Affiliations: Weill Cornell Medicine, United States
Journal reference: https://doi.org/10.1053/j.gastro.2020.05.010
Summary: One of the most surprising facts about COVID-19 is the variety of its manifestations. Western populations for instance have been presenting more gastrointestinal manifestations than those in China or the Far East. Here we analyzed the data of all COVID-19 patients in New York during the first year of the pandemic to understand how common these manifestations were and how they could affect a patient’s disease course and recovery.
What did we do?
On February 29th 2020, the first case of COVID-19 was reported in New York City (NYC), marking the beginning of a novel disease in the USA. NYC has been one of the hardest hit cities, and our institution, New York Presbyterian-Cornell, has bore the brunt of the cases. Despite the tragedy, this has given us a unique opportunity to understand the disease and how it presents in the different ethnic, racial and cultural populations that reside in NYC.
In treating these patients, we noticed an important difference in how COVID-19 was affecting our community in comparison to those in China and the Far East. Western populations were presenting with more gastrointestinal manifestations, namely diarrhea, nausea/vomiting, and abdominal pain and liver involvement. Our objective with this study was to investigate how common these gastrointestinal/liver manifestations are, and to explore how these affect a patient’s disease course and recovery.
How did we do it?
We reviewed the electronic medical records of all patients with a positive COVID-19 PCR test recorded between March 4th and April 9th 2020 at our institution. We created a database in which we included patient variables, such as age and sex, clinical characteristics, vital signs and laboratory data, and key aspects of their disease and treatment course. We took special note of whether patients were treated on an outpatient basis, meaning they were observed in the emergency room and discharged within 24 hours of when they first came in, or on an inpatient basis, meaning they were admitted to the hospital for > 24 hours. We performed various statistical tests to help us in predicting which patients ultimately needed a higher level of care or had a more severe disease course.
What did we discover?
Our study group included 1,059 patients diagnosed with COVID-19 of which the average age was 61 years old and 58% were male. 33% of patients had at least one gastrointestinal manifestation, of which diarrhea was the most common (22%) and abdominal pain (7%) was the least common. 62% of patients had evidence of liver involvement, defined as an elevated level of at least one of the liver-specific markers (enzymes = AST, ALT, total bilirubin, Alk Phos).
We found that female patients and patients with chronic liver diseases, such as cirrhosis, were the only two groups of people who had gastrointestinal manifestations on presentation. Older age was associated with liver involvement at presentation to the hospital. Having gastrointestinal manifestations or liver involvement when initially seeking care was associated with higher rates of admission to the hospital, but only liver injury was associated with death or admission to the intensive-care unit (ICU). Older age, a greater number of underlying medical conditions, high respiratory rate on admission to the hospital, and severe lack of oxygen saturation in the blood were also predictors of death or admission to the ICU.
What does this mean and why is this important?
Given our findings, we created a decision tree with the aim of helping providers characterize who is at risk of death or ICU admission after contracting COVID-19. Patients with the highest risk of death or ICU admission at 56% were those with severely low blood oxygen saturation and evidence of liver involvement at the initial point of contact with medical care. In contrast, those with the lowest risk of death or ICU admission at 18% were those under 62 years old and with no blood oxygen saturation issues.
In summary, both gastrointestinal manifestations and liver involvement when initially seeking care was associated with higher rates of admission to the hospital, but only liver involvement was associated with higher risk of death or ICU admission, especially in the setting of vital sign abnormalities or older age as above.
What is next?
We hope that this data will not only help doctors better diagnose COVID-19 infections, but also help them decide earlier on who needs more intense care or observation. We hope it will facilitate better conversation between doctors and patients with COVID-19 on how their infection will likely evolve. Most importantly, it is my hope that it will inspire patients to seek medical care if they believe they are infected with COVID-19 and to follow recommendations and guidelines by health authorities to stop the spread of COVID-19.