The coronavirus disease-2019 (COVID-19), the clinical name of the new severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) infection that originated in the city of Wuhan, province of Hubei in China, in December 2019, was declared a pandemic by the World Health Organization on March 11, 2020.(1)

At the beginning of May 2020, the Centers for Disease Control and Prevention (CDC) in the United States recognized the presence of "gastrointestinal symptoms, like nausea, vomiting and diarrhea, as less common symptoms" of COVID-19 after cough, shortness of breath or difficulty breathing, fever, chills, muscle pain, sore throat, new loss of taste or smell.(2) At the time, there were already several reports from China,(3)(4) Singapore(5) and Europe(6) that had noted the presence of gastrointestinal symptoms (GI) in patients with COVID-19. These digestive manifestations were later confirmed in the United States(7) and more recently in Mexico.(8) At least four papers have reviewed the literature concerning the presence of GI symptoms in patients with COVID-199-12 (see Table 1).

The first review among 2023 patients found GI symptom incidence varied from 3% to 79%.(9) The second review of fifteen articles with 2,800 patients found a frequency of GI symptoms that ranged from 3.0% to 39.6%, of which diarrhea was the most common symptom.(10) A third review and meta-analysis (29 studies including 6064 patients with reported GI symptoms in patients with COVID-19) calculated a pooled prevalence of digestive symptoms of 15% (95% CI: 10-21; range: 2–57%; I²=96%).(11) Among them nausea or vomiting, diarrhea and loss of appetite were the three most common symptoms.(11) Another review and meta-analysis of 4243 patients that included 53 studies from China, two from South Korea, two from Singapore, one from Vietnam, one from the U.S. and one from the United Kingdom reported a pooled prevalence of all GI symptoms of 17.6% (95% CI: 12.3%. 24.5%).(12) There was a higher prevalence of GI symptoms in patients with severe vs. non-severe COVID-19 illness, 11.8% (95% CI: 4.1%-29.1%) vs. 17.1% (95% CI: 6.9%-36.7%), respectively.(12)

It is interesting to note that besides diarrhea, nausea, vomiting, abdominal pain, bloating, burping and anorexia, gastrointestinal bleeding has also been reported in some patients. Upper GI bleeding has been related to herpetic erosions and ulcers,(13) while hematochezia has been present in some patients without any identified source, or due to hemorrhagic colitis with patchy erythema, circumferential wall thickening, mesenteric hypervascularity and pericolic fat stranding (appearance of edema within fat).(14)(15)

The majority of the above analyzed studies included hospitalized patients. However, a study from outpatient settings in New York of patients with SARS-CoV-2 confirmed by PCR of nasopharyngeal swabs found a GI symptom prevalence of 35%. Also, the presence of GI symptoms was associated with a 70% increased risk of testing positive for COVID-19 infection (adjusted OR: 1.7, 95% CI: 1.1-2.5).(16)

GI symptoms are also present in pediatric patients with COVID-19. For example, in the systematic review and meta-analysis by Mo et al., pediatric patients had a similar prevalence of Gl symptoms to those of adult patients.(11) In 10% of the children with COVID-19, GI symptoms were the only ones without respiratory manifestations, and patients with GI symptoms had delayed diagnosis of 2.85 days (95% CI: 0.22–5.48) and had a higher prevalence of complications.(11) In a systematic review and meta-analysis of nine case-series, 75% of the children had household contacts; and while fever and cough were the most common manifestations, GI symptoms were present in 12% of the children.(17)

In addition, it is important to mention that liver function abnormalities are also common in COVID-19 patients,(11) primarily manifested as moderate elevations of serum aminotransferases levels.(18) In a study among 417 patients with COVID-19, 76.3% had abnormal liver test results and 21.5% had liver injury during hospitalization.(19) Also, these patients were at higher risk of progressing to severe disease.(19) However, the exact cause of the liver damage has not been elucidated and there is insufficient evidence for direct viral liver injury.(20)(21) Several reasons have been proposed for the hepatic abnormalities in COVID-19, including viral infection per se, use of potentially hepatotoxic drugs, systemic inflammatory response, hypoxia and multiple organ dysfunction.(18) The effect of COVID-19 on the liver is beyond the scope of this paper and will not be further addressed.

GI Symptoms As Atypical Manifestation of COVID-19

There is evidence that GI symptoms can also be atypical manifestations of COVID-19. For example, Pan L. et al. reported that 7% of their 204 patients in the province of Hubei in China presented with digestive symptoms only.(4) Of this group, 48.5% arrived at the hospital with GI symptoms as their chief complaint and in seven cases they were the only manifestations, that is, these seven patients did not have respiratory symptoms. Patients with digestive symptoms had a significantly longer time interval from onset to admission compared to those with no digestive symptoms (almost two days longer).(4) In another study from the province of Hubei, with more than half of the patients from the city of Wuhan, Liu K. et al. found a prevalence of diarrhea of 8%. Interestingly, 20% of their patients did not develop fever, and in some patients, digestive symptoms were among the initial manifestations.(22)(23) In a series from the city of Zhuhai, in the province of Guandong, more than 61% of their 95 patients presented with GI symptoms, and 81% developed these symptoms during the course of hospitalization.

Furthermore, 11.6% of the patients did not have imaging features of COVID-19 pneumonia, only GI manifestations.(13) In a study on 651 cases from the province of Zhejiang, 28.4% of their patients lacked respiratory symptoms and only presented with GI symptoms such as nausea, vomiting and diarrhea. The median duration of the diarrhea was four days, ranging from one to nine days, and was self-limited in the majority of cases.(24)24 In the city of Veracruz, in the south of Mexico, among 112 patients who consulted a private hospital, 20.5% reported at least one GI symptom, of which the most common ones were diarrhea (17.8%), abdominal pain (9.8%), and vomiting (7.1%). Similarly, in 6.2% of the cases, these were the only clinical manifestations.(8)

GI Symptoms and Severity of COVID-19

In a study from the city of Wuhan, the frequency of GI symptoms was compared between patients hospitalized or not in the ICU. Only abdominal pain was more common in patients in the ICU (8.3% vs. 0.0%, p=0.02).25 Another study from the city of Zhuhai found no differences in hospital stay, discharge or mortality in patients with or without GI symptoms.(13) In the first study from Singapore of 18 hospitalized patients, none of the three patients with diarrhea required supplemental oxygen.(5) In the study from outpatients settings in New York, there was a trend toward lower rate of ICU admission, and a significantly lower rate of death in patients with vs. without GI symptoms (0.0% vs. 5.0%, p=0.03), during a short-term follow-up.(16)

The last three case series described above possibly suggest a milder outcome in patients with digestive symptoms, such as abdominal pain and diarrhea.(25)(26) In contrast, to the previous studies, Pan et al. reported that patients with digestive symptoms were half as likely to be cured and discharged than the 60% without GI symptoms.(4) Also, Jin et al. in Zhejiang, China reported that the presence of acute respiratory distress syndrome and the need for mechanical ventilation were significantly more frequent in patients with GI symptoms compared to those without.(24) A large study from Chile of 7,016 patients who tested positive for SARS-CoV-2, of whom 7.3% had diarrhea and 3.7% abdominal pain, diarrhea was associated with a higher risk of hospitalization (RR: 1.31, 95% CI: 1.1-1.57; p<0.01) but abdominal pain was not.(27)

Others were unable to discriminate the severity of COVID-19 according to the presence or not of GI symptoms.(3)(28) Notwithstanding, in a pooled analysis of 10 studies including 1,989 patients with COVID-19, of whom 30.1% were classified as severe disease, abdominal pain was found to be associated with a nearly four-fold increased odds of severe illness, with a marginally significant increased prevalence for nausea and vomiting, but not for diarrhea.(29)

Potential Underlying Mechanism for GI symptoms in COVID-19

To infect the human cells, the SARS-CoV-2 requires binding to the angiotensin-converting enzyme 2 (ACE2) receptor.(30) This process begins by the interaction of the spike glycoprotein (S) and the ACE2. Following this engagement, SARS-CoV-2 S is processed by a plasma membrane-associated transmembrane serine proteases, TMPRSS2 and TMPRSS4, promoting the virus entry into the host cells.(31)(32) Although ACE2 receptors have been identified in type II alveolar cells of the lung, myocardial cells, kidney proximal tubule cells and bladder urothelial cells,(33)(34) they are also present in cholangiocytes and in the digestive tract.(35) In the GI tract, they are mainly localized in glandular cells of the stomach, duodenum and rectum, and are rarely identified in the esophagus probably because of its predominance of squamous epithelial cells.(36) As ACE2 receptors have been shown to be expressed in the oral cavity mucosa and are highly enriched in the epithelial cells of the tongue, the oral cavity and digestive tract may be a route of infection of this novel virus.(33)(34)(37)(38) Furthermore, viral replication has been demonstrated in enteric cells(39) and SARS-CoV-2 RNA has been detected in the esophagus, stomach, duodenum and rectum.(13)

There is also evidence that the SARS-CoV-2 infection in patients without inflammatory bowel disease (IBD) triggers an inflammatory response in the gut. This response is not only clinically manifested (e.g., diarrhea), but also presents with intestinal infiltration of neutrophils, macrophages and T cells, elevated fecal calprotectin and a systemic IL-6 response.(40) The increased level of fecal calprotectin was shown to be significantly higher in COVID-19 patients with ongoing diarrhea and in those in whom the diarrhea had ceased compared to patients without diarrhea.(40) This inflammatory response may be responsible for the presence of the round herpetic erosions and ulcers in the esophagus that have been found in some patients with GI bleeding,(13) or aphthous lesions in the terminal ileum that mimic Crohn's disease.(41)

SARS-CoV-2 RT-PCR in Feces

SARS-Cov-2 viral nucleic acids have not only been uncovered in respiratory samples, but are also present in saliva and stools.(5)(42) For example, Zhang JC, et al. reported that the nucleic acid detection of COVID"19 in fecal specimens was as accurate as pharyngeal swab specimens, despite the fact that patients with a positive stool test did not necessarily experience GI symptoms.(43) In the Singapore case series, the virus was detected in the stools of 4 out of 8 patients that were tested, regardless of the presence of diarrhea;5 and in another study, the presence of viral RNA in feces was not related to the presence or severity of GI symptoms.(13) More importantly, a group from China found that viral RNA was still present in the stools of more than 20% of the patients infected with SARS-Cov-2 that had a negative conversion of the viral RNA in the respiratory tract.(35)

In a study of 206 patients with mild COVID-19 disease, 117 (56.7%) presented digestive symptoms, with or without respiratory symptoms. Those with digestive symptoms were more likely to be fecal virus positive (73.3% vs. 14.3%, p=0.033) vs. those with only respiratory symptoms, and patients who were positive for viral RNA in stools had longer time to viral clearance compared to those who were negative (44.2 vs. 33.7 days, p=0.003).(44) In addition, there have been reports of viral viability including survival up to two weeks in a maintenance medium containing the virus dried at room temperature.(42)(45) Extended viral viability, together with the presence of virus in the feces, support the possibility of fecal-oral transmission, especially in areas with poor sanitation.(42)(45)

Stool Testing

Despite the reports of fecal shedding of viral RNA, there are, currently, no guidelines for stool testing in patients suspected of being infected with COVID-19. However, considering the high prevalence of positive stools in a group of patients, some experts advocate for routine RT-PCR testing of feces, especially in patients presenting with digestive symptoms.(44) Multi-sample testing may potentially improve diagnostic accuracy. One study reported, for example, positivity of nasopharyngeal swab at 38.13%, sputum at 48.68%, blood 3.03%, feces 9.83% and anal swabs at 10.0%.(46)

Another important issue to consider is that wastewater surveillance may provide epidemiological information about the viral circulation in a specific community.(47) Accordingly, in the region of Murcia in Spain, surveillance of titers of SARS-CoV-2 in the wastewaters determined that citizens were shedding virus even before the first cases were reported.(48) Therefore, wastewater-based epidemiology, by monitoring viral RNA, can be used to assess disease prevalence and spread in defined populations, providing important information for implementing public health policies.(49)

Preventive Measures

Aerosolization of the virus may be a risk during upper GI endoscopic procedures or breath tests.(45)(50) This, together with the possibility of viral shedding in the feces, prompts the need for preventive measures in endoscopy units, as well as in motility labs during the performance of colonoscopies or anorectal manometries and biofeedback, respectively.(51)(52)(53) These measures should be taken to protect healthcare personnel against contracting infections while performing gastrointestinal endoscopic and/or manometric procedures and breath testing. However, they are beyond the scope of the current review but can be summarized as delaying elective procedures up to eight weeks or until the pandemic declines, implementing a pre-procedural triage and using personal protective equipment (PPE) if there is a time-sensitive endoscopic procedure that needs to be performed.(54)

In addition, toilet flushing can produce aerosols, resulting in fomite transmission,(55) with droplet nuclei that are small enough to both contaminate surfaces and become airborne.(50)50 Apparently, the SARS-CoV-2 can live in aerosols for three hours.(56) In China, the presence of GI symptoms was related to family clustering.(24) Therefore, toilet isolation measures should be implemented in households with positive cases of COVID-19.(10) Fecal shedding could also be a mode of transmission in hospital settings or public places, including schools, stadiums or theaters. It is important to underscore that fecal shedding can occur even in asymptomatic subjects or patients with mild symptoms and delayed diagnosis.(44) Therefore, until research studies clarify whether toilet flushing produces bio-aerosolized and viable SARS-CoV-2, it is recommended to use lids on toilettes(50) and to close them before flushing. According to the author, educational campaigns in this regard, as well as on handwashing, are needed, and include pasting instruction posters in public restrooms.

Treatment of GI symptoms In Patients With COVID-19

There are no treatments that have been tested in patients with COVID-19 and GI symptoms. In patients with diarrhea, dehydration is possible and, therefore, fluid replacement is important. In the experience of the author caring for outpatients with COVID-19 and digestive symptoms, symptomatic treatment is useful. Such treatments can include antispasmodics for abdominal pain and probiotics in patients with diarrhea. Still, the most important issue is to be aware that GI symptoms can be a manifestation of COVID-19 even in the absence of fever or respiratory symptoms, and the healthcare professional must consider ruling out this viral infection, especially in patients with novel GI symptoms and in whom other causes are excluded.(10)

Conclusions

It is now well demonstrated that patients with COVID-19 may present with GI symptoms as the initial manifestation of this pandemic or during the course of the illness. In some cases, the GI symptoms can be the sole manifestation of the viral infection, thus considered atypical COVID-19 cases.

The frequency of the GI symptoms varies widely and include diarrhea, nausea, vomiting, abdominal pain, bloating, burping/reflux and even GI bleeding. There is controversy as to whether these digestive symptoms are related to the illness prognosis, although a pooled analysis found a relationship between abdominal pain and severe disease.

The possible explanation for the GI symptoms is the need of SARS-CoV-2 virus to bind to the ACE2 receptor to infect the human cells; a receptor whose largest expression is in the enteric epithelial cells. In addition, fecal viral shedding has been well demonstrated, hence there is a possibility of fecal-oral transmission. More studies are warranted to determine the potential of fecal testing in the diagnosis of COVID-19, especially in those who have negative pharyngeal swabs. In the meantime, bathroom isolation for infected household contacts and preventive measures in public bathrooms are recommended.