key: cord-0818736-2d6htc3s
authors: Elmunzer, B. Joseph; Spitzer, Rebecca L.; Foster, Lydia D.; Merchant, Ambreen A.; Howard, Eric F.; Patel, Vaishali A.; West, Mary K.; Qayed, Emad; Nustas, Rosemary; Zakaria, Ali; Piper, Marc S.; Taylor, Jason R.; Jaza, Lujain; Forbes, Nauzer; Chau, Millie; Lara, Luis F.; Papachristou, Georgios I.; Volk, Michael L.; Hilson, Liam G.; Zhou, Selena; Kushnir, Vladimir M.; Lenyo, Alexandria M.; McLeod, Caroline G.; Amin, Sunil; Kuftinec, Gabriela N.; Yadav, Dhiraj; Fox, Charlie; Kolb, Jennifer M.; Pawa, Swati; Pawa, Rishi; Canakis, Andrew; Huang, Christopher; Jamil, Laith H.; Aneese, Andrew M.; Glamour, Benita K.; Smith, Zachary L.; Hanley, Katherine A.; Wood, Jordan; Patel, Harsh K.; Shah, Janak N.; Agarunov, Emil; Sethi, Amrita; Fogel, Evan L.; McNulty, Gail; Haseeb, Abdul; Trieu, Judy A.; Dixon, Rebekah E.; Yang, Jeong Yun; Mendelsohn, Robin B.; Calo, Delia; Aroniadis, Olga C.; LaComb, Joseph F.; Scheiman, James M.; Sauer, Bryan G.; Dang, Duyen T.; Piraka, Cyrus R.; Shah, Eric D.; Pohl, Heiko; Tierney, William M.; Mitchell, Stephanie; Condon, Ashwinee; Lenhart, Adrienne; Dua, Kulwinder S.; Kanagala, Vikram S.; Kamal, Ayesha; Singh, Vikesh K.; Pinto-Sanchez, Maria Ines; Hutchinson, Joy M.; Kwon, Richard S.; Korsnes, Sheryl J.; Singh, Harminder; Solati, Zahra; Willingham, Field F.; Yachimksi, Patrick S.; Conwell, Darwin L.; Mosier, Evan; Azab, Mohamed; Patel, Anish; Buxbaum, James; Wani, Sachin; Chak, Amitabh; Hosmer, Amy E.; Keswani, Rajesh N.; DiMaio, Christopher J.; Bronze, Michael S.; Muthusamy, Raman; Canto, Marcia I.; Gjeorgjievski, V. Mihajlo; Imam, Zaid; Odish, Fadi; Edhi, Ahmed I.; Orosey, Molly; Tiwari, Abhinav; Patwardhan, Soumil; Brown, Nicholas G.; Patel, Anish A.; Ordiah, Collins O.; Sloan, Ian P.; Cruz, Lilian; Koza, Casey L.; Okafor, Uchechi; Hollander, Thomas; Furey, Nancy; Reykhart, Olga; Zbib, Natalia H.; Damianos, John A.; Esteban, James; Hajidiacos, Nick; Saul, Melissa; Mays, Melanie; Anderson, Gulsum; Wood, Kelley; Mathews, Laura; Diakova, Galina; Caisse, Molly; Wakefield, Lauren; Nitchie, Haley; Waljee, Akbar K.; Tang, Weijing; Zhang, Yueyang; Zhu, Ji; Deshpande, Amar R.; Rockey, Don C.; Alford, Teldon B.; Durkalski, Valerie
title: Digestive Manifestations in Patients Hospitalized with COVID-19
date: 2020-10-01
journal: Clin Gastroenterol Hepatol
DOI: 10.1016/j.cgh.2020.09.041
sha: 84cfe3fe187c614369a777468f8cfdc8b13ed36c
doc_id: 818736
cord_uid: 2d6htc3s

Background and aims The prevalence and significance of digestive manifestations in COVID-19 remain uncertain. We aimed to assess the prevalence, spectrum, severity, and significance of digestive manifestations in patients hospitalized with COVID-19. Methods Consecutive patients hospitalized with COVID-19 were identified across a geographically diverse alliance of medical centers in North America. Data pertaining to baseline characteristics, symptomatology, laboratory assessment, imaging, and endoscopic findings from the time of symptom onset until discharge or death were manually abstracted from electronic health records to characterize the prevalence, spectrum, and severity of digestive manifestations. Regression analyses were performed to evaluate the association between digestive manifestations and severe outcomes related to COVID-19. Findings A total of 1992 patients across 36 centers met eligibility criteria and were included. Overall, 53% of patients experienced at least one gastrointestinal symptom at any time during their illness, most commonly diarrhea (34%), nausea (27%), vomiting (16%), and abdominal pain (11%). In 74% of cases, gastrointestinal symptoms were judged to be mild. In total, 35% of patients developed an abnormal alanine aminotransferase or total bilirubin level; these were elevated to less than 5 times the upper limit of normal in 77% of cases. After adjusting for potential confounders, the presence of gastrointestinal symptoms at any time (odds ratio 0.93, 95% confidence interval 0.76-1.15) or liver test abnormalities on admission (odds ratio 1.31, 95% confidence interval 0.80-2.12) were not independently associated with mechanical ventilation or death. Conclusion Among patients hospitalized with COVID-19, gastrointestinal symptoms and liver test abnormalities were common but the majority were mild and their presence was not associated with a more severe clinical course.

Background and aims: The prevalence and significance of digestive manifestations in COVID-19 remain uncertain. We aimed to assess the prevalence, spectrum, severity, and significance of digestive manifestations in patients hospitalized with COVID-19.

Methods: Consecutive patients hospitalized with COVID-19 were identified across a geographically diverse alliance of medical centers in North America. Data pertaining to baseline characteristics, symptomatology, laboratory assessment, imaging, and endoscopic findings from the time of symptom onset until discharge or death were manually abstracted from electronic health records to characterize the prevalence, spectrum, and severity of digestive manifestations. Regression analyses were performed to evaluate the association between digestive manifestations and severe outcomes related to COVID-19.

Findings: A total of 1992 patients across 36 centers met eligibility criteria and were included. Overall, 53% of patients experienced at least one gastrointestinal symptom at any time during their illness, most commonly diarrhea (34%), nausea (27%), vomiting (16%), and abdominal pain (11%). In 74% of cases, gastrointestinal symptoms were judged to be mild. In total, 35% of patients developed an abnormal alanine aminotransferase or total bilirubin level; these were elevated to less than 5 times the upper limit of normal in 77% of cases. After adjusting for potential confounders, the presence of gastrointestinal symptoms at any time (odds ratio 0.93, 95% confidence interval 0.76-1. 15 ) or liver test abnormalities on admission (odds ratio 1.31, 95% Even though COVID-19 is primarily a respiratory illness, the digestive system has been implicated in disease expression, transmission, and possible pathogenesis. The responsible virus -Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) -gains cellular entry through the Angiotensin Converting Enzyme 2 (ACE2) receptor, which is present in the gastrointestinal tract at higher levels than in the respiratory system (1) (2) (3) . Viral RNA has been detected in the stool of approximately 50% of affected patients (1, (4) (5) , and it is hypothesized that enteric infection might modulate the severity of pulmonary and systemic illness through alterations in the microbiome, dysregulated intestinal immunity, and/or increased gut permeability (6) (7) (8) .

Digestive manifestations may be common in patients with COVID-19, although reports are conflicting, and the true prevalence remains uncertain. Early series from China and two recent meta-analyses suggest that gastrointestinal symptoms occur in less than 10% of patients (9) (10) (11) (12) (13) whereas other studies suggest proportions in the range of 30-60% (14) (15) (16) (17) . The prevalence of abnormal liver tests similarly varies from 15 to 50% (9, 12, 13, 15) . More importantly, the significance of digestive manifestations in COVID-19, in terms of impact on the alimentary tract and liver, and on overall outcomes, is unknown. The presence and magnitude of gastrointestinal symptoms and hepatic abnormalities have mirrored disease severity in some studies, but this observation has been inconsistent (18) (19) (20) (21) .

Reports on the digestive manifestations of COVID-19 have been limited in scope, reflecting the experience of a single hospital or isolated geographic region, and have used varying and potentially biased sampling strategies. We aimed to systematically and rigorously assess the prevalence, spectrum, and severity of digestive manifestations in consecutive patients hospitalized with COVID-19 across geographically diverse medical centers in North America. We also explored the association between the presence of digestive manifestations and overall outcomes.

This was an observational cohort study conducted through an alliance of 36 medical centers in the United States and Canada. Any site in North America was eligible for participation by open invitation. Certain sites were specifically invited to maximize geographic, ethnic, and socioeconomic diversity, and to ensure representation of regions that were disproportionately affected by the early phase of the pandemic.

Institutional review board approval was obtained at each center prior to patient identification and data collection.

Adult patients who were hospitalized with a confirmed diagnosis of COVID-19 according to local real-time PCR testing were considered eligible. To ensure an unbiased sample, we aimed to enroll the first 50-100 consecutive patients meeting eligibility criteria at each participating institution. Potentially eligible patients were identified by site J o u r n a l P r e -p r o o f investigators using multiple methods, including but not limited to, data warehouse queries, electronic research subject identification tools, and lists provided by the infectious diseases service or other relevant hospital entities.

Demographic, clinical, laboratory, radiographic, and endoscopic data from symptom onset until discharge or death were manually abstracted through review of electronic health records by study personnel under the oversight of a designated clinicianinvestigator. Deidentified data were entered directly into an electronic data collection form (Supplement). When patients had not been dispositioned by the end of the study period, data were collected within 3 days of study closure.

Data quality was ensured to the greatest extent possible using a three-tiered system.

First, formal instructions and consistent communications between the data coordinating center and co-investigators emphasized the importance of ensuring response accuracy at the site level and of seeking clinician-investigator input for responses that required clinical interpretation. In addition, a manual of procedures for data collection and for the handling of special scenarios (e.g. readmissions or nosocomial infections) was circulated frequently to the sites throughout the study period. Second, all incoming data were manually reviewed by a data manager to identify missing or duplicate data, to verify that responses fell within accepted boundaries, and to assess for discrepant or conflicting responses. Third, data were reviewed in aggregate by the study team J o u r n a l P r e -p r o o f primarily to assess for inconsistencies and outliers by center. Data concerns prompted direct queries to the sites that were resolved prior to the final database freeze.

Digestive manifestations were divided into gastrointestinal symptoms and liver test abnormalities. Since anorexia is a common and non-specific symptom of viral illness, it was not considered a digestive manifestation in this study. Similarly, we did not include constipation because it has not previously been implicated as a symptom of acute or subacute viral infection. Therefore, the symptoms of interest in this study were diarrhea, nausea, vomiting, abdominal pain, gastrointestinal bleeding, dysphagia, and odynophagia. Patients were judged to have moderate-severe gastrointestinal symptoms when one or more of the following criteria were satisfied: 1) diarrhea with >4 bowel movements in any 24 hour period, 2) bloody diarrhea, 3) hematemesis, melena, or hematochezia, 4) abdominal computed tomography (CT) scan or endoscopic evaluation was performed, or 5) the gastroenterology consult service evaluated the patient. All other patients were considered to have mild symptoms.

Liver test abnormalities were defined as mild when the alanine aminotransferase level (ALT) or total bilirubin (TB) level was elevated between 1.5 and 3 times the upper limit of normal, moderate when there was >3 to 5 times elevation, and severe at >5 times elevation. Liver tests <1.5 times the upper limit were considered normal in this study.

The upper limits of normal for ALT and TB were considered to be 45 units/L and 1.2 mg/dL respectively. Severe acute liver injury was defined as an ALT>1000 units/L with J o u r n a l P r e -p r o o f an international normalized ratio > 2 or a factor 5 level < 25%. For descriptive purposes, the proportion of patients with any liver test abnormality is reported, but only ALT and TB elevation were used in the analyses.

Digestive manifestations were reported using descriptive statistics. Categorical variables were expressed as counts or percentages with 95% confidence intervals (95%CI); continuous variables were expressed as means with standard deviation (SD) or medians with interquartile range (IQR) depending on distribution. Liver test abnormality proportions were calculated using the full cohort as the denominator.

The association between digestive manifestations and the severity of COVID-19 was assessed using a multivariable logistic regression model which included the presence of gastrointestinal symptoms and liver test abnormalities as the independent variables of interest and which adjusted for pre-specified baseline covariates. The primary outcome was the composite endpoint of mechanical ventilation and/or death. Potential covariates that were considered for the model are listed in the Supplement. If a univariable association with the outcome was observed (p<0.10), the covariate was considered for inclusion in the final regression model. We also explored potential interactions between included covariates and the primary independent variables. Only liver tests at admission were assessed, since hepatic injury during critical illness is consistently associated with multi-organ system failure and death regardless of etiology (22) .

In exploratory analyses, the associations between digestive manifestations and intensive care unit admission, the need for vasopressor support, and hospital length of stay (modeled as a continuous variable) were assessed using a similar approach.

All analyses were conducted using SAS 9.4 (SAS Institute, Inc. Cary, North Carolina).

The programming code for the final primary regression model is included in the Supplement.

All authors had access to the study data and reviewed and approved the final manuscript.

Patients From 15 April until 5 June 2020, data were collected from 1992 subjects across 36 centers. The median number of patients enrolled per participating institution was 51 (IQR 41-68). Characteristics of the study cohort are shown in Table 1 . The average age was 60 years (SD 16.3); 57% were men; 42% were black/African American. Eighty-nine percent of patients had at least one non-digestive comorbidity; 9% had a pre-existing digestive disorder. Thirty-two percent of patients required mechanical ventilation and 19% died. The median hospital length of stay was 9 days (IQR 4-17). Thirty patients (1.5%) were still hospitalized at the end of the study period. Western studies (16, 17) . This is in contrast to much lower proportions of gastrointestinal symptoms reported in studies from China (9, 10) . This difference may be because the Our findings are consistent with prior reports demonstrating that liver tests abnormalities are common in COVID-19. Fifty-five percent of patients in this cohort had an elevated liver test at some point during their illness, including many with abnormal aminotransferases, raising the possibility of hepatocyte injury due to SARS-CoV-2.

However, the large majority had ALT or TB levels less than 5 times the upper limit of J o u r n a l P r e -p r o o f normal and only 23 patients had an ALT >1000 units/L, which was never present at the time of presentation, suggesting that clinically important liver injury in COVID-19 is uncommon. Future mechanistic investigation will be necessary to better understand whether infection leads to direct hepatic injury.

The findings of this study should be interpreted in the context of several limitations, some of which are inherent to observational research on COVID-19. As highlighted above, symptom attribution and ascertainment were influenced by several factors related to conducting research during a pandemic, such as retrospective data collection and reliance on medical records review rather than direct patient interviews. Further, validated definitions for gastrointestinal symptom severity in COVID-19 are not available and thus we devised criteria that we believe reasonably reflect disease severity in terms of patient suffering and resource utilization. Alternative definitions of severity may have led to varying interpretations of the findings. Some of these limitations are mitigated by the large and geographically diverse sample, highly systematic approach to patient selection, and multi-layered and rigorous strategy to ensure the veracity of collected data. It is also important to consider that this study was restricted to hospitalized patients and thus does not reflect the prevalence and significance of digestive manifestations in outpatients with COVID-19.

In summary, among patients hospitalized with COVID-19, gastrointestinal symptoms and liver test abnormalities were common, but the majority were mild in nature and their presence was not associated with worse clinical outcomes. 

The electronic data collection form contained 216 items and was divided into 5 sections: 1) patient characteristics; 2) COVID-19 parameters; 3) symptomatology; 4) imaging and endoscopy; and 5) laboratory data and other hepatic considerations. The form was developed by a core group of investigators and was refined and finalized after a 72-hour period of public comment.

1) Please complete the below data collection form (DCF) in REDCap at the time of discharge or death. Data will appear in the DMC19 database once entry and verification are complete.

2) We aim to capture inpatients with a confirmed COVID-19 diagnosis, regardless of whether they have digestive manifestations. After prevalence is defined in hospitalized patients, and as the numbers grow, we may focus on patients who are known to have GI manifestations and/or include outpatients.

3) Please make all efforts to collect data on the first 50-100 consecutive patients at your hospital or health system.

Eligible patients can and should be identified by any means necessary, which may include, but is not limited to, institutional laboratory records, data warehouse queries, electronic health record research subject identification tools/dashboards, and discussions with the infectious disease or critical care services, etc. You may elect to use the emergency ICD-10 code of U07.1 -2019-nCov acute respiratory disease -to help identify eligible patients.

Please triple check data for accuracy before submission. Although we are performing central data monitoring, we cannot verify incoming data against source documents, nor are we performing on-site monitoring visits. Therefore, the overall quality of the data is assured primarily at the site level.

6) Along the lines of #5, coordinators should confer with a clinician during data collection to ensure that clinical context is accounted for as much as possible in the interpretation of questions that involve an element of subjectivity.

7) All data fields should have affirmative, negative, and unknown options. Therefore, missing data will be assumed to be inadvertent and this will generate a query.

8) Please maintain a secure key at your site that allows patient identification on the basis of subject ID#. This may be used in the future for to collect data pertaining to long-term outcomes.

Age ( 

Evidence for Gastrointestinal Infection of SARS-CoV-2

Multi-omics Evaluation of Gastrointestinal and Other Clinical Characteristics of SARS-CoV-2 and COVID-19

ACE2 angiotensin I converting enzyme 2

Digestive Symptoms in COVID-19 Patients with Mild Disease Severity: Clinical Presentation, Stool Viral RNA Testing, and Outcomes

Prolonged presence of SARS-CoV-2 viral RNA in faecal samples

Microbiota-driven tonic interferon

The Gut-Lung Axis in Respiratory Disease

COVID-19, Coronavirus, SARS-CoV-2 and the small bowel

Clinical Characteristics of Coronavirus Disease 2019 in China

Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China

Epidemiological, clinical and virological characteristics of 74 cases of coronavirus-infected disease 2019 (COVID-19) with gastrointestinal symptoms

AGA Institute Rapid Review of the GI and Liver Manifestations of COVID-19, Meta-Analysis of International Data, and Recommendations for the Consultative Management of Patients with COVID-19

Prevalence of Gastrointestinal Symptoms and Fecal Viral Shedding in Patients With Coronavirus J o u r n a l P r e -p r o o f Disease 2019: A Systematic Review and Meta-analysis

Diarrhea during COVID-19 infection: pathogenesis, epidemiology, prevention and management

Current Knowledge and Research Priorities in the Digestive Manifestations of COVID-19

Prevalence and Characteristics of Gastrointestinal Symptoms in Patients with SARS-CoV-2 Infection in the United States: A Multicenter Cohort Study

Gastrointestinal and Hepatic Manifestations of 2019 Novel Coronavirus Disease in a Large Cohort of Infected Patients From New York: Clinical Implications

Liver injury in COVID-19: management and challenges

Clinical characteristics of COVID-19 patients with digestive symptoms in Hubei, China: a descriptive, cross-sectional

Diarrhea is associated with prolonged symptoms and viral carriage in COVID-19

Manifestations and Prognosis of Gastrointestinal and Liver Involvement in Patients With COVID-19: A Systematic Review and Meta-Analysis Lancet Gastroenterol Hepatol

Impact of Intensive Care Unit Organ Failures on Mortality During the Five Years After a Critical Illness

Severe acute respiratory syndrome coronavirus 2: virus mutations in specific European populations

Prevalence and distribution of abnormal liver tests in 1825 patients who were not known to have received any of the following medications with potential hepatotoxicity: remdesivir, oseltamivir, lopinavir/ritonavir, interferon alpha, compared to the full cohort:Excluding subjects known to have received COVID- 19 

The final primary model was developed by manual selection of variables based on univariate analyses, followed by examination of the Akaike Information Criterion (AIC) as potential variables were added and removed from the model. Yes -Mild to moderate respiratory compromise Yes -Severe respiratory compromise Yes -Congestive heart failure/cardiomyopathy Yes -Myocardial infarction No Unknown Endoscopic findings (please list abnormal findings in impression section of endoscopy report(s)) Histologic findings (please list abnormal findings in impression section of pathology report(s)) -What is the upper limit of normal Amylase (highest) -What is the upper limit of normal Creatinine (highest) Absolute lymphocyte count (lowest) C-Reactive Protein (CRP) (highest) Procalcitonin (highest) Troponin (highest) Ferritin (highest) ng/mL or ug/L Interleukin-6 (highest) pg/mL Duration between first onset of symptoms and highest AST (days) Duration between first onset of symptoms and highest ALT (days) Duration between first onset of symptoms and highest total bilirubin (days) Duration between first hospital day and highest AST (days) Duration between first hospital day and highest ALT (days) Duration between first hospital day and highest total bilirubin (days) Abnormal LFTs were Still close to max at discharge/death Improved but not resolved at discharge/death Resolved or close to resolved at discharge/death Not applicable Were the increased LFTs suspected to be due to a drug reaction (based on review of progress/consult notes)?