A case of gastrointestinal-predominant COVID-19 demonstrates value of stool PCR test.

INTRODUCTION

With over 5 million confirmed cases leading to over 165 000 deaths, coronavirus disease 2019 (COVID‐19) has quickly made a large impact in the United States. COVID‐19, which began as a viral outbreak in Wuhan City, China, was first detected in the United States during January 2020. COVID‐19 is now a global pandemic that has caused significant emotional and economic burden across the United States. Although the most common presenting symptoms include cough, fever, dyspnea, and myalgias, many individuals present with gastrointestinal symptoms such as diarrhea (7.7%), nausea and vomiting (7.8%), and abdominal pain (2.7%). The initial diagnosis of COVID‐19 for patients requiring inpatient care may be particularly challenging when false‐negative results arise. Nasopharyngeal (NP) swabs for reverse transcriptase‐polymerase chain reaction (RT‐PCR) analysis are often used for diagnosis; however, the false‐negative rate has been reported up to 38% on the day of symptom onset. Prolonged coronavirus detection in the stool may be integral to diagnosis in these cases, as the prevalence of viral RNA in rectal/anal swabs and stool specimens has been shown to be 32% after negative conversion in NP RT‐PCR Test. Data on the diagnostic yield of stool RT‐PCR for COVID‐19 varies, as the positive rate of detection has been reported between 29% and 87.9%. , We present a case that truly demonstrates the value of stool PCR testing, as our patient had four false‐negative nasopharyngeal swabs, negative COVID‐19 antibodies, and a positive stool PCR test.

CASE DESCRIPTION

A 66‐year‐old Caucasian male with diverticulosis and medical history of latent tuberculosis presented with worsening nausea, bilious, nonbloody vomiting, nonbloody diarrhea, and decreased oral intake for 10 days. He had extensive exposure to the virus as a pulmonary and critical care physician, directly taking care of COVID‐19 positive patients. He denied shortness of breath, cough, nasal congestion, loss of taste or smell, or headaches. His vital signs on admission were as follows: temperature 37.2°C, heart rate 65 beats/min, respiratory rate 18 breaths/min, oxygen saturation 94% on 2 L nasal cannula, and blood pressure 107/65mm Hg. Labs on admission were significant for lymphocytes 0.6 K/μL (normal range, 0.90‐3.50 K/μL), lactate dehydrogenase 288 unit/L (normal range, 87‐241 unit/L), ferritin 863.1 ng/mL (normal range, 26‐388 ng/mL), C‐reactive protein 45.1 mg/L (normal range, <3.0 mg/L). On day 2 of admission, erythrocyte sedimentation rate was 18 mm/h (normal range, 0‐20 mm/h), and D‐dimer was 341.8 ng/mL (normal range, 0‐330 ng/mL in D‐DU adjusted for age). He was started on intravenous (IV) maintenance fluids with normal saline at 120 cc/h for 1 L, followed by 100 cc/h, and IV ondansetron 4 mg IV every 6 hour as needed to control his nausea. Oral acetaminophen was prescribed as needed to control his fevers. Stool studies (including stool white blood cells, stool culture, stool ova, and parasite exam) and blood cultures were sent, which were unremarkable. Initial chest X‐ray was normal, however, a computed tomography scan of the chest demonstrated classic findings of COVID‐19 viral bronchopneumonia (Figure 1). Four total nasopharyngeal swabs for RT‐PCR analysis were sent—the first swab was sent 2 days before admission during an ED visit, the second and third swabs were sent on admission day 1, and the fourth swab was sent on admission day 3. Despite four negative NP swabs, there was high suspicion for COVID‐19 infection. In an effort to procure a molecular diagnosis of COVID‐19 infection, one stool swab was sent (by swabbing a collecting sample of stool, rather than a rectal swab) for RT‐PCR analysis on day 4 of admission and was positive. On day 4 of supportive care, he was given methylprednisolone 40 mg IV every 12 hour and rapidly improved over a 4‐day period. On day 5, he no longer required supplemental oxygen with 2 L nasal cannula, and his oxygen saturation was 92% to 96% on room air up until day 7, when he was discharged home safely without home oxygen.

Figure 1

CT chest scan revealing right greater than left lower lobe patchy ground‐glass opacities as well as alveolar interstitial opacities. There is right greater than left posterior dependent segmental lower lobe consolidation, as well as areas of subsegmental atelectasis or scarring. There are also patchy peripheral ground‐glass opacities, mainly in the right middle lobe and right upper lobe (not well represented from this slice). CT, computed tomography

DISCUSSION

Coronavirus enters host cells via the angiotensin‐converting enzyme 2 receptors, which is expressed not only in the respiratory tract, but also throughout the gastrointestinal tract, from the upper esophagus and liver to the small intestinal epithelia and the colon. In a large meta‐analysis of 8135 pooled clinical specimens, rectal swabs were found to have a high positive rate of detection (87.8%), second only to bronchoalveolar fluid specimens, which had a positive rate of 91.8%. Surprisingly, NP swabs were found to have only a moderate detection rate (positive rate of 45.5%). In addition, viral RNA shedding of coronavirus been shown to persistent in feces longer than in the respiratory tract. This allows for a prolonged window of detection and may suggest why rectal swabs have been shown to have a higher positive rate of detection. Although not yet confirmed, these factors definitely add to a growing body of evidence to suggest that COVID‐19 may be spread by fecal‐oral transmission. Hindson reports a scenario where eight patients had persistently positive rectal swabs even after their NP swabs were negative and stressed the importance of spreading awareness that stool may possibly be infectious, while larger cohort studies are developed to confirm the suspicion of fecal‐oral transmission. Given this growing body of evidence, it is likely that our patient had increased virus activity and shedding in his gastrointestinal tract. Thus, stool PCR was particularly integral to the management of our patient, as it was our last chance to obtain a molecular diagnosis, which is required for certain therapeutic options to be considered at our institution (eg, administration of convalescent antibodies or an antiviral medication).

CONCLUSION

We propose that stool PCR is a valuable asset in diagnosing COVID‐19 infection, especially in gastrointestinal‐predominant cases that prove to be difficult to diagnose with the classic nasopharyngeal assay. Future studies with larger sample sizes should be conducted to further elucidate the value of stool PCR in the diagnosis of COVID‐19 in such cases.

CONFLICT OF INTERESTS

The authors do not have any conflict of interests or funding sources associated with this manuscript. This information has not yet been presented at any meeting.

ETHICS STATEMENT

Informed patient consent was obtained for case publication.