of Science were used to search for articles for this review. Articles with the key words: “asymptomatic”, “COVID-19”, “[18F]FDG PET/CT”, and “nuclear medicine” were searched for from 1 January 2020 to 20 May 2021. Thirty asymptomatic patients with COVID-19 were included in the eighteen articles. These patients had a mean age of 62.25 ± 14.85 years (male: 67.71 ± 12.00; female: 56.79 ± 15.81). [18F]FDG-avid lung lesions were found in 93.33% (28/30) of total patients. The major lesion was [18F]FDG-avid multiple ground-glass opacities (GGOs) in the peripheral or subpleural region in bilateral lungs, followed by the consolidation. The intensity of [18F]FDG uptake in multiple GGOs was 5.605 ± 2.914 (range from 2 to 12) for maximal standardized uptake value (SUVmax). [18F]FDG-avid thoracic lymph nodes (LN) were observed in 40% (12/40) of the patients. They mostly appeared in both mediastinal and hilar regions with an SUVmax of 5.8 ± 2.93 (range from 2.5 to 9.6). The [18F]FDG uptake was observed in multiple GGOs, as well as in the mediastinal and hilar LNs. These are common patterns in PET/CT of asymptomatic patients with COVID-19.


Introduction
Coronavirus disease 2019 (COVID- 19) was first detected in Wuhan, China at the end of December 2019 [1]. It is an infectious lung disease causing severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) [2,3]. Since 2019, COVID-19 has rapidly spread across the entire world. As of 23 June 2021, the World Health Organization (WHO) has estimated that there are over 178.5 million confirmed positive patients and over 3.8 million deaths [4]. Despite ongoing vaccination in many countries, more time is needed to achieve global herd immunity. A major prevention strategy for COVID-19 is still to control the transmission of SARS-CoV-2.
The common clinical symptoms of COVID-19 include fever, fatigue, dry cough, and pneumonia, as well as other rare signs such as headache, nausea, vomiting, and diarrhea [5,6]. However, some SARS-CoV-2 infected individuals are asymptomatic during COVID-19 illness [7,8]. Asymptomatic cases with COVID-19 are a source of SARS-CoV-2 transmission in social situations. They have infected more than 40% of positive cases. Since asymptomatic individuals do not present any clinical symptoms of COVID-19, it is not possible to quickly identify them for intervention. Therefore, the prevention strategy for COVID-19 is complicated due to these asymptomatic cases.
The radiopharmaceutical 2-deoxy-2-[fluorine -18] fluoro-D-glucose ([ 18 F]FDG) is a glucose analog in which the normal hydroxyl group at the C-2 position in the glucose molecule is replaced with the positron-emitting radionuclide fluorine-18. Imaging is based on the detection of gamma rays after positron-electron annihilation with a positron emission tomography (PET) machine. Combined with computed tomography (CT) to assist in the localization of a lesion, PET/CT imaging with [ 18 F]FDG can detect metabolic status based on degree of glucose utility in a variety of tissues. [ 18 F]FDG PET/CT is commonly used for diagnosis, staging, or restaging of malignant disease due to increased glucose uptake and glycolysis in tumor cells [9,10]. Apart from malignant disease, it is also used to characterize infection and aseptic inflammation based on the high glucose uptake of activated inflammatory cells [11].
Recently, several case studies demonstrated that many asymptomatic patients with COVID-19 underwent [ 18 F]FDG PET/CT examination due to other clinical indications. However, there is a lack of literature to characterize the disease pattern of [ 18 F]FDG PET/CT in asymptomatic COVID-19 patients. Herein, we present a systematic review to analyze the pulmonary findings of [ 18 F]FDG PET/CT in asymptomatic COVID-19 patients.

Data Sources and Review Approaches
The guidelines of PRISMA were used to conduct this review ( Figure 1). A literature review was performed from 1 January 2020 and up to 15 May 2021 by searching in electronic databases-mainly PubMed, Medline, and Web of Science. The central aim of this review was to detect findings of [ 18 F]FDG PET/CT in asymptomatic patients with COVID-19. The types of studies in this review were restricted to case reports, case series, and retrospective case reports. The key words "asymptomatic", "COVID-19", "FDG PET/CT", and "nuclear medicine" were used to search the literature for this review. Duplicated articles and articles with only abstracts were removed. Irrelevant articles were deleted after screening the titles and abstracts. All authors reviewed relevant articles and excluded those without abstract and PET/CT images, or those whose subjects did not have a laboratory-confirmed test for COVID-19. The language of all articles was limited to English.

[ 18 F]FDG PET/CT Findings
The pulmonary findings of asymptomatic patients with COVID-19 showed that increased [ 18 [15], 21 [20], and 24 [27].) Patients numbered P11 and P21 demonstrated different oncological diseases with no [ 18 F]FDG-avid lung lesion and no above pulmonary alternation, respectively [15,22]. P24 was a non-oncological patient with unknown anosmia, and the abnormal [ 18 F]FDG uptake was only detected on her right hilar LN [27]. Of these 27 patients, 11 of them had [ 18 F]FDG-avid lung lesions and thoracic involvement (P6, 7, 8, 9, 10, 12, 15, 17, 18, 25, and 28). Fifteen patients with [ 18 F]FDG-avid lung lesions had previously unknown performance of thoracic LN involvement (P1, 3, 4, 5, 13, 14, 16, 19, 20, 22, 23, 26, 27, 29, and 30). [ 18 F]FDG-avid lung lesions alone were only detected in patient P2, with no thoracic LN involvement. Results from extrapulmonary and pulmonary findings showed that many asymptomatic COVID-19 patients with or without oncological disease had [ 18 F]FDG-avid lung lesions. Some of that group also had thoracic LN involvement, mostly in patients without non-thoracic LN involvement (n = 9: P7, 9, 10, 12, 15, 17, 18, 24, and 28). A rare percentage of asymptomatic COVID-19 patients presented thoracic LN involvement alone or with no pulmonary lesions.  Table 2. Ground-glass opacities (GGOs) with other lesion patterns such as consolidations, curvilinear lines, crazy paving, and lobar thickening, etc., were found on 50% (15/30) of total patients. This was slightly more than the percentage of patients with GGO alone (43.44%; 13/50). It was revealed that GGO was a major lung lesion for asymptomatic patients with COVID-19. Lung lesions were found in 93.33% (28/30) of the total patients, and 70% (21/30) of these presented with multiple lesions on their lungs, compared with 13.33% (4/30) with unique lesions and 10% (3/30) with unique lesions at multiple sites. Unique lesions were only seen in patients who had GGOs with other patterns (n = 7: P5, 7, 9, 10, 23, 26 and 30). This indicated that asymptomatic patients with COVID-19 frequently presented multiple lung lesions.   (6/30). This suggests that lesions frequently occurred on more than 3 lobes in asymptomatic patients with COVID-19. The distribution of lesions in affected lobes was studied in half of the total patients. Lesions were mostly distributed on the peripheral or subpleural region of affected lobes. In addition, the results of the location distribution of affected lobes from 73.33% (22/30) of total patients demonstrated that the infection and/or inflammation caused by COVID-19 could occur in any region in both lungs. The finding was similar to that of pulmonary involvement for asymptomatic patients with COVID-19. Of the total patients, 70% (21/30) had bilateral lung involvement. These findings indicated that lung lesions were detected on any lobe, either in the peripheral or subpleural regions of lung, in asymptomatic patients with COVID-19.

Characteristics of Pulmonary Findings
In this systematic review, most COVID-19 patients were incidentally detected by [ 18 F]FDG PET/CT scan, which was originally used for tumor staging or determination of suspicious disease progression. Differentiation between lung metastasis and COVID-19 infection is important but difficult to achieve for physicians, especially for patients who underwent [ 18 F]FDG PET/CT. The number of affected lung nodules may be single or multiple in metastasis and/or COVID-19 patients. However, metastatic lung nodules may appear round-shaped, well-circumscribed, with variability in size and soft tissue attenuation, mainly in the peripheral region of the lung. In our observations, 70% of COVID-19 patients presented multiple GGOs involving bilateral lung abnormalities with corresponding mild-to-hot [ 18 F]FDG uptake. This indicated that COVID-19 infection was related to glucose hypermetabolism. The location of GGOs could be peripherally distributed with lower lobe involvement. Other patterns were observed in COVID-19 patient's lungs, including lung consolidation, linear opacity, septal thickening, tree-in-bud opacity, or pleural thickening; all of these showed mild-to-moderate [ 18 F]FDG uptake. It is different in main lung involvement; cavitation is not frequently noted in these patients, dissimilarly to patients with tuberculosis infection or squamous cell carcinoma. Nevertheless, its appearance suggests lung injury. We suggest suspecting individuals with characteristics of [ 18 F]FDG uptake appearance in lung and specific CT scan morphology of being COVID-19 patients. Accordingly, these patients should proceed to screening for SARS-CoV-2 infection. Table 2 shows the features of LN involvement, which randomly occurred as diverse patterns in lung lesions. The location of thoracic LN involvement was frequently in both sides of the mediastinal and hilar regions with or without other thoracic LN, i.e., subclavian or carinal LNs (Table 2), except for the above characterization of [ 18 F]FDG-avid lung lesions or thoracic LN. The intensity of [ 18 F]FDG uptake was separately analyzed on lung lesions and thoracic LNs. Maximal standardized uptake value (SUVmax), maximal SUV based on body weight (SUV bw max), or maximal SUV based on lean body mass (SUV lbm max) were used to represent the degree of [ 18 F]FDG uptake in the lung lesion or thoracic LNs. The range of [ 18 F]FDG uptake in lung lesions was from 2 to 12 for SUVmax, 3.3 to 10.7 for SUV bw max, and 3.7 to 6.8 for SUV lbm max, respectively. The SUVmax in thoracic LNs ranged from 2.5 to 9.6, while that of SUV bw max ranged from 3.9 to 5.9. These findings demonstrate that asymptomatic patients with COVID-19 presented various degrees of [ 18 F]FDG uptake in both lung lesions and regions of involved thoracic LNs.  [12][13][14][15]19,20,[22][23][24][25][26][27][28]. However, some hospitals did not emphasize the status of being under quarantine [12,13,19,22,[24][25][26]. Of these cases, P27 was home quarantined before the laboratory confirmation of COVID-19 infection [20]. Subsequently, P27 was confirmed later due to symptom onset. In addition, some hospitals in Europe referred the patients to a dedicated COVID-19 medical facility for laboratory COVID-19 confirmation [13,15,22] and COVID-19 treatment [13,22]. Patients at registered hospitals were also provided with COVID-19 treatment [12,18,20]. Oncological therapy was postponed for some patients with COVID-19 [12,22].

Patient Management and Hospital Infection Control
As the radiologists had insufficient experience in COVID-19 imaging on [ 18 F]FDG PET/CT, two patients (P5 and P6) from hospitals in Italy were not immediately confirmed by laboratory COVID-19 testing [16,21]. This also happened to P30 at another hospital in Spain [17]. P6 and P30 were confirmed to have SARS-CoV-2 infection later due to symptom onset. However, this was over four days after the [ 18 [14,19,28]. These three hospitals were in Portugal [14], France [19], and the USA [28]. The guidelines for hospital infection control are presented in Table 3. Two hospitals screened for the COVID-19 infection risk of every patient before and upon entering the facility [14,19]. All hospitals canceled radiopharmaceutical imaging investigations and cleaned the imaging room after having suspected or confirmed cases of COVID-19 [14,19,28]. The hospital in Portugal even cleaned the imaging equipment after each use [14]. In addition, employees were required to self-monitor after having contact with suspected cases of COVID-19 [19,28]. These guidelines expanded the prevention targets to each patient and also increased the disinfection level. Overall, it is suggested to take extensive measures to avoid the spread of COVID-19 via contact with asymptomatic patients.

Discussion
This systematic review aimed to analyze the pulmonary findings of [ 18  GGOs on both peripheral and subpleural lungs was the most common feature of chest CT on asymptomatic COVID-19 patients [30] and is consistent with our findings on PET/CT. Besides in asymptomatic COVID-19 patients, GGOs are also found in other infectious diseases with a variable degree of [ 18 F]FDG uptake [31]. Abnormal [ 18 F]FDG uptake in mediastinal and hilar LNs is found in malignant diseases such as lung cancer and lymphoma [32,33], and in acute primary pulmonary histoplasmosis [34]. LN alternation on chest CT is commonly found in patients with severe COVID-19 [35]. It was first found on [ 18 F]FDG PET/CT for asymptomatic patients with COVID-19.
Some limitations in this review should be considered. First, during the time of this review's preparation, new and updated findings were released which are not included. Second, most asymptomatic cases with COVID-19 were from Europe, and cases from other areas should also be investigated.

Conclusions
This review demonstrates that asymptomatic COVID-19 patients present a wide range of [ 18 F]FDG uptake patterns in multiple GGOs and on both sides of mediastinal and hilar LNs. Although these patterns can be also found in other infectious or malignant diseases, the [ 18 F]FDG PET/CT discloses the increased metabolic status that is revealed in asymptomatic COVID-19 patients. Nuclear medicine staff should consider the risk of the spread of COVID-19 during [ 18 F]FDG PET/CT examination. Every patient should be considered possibly infectious and treated with universal precautions. This study emphasizes the need for such universal precautions.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.