key: cord-0840488-7flv9yed
authors: dos Reis, José Maciel Caldas; Melo, Glauco dos Santos; de Oliveira, Murilo Vasconcelos; Fernandez, Mariana Morgado; da Silva, Tereza Maria Meireles Fernandes; Ferreira, Hugo Luis da Silva; de Andrade, Mariseth Carvalho
title: Incidental cardiovascular findings on chest CT scans requested for suspected COVID-19
date: 2022-01-07
journal: J Vasc Bras
DOI: 10.1590/1677-5449.210052
sha: 8af507388562e7d230bb32d773d10a45323bd157
doc_id: 840488
cord_uid: 7flv9yed

BACKGROUND: Computed tomography scans of the chest are often requested as a complementary examination to investigate a clinical suspicion of pulmonary disease caused by the novel coronavirus 19 (COVID-19). OBJECTIVES: Our objective was to analyze the prevalence of incidental cardiovascular findings on chest CT scans requested to assess radiological signs suggestive of COVID-19 infection. METHODS: This cross-sectional, descriptive, and retrospective study reviewed 1,444 chest tomographies conducted in the Radiology department of the Hospital de Clínicas Gaspar Vianna, from March 1 to July 30, 2020, describing the prevalence of images suggestive of viral pneumonia by COVID-19 and incidental pulmonary and cardiovascular findings. RESULTS: The mean age of the patients was 50.6 ± 16.4 years and female sex was more frequent. Computed tomography without contrast was the most frequently used method (97.2%). Aortic and coronary wall calcification and cardiomegaly were the most prevalent cardiovascular findings. CT angiography revealed aortic aneurysms (9.7%), aortic dissection (7.3%) and thoracic aortic ulcers (2.4%). CONCLUSIONS: Incidental cardiovascular findings occurred in about half of the chest CT scans of patients with suspected COVID-19, especially aortic calcifications, cardiomegaly, and coronary calcification.

Incidental findings in imaging exams are defined as unexpected findings unrelated to the scope of the clinical indication. 1, 2 They are a common and wellknown phenomenon in computed tomography (CT) examinations. 3 However, the decisive factor is not the occurrence of a random finding, but its clinical relevance. 1, 3 In the literature on the subject, it is reported that incidental findings can be detected in up to 70% of all imaging investigations and, thankfully, the majority are of little clinical relevance. [1] [2] [3] These findings can be classified according to their clinical importance, as major, moderate, or minor. 1 From this perspective, newly-discovered tumors or aneurysms are of particular relevance, gallstones or pleural effusions can be considered moderately relevant, and uncomplicated liver cysts or simple vascular calcifications are irrelevant findings. The majority of incidental findings are clinically insignificant. 1, 2 Cardiovascular incidental findings include disorders such as aneurysms, calcifications of cardiac valves and arteries, and thromboembolisms, which can sometimes be clinically asymptomatic and require further investigation, or may need therapeutic interventions, as is the case with giant aortic aneurysms. [3] [4] [5] Currently, because of the global health emergency faced by humanity and the scarcity of resources and limitations of large-scale confirmatory tests for the coronavirus 2019 disease (COVID- 19) , such as reverse transcription followed by polymerase chain reaction (RT-PCR), 6 chest CT scans have come to play an important role supporting diagnosis because of their high sensitivity for detection of viral pneumonia and their potential to aid in assessment of disease progression and monitoring of treatment response. 7, 8 As a result, it has become necessary to increasing use imaging techniques with satisfactory accuracy and ensure availability of tomographic images in the majority of emergency units, since this is the preferred diagnostic examination in cases of suspected viral infection by COVID-19. Another advantage of CT, in addition to investigations related to the clinical complaint, is its capacity to identify incidental abnormalities, i.e., findings unrelated to the indications for requesting it. [6] [7] [8] [9] [10] Considering the above, in the present study we intend to analyze the prevalence of incidental cardiovascular findings in chest CTs performed because of suspected COVID-19 at a tertiary hospital in the state of Pará, Brazil.

This is a retrospective and descriptive cross-sectional study conducted at the radiology service of the Hospital de Clínicas Gaspar Vianna (HCGV). The research corpus was collected at a general tertiary hospital specialized in cardiology, nephrology, and cardiovascular surgery, located in the city of Belém (PA).

This project was approved by the Research Ethics Committee at the institution and registered on the Plataforma Brasil (CAAE: 33706220.9.0000.0016) under ruling number 4.142.701.

Patients were identified using a computerized radiographic database that records all of the radiological studies performed by the radiology department. It was possible to include all chest CTs performed at the institution on patients from its various different sectors (wards, emergency department, or outpatients) from March 30 to July 2020 that had been requested because of indications related to respiratory symptoms, flulike syndromes, or suspected COVID-19 pneumonia. CTs conducted because of other indications were excluded.

All CTs were performed using a Siemens Somatom Plus 16 multidetector scanner (Siemens Medical Systems Inc, Iselin, New Jersey, United States), and the tomographic images were recorded on a Vitria workstation and viewed with windows for bone (C800 W2000) and/or lung (C40 W80), both appropriate for the thorax. The examination was performed with the patient in dorsal decubitus with the following CT scan parameters: tube potential, 120 kVp; automatic tube current modulation, 30-70 mAs; pitch, 0.99-1.22 mm; matrix, 512 x 512; slice thickness, 1.0 mm; and field of view, 350 x 350 mm. Examinations reports were prepared by the institution's three staff radiologists. When patients underwent more than one chest CT during the study period, only the results of the first examination were included in the analysis.

The research protocol comprised a list of 35 questions covering three principle topics: sociodemographic aspects, cardiovascular findings, and distribution of incidental findings.

Information was collected on the sample characteristics and organized in a spreadsheet in Microsoft ® Office Excel ® 2016.

The sample size was calculated based on an estimated prevalence of incidental cardiovascular findings of up to 60%, a standard error of 3.0%, and an alpha error of 5%, resulting in a minimum sample of 1,024 CT examinations.

Application of descriptive statistics involved drawing up tables and plotting graphs to present the results and calculating measures of position, such as arithmetic means and standard deviations.

The analytical statistics used to evaluate the results of variables for the sample were G tests and chi-square tests of fit for univariate tables. Descriptive and analytical statistics were computed using BioEstat ® 5.4 software. A significance level of α = 0.05, or 5%, was adopted for decision making, indicating significant results with an asterisk (*).

During the period from March 1 through July 30, 2020, the HCGV radiology department performed a total of 1,444 chest CTs for suspected COVID-19 pulmonary infection.

Mean (standard deviation) patient age was 50.6±16.4 years. The youngest patient was 2 years old and the oldest was 99. There was a statistically significant difference (*p < 0.0001) between the proportions of age groups, with the greatest proportion of patients aged 40 to 49 years (25.7%).

There was a predominance of female patients (738; 51.1%), but the difference between sexes was not statistically significant (p = 0.4146).

The statistically significantly greatest proportion of requests for CT originated from the emergency department (48.8%), as shown in Table 1 .

The most common method used was CT without contrast (97.2%), used for a statistically significant proportion of the sample. Similarly, tomographic findings compatible with COVID-19 pulmonary involvement were observed in a significant 56% of the sample (*p = 0.0002) and the distribution pattern of pulmonary injury was < 25% ground-glass opacity in a statistically significant majority (*p < 0.0001) of patients, as shown in Table 2 .

Analysis of the tomographic findings compatible with COVID-19 pulmonary involvement revealed that a statistically significant percentage of patients had ground-glass opacity (56%). Additional compatible findings included consolidation (18.4%), pleural effusion (12.6%), and parenchymal bands (7.5%); tomographic characteristics suggestive of COVID-19, among the other findings identified in the CT reports, as shown in Table 3 .

Incidental pulmonary findings were reported in 63.2% of chest CTs. The finding with the greatest proportion was pulmonary nodule (18.3%), which was statistically the most significant (*p < 0.0001). The most frequent other findings were peribronchial thickening (13.1%), band atelectasis (11.4%), atelectasis (9.5%), and emphysema (5.7%), as shown in Table 4 .

The database analysis was able to catalogue and distribute the cardiovascular findings present in 51.2% [confidence interval (CI) 48.7-53.8%] of the patients analyzed. The finding seen in the largest proportion of the sample was aortic wall calcification (21.8%), followed by cardiomegaly (10.5%), as illustrated in Table 5 and Figures 1 and 2. In turn, Table 6 shows the distribution of cardiovascular findings in the population over the age of 30 (1,299 patients).

The most prevalent finding identified was aortic wall calcification (16.2%), followed by cardiomegaly (10.7%), and coronary calcification (4.5%).

With regard to aortic conditions, there were 12 incidental diagnoses of thoracic aorta aneurysm (0.83%), seven in the ascending aorta and five in the thoracic descending aorta. These 12 aneurysms were diagnosed in 10 patients (two male patients had two aneurysms each, in the ascending and descending aorta). CT with contrast was conducted for 41 patients (2.8% of all scans), and yielded alternative diagnoses of aortic dissection in 7.3% (three of 41 scans) and thoracic aortic aneurysm in 9.7% (four of 41 scans). There were also findings of aortic ulcers in 2.4% (two of 41 scans). Figures 3 and 4 show examples of incidental aortic conditions detected with chest CT.

The prevalence of pulmonary thromboembolism (PTE) detected with computed tomography angiographies of the chest was 24.4% (10 of 41 examinations with contrast). Twenty-eight of these scans were of patients already in hospital and 13 were of patients referred from the emergency department. The incidental PTE findings were in two cases from the emergency department and eight of those already in hospital.

The important role played by diagnostic imaging has increased the number of CT examinations performed and, as CT became more widely used, radiologists began to observe a wave of findings unrelated to the clinical indications for which the examinations were ordered. 1, 3, 4 Confirmation of a COVID-19 diagnosis is based on RT-PCR of nasal or oropharyngeal samples collected with swabs. Patients infected by the SARS-Cov2 virus may present with changes seen on CT that are typical of the disease, such as ground-glass opacities. However, according to the Brazilian College of Radiology's recommendations, CT should only be performed on symptomatic patients who have been admitted to hospital. 11 Incidental findings are a well-known and common phenomenon in imaging exams. 1, 4, 7, 12 They can be classified according to their clinical importance, as major, moderate, or minor. 12 Major findings include lesions suspected to be malignant diseases, such as thyroid nodules, changes to the thickness of the intestinal wall, and solid pancreatic or renal masses. 12 Moderate findings, i.e. findings possibly of clinical relevance, include gallstones and pleural fluid accumulations or pleural effusion. Minor findings, or findings without clinical relevance, include simple renal or hepatic cysts, degenerative changes to the spinal column, and calcification of vessels. Many incidental findings are clinically insignificant. 1, 2, 4, 7 This study found that female patients predominated and the most frequent source of patients was the institution's emergency department, which requested 48.8% of the examinations.

The routine chest CT protocol used at the HCGV is a low-dose, 1.0 mm slice program without intravenous contrast. The great majority of chest CTs (97.2%) were performed without contrast, with contrast reserved for cases in which PTE or other vascular disorders requiring more detailed analysis were suspected.

The most common source of Chest CTs requests was the emergency department (48.8%). This elevated demand from external patients (rather than patients already admitted to the wards or intensive care unit) is probably because of the hospital's characteristics, since it has an open-door emergency service for cardiovascular and nephrology patients.

Although CT is far from being the only examination of choice recommended for diagnosis by the various different medical specialty societies, it has become a valuable tool to support diagnosis in these patients, in addition to its utility for monitoring progress and detecting possible complications. 12, 13 Thus, 56% of chest CTs requested had findings compatible with COVID-19 pulmonary involvement. The most characteristic findings were ground-glass opacity and consolidation; but many other findings can suggest pulmonary damage, depending on the disease phase and time since onset. 13 A total of 63.2% of the CTs requested for diagnosis of presumed COVID-19 had incidental findings and the majority of these comprised pulmonary nodules (18.3%), peribronchial thickening (13.1%), band atelectasis (11.4%), and atelectasis (9.5%). Nodules were statistically significantly more common than the other pulmonary findings (p < 0.0001). It is believed that small nodules are not clinically relevant. However, eight patients exhibited findings of pulmonary masses larger than 3 cm (0.5% of the sample). This supports the hypothesis that the great majority of incidental findings are irrelevant, but the method can also detect asymptomatic injuries that need further management. 11 Thus, although bilateral ground-glass opacity and consolidation are described as the predominant findings characteristic of imaging exams conducted for COVID-19, the manifestations seen on chest CT can vary from patient to patient and at different disease stages. 11 Incidental cardiovascular findings were found in 51.2% (CI 48.7-53.8%) of the CT scans conducted and more than 40% had more than one finding. Calcification of aortic and coronary walls and cardiomegaly were the most common in the radiological reports. Less common findings involved hemorrhagic pericarditis and dilatation of the aortic root. These findings can be explained by the epidemiological profile observed in the study, since approximately 30% of the patients were over the age of 60, and also by the profile of the hospital's patients in general, since it specializes in cardiovascular conditions and nephrology.

In an attempt to form more homogenous groups, a subset analysis was conducted after excluding data from 145 examinations of individuals less than 30 years old, leaving a subset of 1,299 CTs and revealing a similar distribution, with predominance of coronary and aorta wall calcification, cardiomegaly, and hemorrhagic pericarditis. Thus, regardless of its incidental nature early detection of potentially clinically relevant diseases, such as coronary calcification and aortic aneurysms or dissections, can change the prognosis of this population and have a positive impact on reduction of mortality and increase of quality of life.

Surov et al. 2 showed that cardiovascular findings can be identified in 6.8% of patients with malignant diseases investigated with CT. In contrast, Jacobs et al. 4 reported that in their study the frequency of aortic aneurysm varied from 0.07% to 3.4%, 4 while the frequency of aortic dissection varied from 0.06% to 0.2%.

Although incidental cardiac findings may not be relevant to the immediate clinical management of patients with suspected COVID-19 pulmonary involvement, they can influence long-term clinical management and improve prognosis. For example, coronary and aorta wall calcifications are known markers of atherosclerosis and signs of underlying cardiovascular disease, very often subclinical, which can influence primary prevention of atherosclerotic events. [14] [15] [16] [17] On the other hand, cardiomegaly constitutes a late characteristic of left ventricular dysfunction and heart failure, both with poor prognosis, and could trigger further investigation and more effective cardiac treatment. 14 Several different studies have shown that atherosclerotic plaques, calcifications, and aorta wall irregularities are very prevalent among patients with cardiovascular disease. [15] [16] [17] Others have demonstrated that calcification of the descending aorta is related to calcification of the coronary arteries, an important predictor of cardiovascular disease. However, to date, no follow-up studies have been conducted to investigate the prognostic value of these abnormalities. 16, 17 Sverzellati et al. 18 reported that 50% of 286 CT examinations ordered for pulmonary fibrosis, suspected pulmonary embolism, or lung cancer staging revealed potentially significant cardiovascular findings. Along the same lines, Choy et al. 19 demonstrated that 61% of a consecutive series of routine chest CTs exhibited cardiac findings that merited reporting.

Previous studies of COVID-19 have shown that 63 to 67% of patients who died had cardiovascular comorbidities, most commonly hypertension, diabetes, and coronary cardiac disease, [20] [21] [22] which are all factors that can be linked to vascular diseases. [19] [20] [21] [22] Aortic conditions were detected incidentally on chest CT in up to 2.4% of cases, among which aneurysms were the most prevalent abnormality, especially in CTs without contrast. In the absence of intravenous contrast, significant aortic dissections and ulcerations are generally undetectable, which can be considered a limitation of the method. 4 In the present study, just 2.8% of the CT scans were conducted with contrast. Aneurysms and dissections were observed in 9.7% and 7.3% respectively and suggested in the radiologist's report in 1.2% of scans without intravenous contrast. On the other hand, PTE was observed in 24.4% of the patients who had scans with contrast, constituting a significant sample of cases admitted to the intensive care unit, where the more severe COVID-19 cases predominate.

Thus, is it is important to emphasize that incidental findings constitute an important event for patient clinical outcomes, since a considerable proportion of them have comorbidities concomitant to the novel coronavirus infection and, consequently, are more susceptible to the disease's complications. If the incidental diagnosis is made early and is of a relevant nature, it increases the likelihood of management and prompt treatment and, consequently, of better prognosis for the infected population.

The present study is subject to limitations, since it is a retrospective study with selection bias, because it analyzed a sample of outpatients and inpatients from a specialist cardiovascular service and the data collection period was relatively short. The CT reports were prepared by three different radiologists and there was no analysis by race or other subsets. Some cardiovascular findings were observed in scans with contrast, including ulcers and dissections, which could indicate that CTs without contrast are limited in this respect.

No analysis was conducted of correlations between cardiovascular findings and risk factors because the study is based on CT scans and reports that were requested because of diagnosis suggestive of COVID-19. Additional studies are ongoing at the institution with research protocols that include better stratification of outpatients and inpatients in order to obtain more trustworthy and explanatory results.

Incidental cardiovascular findings were observed in approximately half of the chest CTs of patients with suspected COVID-19; more specifically, aortic calcifications, cardiomegaly, and coronary calcification. 

Achados incidentais em exames de imagem são definidos como achados inesperados, não relacionados ao escopo da indicação clínica 1,2 . São fenômenos conhecidos e comuns em exames de tomografia computadorizada (TC) 3 . No entanto, o fator decisivo não é a ocorrência de um achado aleatório, mas a sua relevância clínica 1,3 . Na literatura pesquisada a respeito, observa-se que podem achados incidentais ser detectados em até 70% de todas as investigações por imagem e, felizmente, a maioria tem baixa relevância clínica 1-3 .

Esses achados podem ser classificados, de acordo com sua importância clínica, em maiores, moderados e menores 1 . Assim, as neoplasias recém-descobertas ou aneurismas são particularmente relevantes, cálculos biliares ou derrames pleurais podem ser considerados moderadamente relevantes, enquanto que cistos não complicados de fígado ou rim e calcificações vasculares simples são achados irrelevantes. A maioria dos achados incidentais são clinicamente não significativos 1,2 .

Os achados incidentais cardiovasculares compreendem distúrbios como aneurismas, calcificações das válvulas e artérias cardíacas, e tromboembolismos, que às vezes podem ser clinicamente assintomáticos e exigir mais investigação ou necessitar de intervenções terapêuticas, como é o caso de aneurismas aórticos gigantes [3] [4] [5] .

Atualmente, devido à emergência de saúde global enfrentada pela humanidade e à escassez de recursos e limitações de exames confirmatórios em larga escala para a doença do coronavírus 2019 (coronavirus disease, COVID-19), como a transcrição reversa seguida de reação em cadeia da polimerase (reverse transcription-polymerase chain reaction, RT-PCR) 6 , a TC de tórax passou a exercer importante função no auxílio diagnóstico, em virtude da sua alta sensibilidade na detecção de pneumonia viral, podendo colaborar potencialmente na avaliação de progressão da doença e no monitoramento da resposta à terapia 7, 8 .

Dessa forma, o aumento do uso de técnicas de imagem com acurácia satisfatória e a frequente disponibilidade das imagens tomográficas para o exame diagnóstico preferencial em caso de suspeita de infeção viral pelo COVID-19, na maioria das unidades de emergência, tem sido uma prática necessária. Outra vantagem da TC, além da investigação relacionada à queixa clínica, é a capacidade de identificar anormalidades incidentais, ou seja, achados não relacionados à indicação do estudo [6] [7] [8] [9] [10] .

Considerando essa discussão, com o presente estudo pretendemos avaliar a prevalência de achados cardiovasculares incidentais em TCs de tórax realizadas por suspeitas de COVID-19 em um hospital de referência no Pará. Os pacientes foram identificados por meio de um banco de dados radiográfico computadorizado que registra todos os estudos radiológicos realizados pelo departamento de radiologia. Foi possível incluir no estudo todas as TCs de tórax realizadas na instituição, oriundas dos diversos setores (unidades de internamento, pronto atendimento, ou de pacientes ambulatoriais), no período de 1° de março a 30 de julho de 2020, cuja indicação tenha relação com sintomas respiratórios da síndrome gripal ou suspeita de pneumonia por COVID-19. Foram excluídas TCs realizadas por outras indicações.

As TCs foram obtidas com um scanner multidetector Siemens Somatom Plus 16 (Siemens Medical Systems Inc, Iselin, Nova Jérsei, EUA), e os cortes tomográficos gravados em uma estação de trabalho (workstation -Vitria), visualizados com janelas para osso (C800 W2000) e/ou janela para pulmão (C40 W80), ambas adequadas à avaliação de tórax. Para a realização do exame, o paciente era posicionado em decúbito dosal, e os principais parâmetros de varredura foram os seguintes: tensão do tubo, 120 kVp; modulação automática da corrente do tubo, 30-70 mAs; passo, 0,99-1,22 mm; matriz, 512 × 512; espessura do corte, 1,0 mm; e campo de visão, 350 × 350 mm. Os exames foram laudados pelos três radiologistas titulados da instituição. Quando os indivíduos do estudo submeteram-se a mais de uma TC de tórax durante o acompanhamento, foi considerado apenas o resultado do primeiro exame.

O protocolo da pesquisa consistiu em 35 perguntas que envolveram três tópicos principais: aspectos sociodemográficos, achados cardiovasculares e distribuição dos achados incidentais.

As informações da caracterização amostral foram apuradas e sistematizadas em planilha elaborada no software Microsoft ® Office Excel ® 2016.

O dimensionamento amostral foi baseado na estimativa de uma prevalência de até 60% de achados cardiovasculares incidentais, erro padrão de 3,0% e nível alfa de 5%, resultando em uma amostra mínima de 1.024 exames.

Na aplicação da estatística descritiva, foram construídos tabelas e gráficos para apresentação dos resultados, e calculadas as medidas de posição, como média aritmética e desvio padrão.

A estatística analítica utilizada para avaliar os resultados das variáveis da amostra ocorreu por meio dos testes G e qui-quadrado de aderência para tabelas univariadas. As estatísticas descritiva e analítica foram realizadas pelo software BioEstat ® 5.4. Para a tomada de decisão, adotou-se o nível de significância α = 0,05, ou 5%, sinalizando com asterisco (*) os valores significantes.

No período de 1° de março a 30 de julho de 2020, foram realizadas 1.444 TCs de tórax no setor de radiologia do HCGV por suspeita de acometimento pulmonar por COVID-19.

A média (desvio padrão) de idade dos pacientes foi de 50,6±16,4 anos. O paciente mais jovem tinha 2 anos, e o de maior idade tinha 99 anos. Houve diferença estatisticamente significante (*p < 0,0001) entre as faixas etárias dos pesquisados, com maior proporção de pacientes entre 40 a 49 anos (25,7%).

Houve predominância do sexo feminino (738; 51,1%). No entanto, não houve diferença estatisticamente significante (p = 0,4146) entre os sexos.

A parte estatisticamente mais significativa das solicitações para realização do exame foi provenientes do pronto atendimento (48,8%), como mostra a Tabela 1.

A TC sem contraste foi o método mais comum (97,2%), realizado em proporção estatisticamente significante da amostra. Da mesma forma, achados tomográficos compatíveis com acometimento pulmonar pela COVID-19 ocorreram de forma significante (*p = 0,0002) em 56% da amostra, e a maioria estatisticamente significante (*p < 0,0001) apresentou opacidade em vidro fosco < 25% como padrão de distribuição da lesão pulmonar, conforme Tabela 2.

A análise dos achados tomográficos compatíveis com comprometimento pulmonar pela COVID-19 evidenciou um percentual estatisticamente significante de pacientes com opacidade em vidro fosco (56%). Achados adicionais compatíveis incluíram consolidação (18,4%), derrame pleural (12,6%) e bandas parenquimatosas (7,5%), características tomográficas sugestivas de COVID-19, entre outras identificadas nos laudos, conforme Tabela 3.

Achados pulmonares incidentais foram evidenciados em 63,2% das TCs de tórax, sendo o nódulo pulmonar o de maior proporção (18,3%) e estatisticamente mais significativo (*p < 0,0001) entre os demais. Achados adicionais incluíram espessamento peribrônquico (13,1%), faixas fibroatelectásicas (11,4%), atelectasia Em relação às afecções aórticas, houve 12 diagnósticos incidentais de aneurisma de aorta torácica (0,83%), sendo sete na aorta ascendente e cinco na aorta torácica descendente. Os 12 aneurismas acometeram 10 pacientes (dois do sexo masculino apresentaram dois aneurismas simultaneamente nas aorta ascendente e descendente).

O exame contrastado foi realizado em 41 pacientes (2,8% do total de exames), e favoreceu diagnósticos alterantivos de dissecção aórtica em 7,3% (três em 41 exames) e de aneurisma da aorta torácica em 9,7% (quatro em 41 exames). Houve 2,4% de achados de úlceras da aorta (dois em 41 exames). As Figuras 3 e 4 ilustram exemplos de afeccões aórticas incidentais evidenciadas em TCs de tórax.

A prevalência de tromboembolismo pulmonar (TEP) em angiotomografias de tórax foi de 24,4% (10 em 41 exames contrastados). Dos exames, 28 foram realizados em pacientes já hospitalizados e 13 em pacientes do pronto atendimento. O achado de TEP ocorreu em dois casos do pronto atendimento e em oito entre os hospitalizados.

O expressivo papel do diagnóstico por imagem levou a um aumento no número de exames de TC realizados e, juntamente com sua maior utilização, os radiologistas passaram a observar uma onda de achados não relacionados à indicação clínica específica, os chamados achados incidentais 1, 3, 4 .

O diagnóstico confirmatório de COVID-19 é baseado no RT-PCR em amostras nasais ou da região de orofaringe coletadas com um swab. Pacientes infectados pelo vírus SARS-Cov2 podem apresentar alterações na TC de tórax típicas da doença, como opacidades em vidro fosco. Porém, de acordo com as recomendações do Colégio Brasileiro de Radiologia, o exame deve ser realizado apenas em pacientes sintomáticos e hospitalizados 11 .

Achados incidentais são um fenômeno bem conhecido e comum em exames de imagem 1, 4, 7, 12 . Podem ser classificados, de acordo com sua importância clínica, em maiores, moderados e menores 12 . Os achados maiores incluem lesões suspeitas de doenças malignas, como nódulos da tireoide, alterações da espessura da parede intestinal e massas pancreáticas ou renais sólidas 12 . Achados moderados, isto é, possíveis achados clinicamente relevantes, compreendem cálculos biliares, coleção de fluido pleural ou derrame pleural. Pequenos achados ou achados sem relevância clínica incluem cistos renais ou hepáticos simples, alterações degenerativas na coluna vertebral e calcificação de vasos. Muitos achados incidentais são clinicamente não significativos 1, 2, 4, 7 .

A presente pesquisa evidenciou predomínio de solicitações envolvendo pacientes do sexo feminino e que buscaram assistência no pronto atendimento da instituição, fonte de 48,8% das requisições de exames.

No HCGV, o protocolo usual é o de TC de tórax com espessura de corte de 1,0 mm, baixa dose de radiação e sem contraste endovenoso A grande maioria das TCs de tórax (97,2%) foi realizada sem contraste, sendo reservado o uso de contraste nas particularidades de suspeita de TEP ou de outras afecções vasculares que demandassem melhor análise.

Quanto à origem das solicitações de TCs de tórax, a maioria teve origem no serviço de pronto atendimento (48,8%). Essa expressiva demanda de pacientes externos, em detrimento dos internados em enfermaria ou unidade de terapia intensiva, deve-se provavelmente às características do serviço, que conta com uma urgência porta aberta para pacientes cardiovasculares e nefrológicos.

A TC, muito embora não seja indicada como exame de escolha isolado para o diagnóstico pelas diversas sociedades de especialidade, tornou-se valiosa ferramenta de auxílio diagnóstico nesses pacientes, além de ser útil no acompanhamento da evolução e na detecção de possíveis complicações 12, 13 . Assim, 56% das TCs de tórax solicitadas foram compatíveis com afecção pulmonar pela COVID-19. Os achados mais característicos foram opacidade em vidro fosco e consolidação; no entanto, diversos outros achados podem sugerir afecção pulmonar, dependendo da fase e do tempo de evolução da doença 13 .

Entre as TCs que foram solicitadas para diagnóstico presuntivo de COVID-19, 63,2% tiveram achados incidentais, e a maioria deles consistiram em nódulos pulmonares (15,8%), espessamentos peribrônquicos (13,1%), faixas fibroatelectásicas (11,4%) e atelectasia (9,5%). Entre os achados pulmonares, os nódulos apresentaram significância estatística de p < 0,0001 entre os demais. Nódulos pequenos, acredita-se, não apresentam relevância clínica; no entanto, oito pacientes apresentaram achados de massa pulmonar superior a 3 cm (0,5% da amostra). Isso reforça a hipótese de que a grande maioria dos achados incidentais são irrelevantes; porém, o método também permite detectar lesões assintomáticas que necessitam de melhor gerenciamento 11 . Assim, embora opacidade em vidro fosco bilateral e consolidação tenham sido relatadas como as características predominantes dos exames de imagem por COVID-19, as manifestações da TC no tórax podem variar em diferentes pacientes e estágios 11 .

Achados cardiovasculares incidentais foram evidentes em 51,2% (IC 48,7-53,8%) dos exames tomográficos realizados, e mais de 40% apresentaram mais de um achado. Calcificação das paredes aórtica e coronária, assim como cardiomegalia, foram os mais evidentes em laudos radiológicos. Achados menos comuns envolveram derrame pericárdico e dilatação da raiz aórtica. Esses achados podem ser justificados devido ao perfil epidemiológico apresentado no estudo, no qual aproximadamente 30% dos sujeitos pesquisados têm mais de 60 anos, além do próprio perfil dos pacientes atendidos no hospital, que possui foco cardiovascular e nefrológico.

Na tentativa de homogeneizar o grupo, foi realizada a análise em subgrupo, com a exclusão de 145 exames de indivíduos com menos de 30 anos, perfazendo um total de 1.299 TCs e evidenciando distribuição similar, com predomínio de calcificação coronária e da parede aórtica, cardiomegalia e derrame pericárdico. Portanto, a detecção precoce, ainda que incidental, de doenças com potencial clínico relevante, como calcificação coronária, aneurismas ou dissecções aórticas, pode mudar o prognóstico nessa população e impactar positivamente na redução da mortalidade e no aumento da qualidade de vida.

Surov et al. 2 mostraram que os achados cardiovasculares podem ser identificados em 6,8% dos pacientes com doenças malignas investigadas por TC. Por outro lado, Jacobs et al. 4 relataram em seu estudo que a frequência de aneurisma aórtico variou de 0,07% a 3,4%, enquanto que a frequência de dissecção aórtica variou de 0,06% a 0,2% 4 .

Embora os achados cardíacos incidentais possam não ser relevantes para o manejo clínico imediato de pacientes com suspeita de afecção pulmonar pela COVID-19, eles podem influenciar o manejo clínico a longo prazo e favorecer melhores prognósticos. Por exemplo, calcificações coronárias e da parede aórtica são sabidamente marcadoras de aterosclerose e sinais subjacentes de doença cardiovascular, muitas vezes subclínica, podendo influenciar na prevenção primária de eventos ateroscleróticos [14] [15] [16] [17] . Por outro lado, a cardiomegalia representa uma característica tardia da disfunção ventricular esquerda e da insuficiência cardíaca, ambas com prognóstico desfavorável, podendo desencadear mais investigações e tratamento cardíaco mais efetivo 14 .

Vários estudos têm mostrado que placas ateroscleróticas, calcificações e irregularidades da parede aórtica são muito prevalentes em pacientes com doença cardiovascular [15] [16] [17] . Outros demonstraram que a calcificação da aorta descendente está relacionada à calcificação das artérias coronárias, importante preditor de doença cardiovascular. No entanto, até o momento, não há estudo de acompanhamento de investigações do valor prognóstico dessas anormalidades 16, 17 .

Sverzellati et al. 18 relataram que 50% dos 286 exames de TC realizados para fibrose pulmonar, suspeita de embolia pulmonar ou estadiamento do câncer de pulmão tiveram achados cardiovasculares potencialmente significativos. Da mesma forma, Choy et al. 19 demonstram que 61% dos exames consecutivos de TC de tórax de rotina apresentaram achados cardíacos relatáveis.

Estudos prévios sobre COVID-19 apontam que 63 a 67% dos pacientes que evoluíram a óbito sofriam de comorbidades cardiovasculares, mais comumente hipertensão, diabetes e doença cardíaca coronária [20] [21] [22] , fatores estes que podem estar relacionados a doenças vasculares [19] [20] [21] [22] .

Asim, as afecções aórticas foram detectadas incidentalmente em TC de tórax em até 2,4% dos casos, sendo aneurisma a anormalidade mais prevalente, especialmente com a TC sem contraste. Na ausência do uso de contraste intravenoso, dissecções aórticas e ulcerações significativas geralmente são indetectáveis, podendo caracterizar um fator limitante do método 4 . No presente estudo, apenas 2,8% dos exames tomográficos foram realizados com uso de contraste. Aneurismas e dissecções foram observadas respectivamente em 9,7% e 7,3% dos exames contrastados e sugeridas pelos radiologistas em laudo de 1,2% dos exames sem uso de contraste endovenoso. Por outro lado, TEP foi observado em 24,4% dos pacientes que realizaram exame contrastado, representando uma significativa amostra de casos internados em unidade de terapia intensiva, onde predominam os casos de maior gravidade de COVID-19.

Assim, é importante salientar que os achados incidentais configuram-se como um evento importante para o desfecho clínico dos pacientes, visto que boa parte destes possuem comorbidades concomitantes à infecção pelo novo coronavírus e, por conseguinte, tornam-se mais suscetíveis a complicações por essa doença. O diagnóstico incidental, se realizado precocemente e tiver característica relevante, proporciona maiores chances de gerenciamento e tratamento oportuno, e também, por conseguinte, melhores prognósticos à população infectada.

A presente pesquisa apresenta limitações, pois trata-se de estudo retrospectivo e com viés de seleção, visto que analisou uma amostra de pacientes ambulatoriais e hospitalizados em um serviço de referência cardiovascular e em um período de coleta de dados relativamente curto. As TCs foram avaliadas por três radiologistas e não houve, na amostra coletada, caracterização racial ou análise de subgrupos. Alguns achados cardiovasculares foram evidenciados em exames contrastados, como úlceras e dissecções, o que pode limitar exames tomográficos não contrastados.

Não foi realizada a correlação entre os achados cardiovasculares e fatores de risco, pois a pesquisa foi baseada em laudos e exames tomográficos cuja indicação fosse diagnóstico sugestivo de COVID-19. Estudos futuros estão sendo conduzidos na instituição e consideram, na aplicação do protocolo, uma melhor estratificação entre casos ambulatoriais e internados, para resultados mais fidedignos e esclarecedores.

Achados cardiovasculares incidentais ocorreram em aproximadamente metade das TCs de tórax de pacientes com suspeita de COVID-19, mais especificamente, calcificações de aorta, cardiomegalia e calcificação coronária.

Incidental findings in imaging diagnostic tests: a systematic review. Br J <jrn>1. Lumbreras B, Donat L, Hernandez-Aguado I. Incidental findings in imaging diagnostic tests: a systematic review

Clinically relevant cardiovascular findings detected on staging computed tomography in oncological patients

Thrombotic events as incidental finding on computed tomography in intensive care unit patients

Prevalence of incidental findings in computed tomographic screening of the chest: a systematic review

Whole-body CT screening: spectrum of findings and recommendations in 1192 patients

Findings in 2019 Novel Coronavirus (2019-nCoV) Infections from Wuhan, China: Key Points for the Radiologist

Incidental findings in imaging research: evaluating incidence, benefit, and burden

Incidental breast lesions detected on CT: what is their significance?

Clinical significance of incidental thyroid nodules identified on low-dose CT for lung cancer screening

Prevalence and clinical importance of aortic valve Incidental cardiovascular findings on tomography

calcification detected incidentally on CT scans: comparison with echocardiography

A importância da Tomografia Computadorizada no diagnóstico da COVID-19

Incidental findings on knee radiographs in children and adolescents

Achados da COVID-19 identificados na tomografia computadorizada de tórax: ensaio pictórico

Incidental cardiac findings on computed tomography imaging of the thorax

Relationship between coronary artery and descending thoracic aortic calcification as detected by computed tomography: the Multi-Ethnic Study of Atherosclerosis

CT in coronavirus disease 2019 (COVID-19): a systematic review of chest CT findings in 4410 adult patients

Relationships of thoracic aortic wall calcification to cardiovascular risk factors: the Multi-Ethnic Study of Atherosclerosis (MESA)

Under-reporting of cardiovascular findings on chest CT

Pertinent reportable incidental cardiac findings on chest CT without electrocardiography gating: review of 268 consecutive cases

PMid:29941240. REFERÊNCIAS 1. Lumbreras B, Donat L, Hernandez-Aguado I. Incidental findings in imaging diagnostic tests: a systematic review

Clinically relevant cardiovascular findings detected on staging computed tomography in oncological patients

Thrombotic events as incidental finding on computed tomography in intensive care unit patients

Prevalence of incidental findings in computed tomographic screening of the chest: a systematic review

Whole-body CT screening: spectrum of findings and recommendations in 1192 patients

Findings in 2019 Novel Coronavirus (2019-nCoV) Infections from Wuhan, China: Key Points for the Radiologist

Incidental findings in imaging research: evaluating incidence, benefit, and burden

Incidental breast lesions detected on CT: what is their significance?

Clinical significance of incidental thyroid nodules identified on low-dose CT for lung cancer screening

Prevalence and clinical importance of aortic valve calcification detected incidentally on CT scans: comparison with echocardiography

A importância da Tomografia Computadorizada no diagnóstico da COVID-19

Incidental findings on knee radiographs in children and adolescents

Achados da COVID-19 identificados na tomografia computadorizada de tórax: ensaio pictórico

Incidental cardiac findings on computed tomography imaging of the thorax

Relationship between coronary artery and descending thoracic aortic calcification as detected by computed tomography: the Multi-Ethnic Study of Atherosclerosis

CT in coronavirus disease 2019 (COVID-19): a systematic review of chest CT findings in 4410 adult patients

Relationships of thoracic aortic wall calcification to cardiovascular risk factors: the Multi-Ethnic Study of Atherosclerosis (MESA)

Under-reporting of cardiovascular findings on chest CT

Pertinent reportable incidental cardiac findings on chest CT without electrocardiography gating: review of 268 consecutive cases

Managing Incidental Findings on Thoracic CT: Mediastinal and Cardiovascular Findings. A White Paper of the ACR Incidental Findings Committee

Clinical course and risk factors for mortality of adult inpa-tients with COVID-19 in Wuhan, China: a retrospective cohort study

Clinical characteristics of 113 deceased patients with coronavirus disease 2019: retrospective study

CEP 66093-040 -Belém (PA), Brasil Tel.: (91)

macielreis.angiovasc@gmail.com Informações sobre os autores JMCR -Membro Titular

Hospital Beneficente Portuguesa de São Paulo; Coordenador; Cirurgia Vascular

Programa de Residência em Cirurgia Geral, Hospital de Clínicas Gaspar Vianna (HCGV)

MCA Responsabilidade geral pelo estudo: JMCR *Todos os autores leram e aprovaram a versão final submetida do J Vasc Bras