Introduction

Nearly 17 million people have been infected with the coronavirus disease (COVID-19) as of July 29, 2020, and more than 660,000 people have died [1]. The most common COVID-19 symptoms include fever (91%), cough (67%), fatigue (51%), and dyspnea (30%) [2]. However, gastrointestinal (GI) symptoms have been increasingly recognized in these patients [3,4,5,6] and were likely under-reported in early studies [7]. Some COVID-19 patients have presented only with GI symptoms at initial presentation, or even in the complete absence of respiratory symptoms in some cases [8, 9]. While there is strong evidence that the digestive system can be infected by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) [10, 11], there has been less focus on the effects of SARS-CoV-2 on extra-pulmonary systems. Few studies have identified the potential abdominal imaging features of these symptoms [12]. This scoping review aims to discuss the current literature on the imaging features in the abdomen in COVID-19, as well as the relevant clinical manifestations of these image findings.

Methods

This scoping review used the 22-item Preferred Reporting Items for Systematic Review and Meta-Analysis – Scoping Review (PRISMA-ScR) for guidance [13, 14]. A pre-registered protocol was not submitted prior to review.

Literature Search and Selection Criteria

MEDLINE, EMBASE, Scopus, and Web of Science databases were searched for eligible publications from inception to July 15, 2020. Variations of “COVID-19” AND “GASTROINTESTINAL” AND “IMAGING”-related title/abstract/keywords or medical subject heading terms were performed with individual search strategies outlined in Appendix 1–4. No language restrictions were applied and language translation was performed when required. Search results were combined from each database and duplicates were removed using RefWorks. Titles and abstracts were screened for relevance, and full-text review for potentially relevant articles was then performed. To be included in the review, papers needed to address abdominal imaging findings in COVID-19-positive patients. Studies involving animal subjects were excluded. Publications from inception to March 31, 2020, were reviewed by MPW, while articles published between March 31 and July 15, 2020 were reviewed by KL. Given the broad scope of this review, a charting form was not established at the start, but one was developed after full-text review. Publications of unclear relevance were resolved by discussion and consensus between KL and MPW. Reference lists from included studies were screened to identify additional articles of interest. Forward searching of included studies was performed in Google Scholar. Ethics approval was not necessary for this review.

Results

A total of 469 articles were identified from the database search after duplicate removal. After title and abstract screening, 137 articles were identified for full-text review. In addition to studies found from included article reference lists and manual searches, 36 articles were included in qualitative synthesis. The PRISMA flow diagram is shown in Fig. 1.

Fig. 1
figure 1

PRISMA flow diagram of the included studies on abdominal imaging findings in patients with SARS-CoV-2 infection. Flow diagram adapted from [14]

Full size image

Incidental COVID-19 findings based on abdominal computed tomography (CT) indicated by gastrointestinal symptoms

A wide range of GI symptoms have been reported in patients with COVID-19. A multicenter study by Pan et al. found that more than 50% of their 204 patients reported digestive symptoms including lack of appetite, diarrhea, vomiting, and abdominal pain [15]. A recent meta-analysis of 4234 patients found the prevalence of GI symptoms to be 17.6% [16]. Three case reports included hematochezia as a symptom [17,18,19]. There are also several case reports that describe patients that present with only GI symptoms initially or throughout the course of their entire infection [20,21,22,23,24]. The largest study, a single-center study by Luo et al. found that of 1141 confirmed cases of COVID-19, 16% presented with only GI symptoms [25]. Additionally, many cases of COVID-19 have been found incidentally through abdominal imaging scans indicated by patients presenting mainly or only with GI symptoms [20, 21, 26,27,28,29]. Hossain et al found that more than 50% of their 119 patients first demonstrated evidence of COVID-19 incidentally through either an abdominal or cervical spine/neck CT [30]. Some authors now recommend radiologists to carefully review abdominal CT scans for typical COVID-19 findings at the lung bases [31, 32].

Abnormal abdominal imaging findings

Twenty-one studies have now investigated abnormal abdominal imaging findings in patients tested positive for SARS-CoV-2. A summary of findings are shown in Table 1. Ten studies of 23 patients (2 of which are pediatric cases) demonstrate thickening of various regions of the small and large bowel wall [17, 18, 28, 33,34,35,36,37,38,39]. Hyperemia and mesenteric thickening have also been observed in tandem with bowel wall thickening. Different mechanisms have been postulated for the cause of thickening in these cases. Such causes include the SARS-CoV-2 infection of bowel epithelia leading to cytokine storm and inflammation, edema, or ischemia indicated by the presence of fibrin clots.

Table 1 Characteristics and findings of included studies of patients with abnormal abdominal radiologic features
Full size table

Pneumatosis or portal vein gas has been reported on abdominal CT in 6 patients [33, 40, 41]. An additional 3 patients were reported to have pneumatosis though the diagnostic modality used is not reported [42]. The presence of pneumatosis was reported in various regions of the GI tract, including the jejunum, ileum, caecum, and right colon.

There has been one case report of abdominal CT demonstrating extensive pneumoperitoneum with free intracavitary fluid [43]. Subsequent laparotomy revealed the cause to be perforation of the sigmoid colon accompanied by reduced perfusion of the entire GI tract. The authors suggest this perforation to be the result of septic and thromboembolic phenomena, caused by the viral infection.

A rare ultrasound finding reported in pediatric cases of SARS-CoV-2 infection is ileocolic intussusception. Three separate cases have reported this imaging finding [44,45,46]. Another two SARS-CoV-2-positive patients have been reported to have intussusception, though it is not known if their condition was identified through imaging modalities [47, 48].

One case report involved a patient with colonic ileus and air in bowel wall on abdominal CT [35]. Behzad et al similarly reported a patient that showed ileus pattern on abdominal radiograph, along with nonspecific stranding surrounding mildly distended fluid-filled ascending colon [49]. Lastly, there has been one case of ascites found on ultrasound in a SARS-CoV-2 patient [50]. Drainage yielded 6L of fluid, and the fluid sample was tested positive for SARS-CoV-2 RNA. The authors suggest that ascites infection may involve infection of ACE2 expressing cells of the esophageal epithelial cells and absorptive enterocytes [3].

Discussion

This scoping review identified 36 primary studies addressing the GI symptoms and radiologic manifestations of SARS-CoV-2 infection of the GI system. On imaging, typical GI findings have included nonspecific small and large bowel wall thickening and liquid stool throughout the bowel. Other more rare presentations such as pneumatosis intestinalis, pneumoperitoneum, and large volume ascites have also been reported. Additionally, the more typically described lung parenchymal findings have also been seen incidentally in the periphery of cross-sectional abdominal imaging. This review highlights features which can cue imaging experts to consider COVID-19 and evaluate for other features which may lead to a COVID-19 diagnosis, particularly in patients with few or no typical COVID-19 symptoms.

This scoping review is subject to some limitations. Given the broad scope of this review and the lack of detailed exclusion criteria, some imaging features may have been potentially missed. Additionally, this review may have missed relevant studies by not utilizing other search databases and exclusion of the gray literature. Furthermore, the lack of early recognition of GI symptoms [7] may limit the available literature in terms of imaging investigations indicated by GI symptoms. Investigating abdominal imaging abnormalities in COVID-19 cases presenting with both respiratory and GI symptoms may be a topic for future investigations.

Conclusion

Gastrointestinal manifestations of COVID-19 have been increasingly recognized. Although rare, imaging features have included bowel wall thickening, fluid-filled colon, pneumatosis, pneumoperitoneum, intussusception, and ascites. Identifying these features on abdominal imaging highlight the need to consider and evaluate for other manifestations of COVID-19 such as lung parenchymal findings. In the appropriate setting, COVID-19 can be raised as a consideration, particularly in patients presenting with atypical symptoms.