Primary presentation of COVID-19 solely with gastrointestinal symptoms: a problem for the containment of the disease

Abstract

An oligosymptomatic patient with initially exclusively gastrointestinal symptoms (massive nausea and mild pancreatitis) turned out to suffer from COVID-19 during the hospital stay. The patient did not exhibit the typical COVID-symptoms cough or fever despite lung involvement. The gastroenterological diagnostic investigations comprised abdominal ultrasound, gastroscopy and endoscopic ultrasound and first revealed no clear cause of these symptoms. In an abdominal computed tomography (CT) scan, patchy ground-glass opacities in both lungs were discovered and the following chest CT scan and a virus PCR test revealed the right and surprising diagnosis. This case report depicts the clinical course of this atypical case and discusses gastrointestinal COVID-manifestations and clinical consequences as well as consequences of this atypical presentation for disease control.

Introduction

In December 2019, a new infectious lung disease was described in Wuhan, China [^1,2] and subsequently termed COVID-19 [³]. A novel coronavirus was identified as causative for this disease and, because of similarities to the already known coronavirus SARS-CoV [⁴], initially named 2019-nCoV [⁵] and later on designated as SARS-CoV-2 [⁴].

The disease was first described as a bilateral pneumonia and eventually severe acute respiratory syndrome [¹] with high fever, but during the first 6 months after the identification of the disease, it has become clear that also extrapulmonary symptoms play an important role in COVID-19 [^6,7]. Because COVID-19 is a highly infectious disease, it is a mainstay of disease control to immediately identify sick persons and isolate them in a clinic or at home [⁸]. Therefore, atypical presentation of COVID-19 is a serious problem for this strategy of containment. We report a case of COVID-19 with solely gastrointestinal symptoms and (initially) without any classical symptoms of the disease.

Case report

On 15 March 2020, the pandemia started to accelerate in Europe. About 6500 cases with positive tests for SARS-CoV-2 were registered in Germany and about 150 in the region of Aachen. On that day, a 72-year-old male presented to the emergency department with a 7-day history of progressing nausea and mild abdominal pain. Apart from being overweight (87 kg on 172 cm, BMI 29.4 kg/m²) and having arterial hypertension treated with a ß-blocker, the patient had no serious diseases in the past and he was a nonsmoker. The intensity of the nausea was very high in the last days before admission so that the patient refused any uptake of food and consumed only water. There was no known contact to any (possible) COVID-19-patient in the weeks before admission, as was figured out later after the diagnosis was made.

His wife reported fatigue, lethargy and listlessness of her husband in the last days. The patient slept very well, but did not leave the house. In the emergency room, he denied any fever or respiratory symptoms such as cough and shortness of breath.

At admission, the patient was calm and responded only shortly to questions. Head, neck and throat were normal. Heart and lung revealed no abnormal murmurs or wheezing. The abdomen was bloated and tender, with diffuse pain. The patient had no edema, spine and limbs were unremarkable.

Diagnostic procedures

The laboratory values at admission (and further on) are given in Table 1 and showed a mild elevation of glutamate oxalacetate transferase y-glutamyl transferase and lipase, and alkaline phosphatase was normal.

Table 1. - Laboratory values

Parameter (unit)	At admission (normal values)	Day 5 (diagnosis)	Day 11 (ICU)	Day 17	Day 60
WBC (/nl)	4.8 (4–10)	7.0	7.0	7.6	10.8
CRP (mg/dl)	2.3 (< 0.5)	12.5	13.2	6.0	18.8
GOT (U/L)	55 (< 50)	63	n.d.	74	31
yGT (U/L)	80 (< 60)	70	99	102	13
Lipase (U/L)	185 (< 67)	69	n.d.	45	42
LDH (U/L)	382 (< 250)	611	n.d.	477	203
Creatinine (mg/dl)	1.9 (< 1.4)	1.1	0.8	0.9	0.58
PaO2/FiO2	300	225	146	165	>300

CRP, C-reactive protein; GOT, glutamate oxalacetate transferase, LDH, lactate dehydrogenase; n.d., not determined; WBC, white blood count; yGT, y-glutamyl transferase.

Abdominal ultrasound revealed cholecystolithiasis, but no signs of obstructive cholestasis. The pancreas was barely visible, but the parenchyma seemed to be inhomogenous.

The patient received a gastroscopy and endosonography. Here, inhomogenous pancreatic tissue without any focal mass and a lean bile duct with no signs of an intraluminal stone or papilla passage of a stone were seen. In the stomach, a lipoma and erosive gastritis was diagnosed.

With no improvement under treatment with proton pump inhibitor or procinetic substances, we conducted an abdominal CT on 18 March (day 4). No significant abdominal findings could be obtained, but the basal lung showed patchy infiltrative lesions. In a cerebral CT, only older ischemic lesions were found.

The next day, we conducted a thoracic CT (Fig. 1) and obtained pictures with opacities and glass-ground infiltrates diffusely distributed in both lungs. The patient immediately was isolated at our quarantine ward and swabs were taken from the throat and the nose the same day and analyzed for influenza and Sars-CoV2.

Fig. 1.: Representative chest computertomogram of the patient on day 5 (day 12 since beginning of gastrointestinal symptoms).

The patient now developed rising C-reactive protein (CRP) values (see Table 1) and a transient modest fever of 38.5°C, but still did not cough. However, he reported exertional dyspnea. After obtaining blood gases and blood cultures, the patient received ceftriaxone and clarithromycine. The patient tested positive for SARS-CoV-2.

Diagnosis: COVID-19 disease due to the novel corona virus SARS-CoV-2

Treatment

The patient received oxygen and breathing exercises. On the second day after PCR diagnosis, the CRP fell and the patient had no fever anymore.

The patient still reported unchanged nausea and further on refused eating and also breathing exercises. On day 10 (day 5 after the diagnosis), the patient suddenly developed respiratory distress and was taken to the ICU, where intubation was done at a PaO₂/FiO₂ (p/F ratio) of 146.

The patient was transferred to the university hospital on day 20. Under lung protective ventilation, high positive end-expiratory pressure settings, intermittent prone positioning and fluid restriction, mild hypercapnia (with paCO₂ < 60 mmHg) could be maintained and oxygenation (with p/F ratio > 200) improved within 48 h; however, on day 25, BNP elevated significantly and a thoracic CT scan was performed. Bilateral pulmonary artery embolism could be diagnosed and a therapeutic anticoagulation was started. In a cerebral CT, small infarction signs in the left frontal lobe and occipitoparietal region were seen.

On day 26, analgosedation could be reduced and the patient showed first efforts of spontaneous breathing. Under pressure support ventilation, intermittent t-piece trials could be performed since day 28 of treatment.

As another relapse, a ventilator-associated pneumonia was diagnosed on day 34, resulting in a new period of controlled mechanical ventilation. Antiinfective therapy was given for 7 days. Repetitive bronchoscopies were performed. Yet, measures of proning were not required during this episode. On day 40, short periods of t-piece trials could again be established and further expanded within the next days. Vigilance also improved remarkably, however, due to persistent dysphagia, a percutaneous endoscopic gastrostomy tube was installed on day 46. Thereafter, ascending, activating physiotherapy and speech therapy led to further neurological improvement. In sum, in terms of ventilatory support, the process of prolonged weaning could be completed on day 64, but a tracheostoma was still needed. After regaining vigilance, the patient denied any gastrointestinal (GI) symptoms such as nausea or abdominal pain, and his lipase remained normal.

Discussion

COVID-19 is primarily a respiratory disease caused by the novel corona virus SARS-CoV-2. Typical symptoms of COVID-19 comprise dry cough and high fever [³]. In several populations, other more rare symptoms such as diarrhea and dysgeusia have been reported in up to 50% [^9,10]. GI symptoms in general seem to be frequent in COVID-19 and have been reported in a cohort from the USA in 60% of 318 patients [¹¹]. Fatigue obviously is a frequent symptom in COVID-19-patients (from 12 to almost 60% of patients [^11,12]), but also accompanies infections in general. In the course of his long ICU stay, the patient also suffered from other complications described in case series of COVID-19 such as thromboembolic events [⁷] and recurrent ventilator-associated pneumonia.

To our knowledge, nausea as a single presenting symptom has not been reported for COVID-19 so far. In the above cited US cohort study and in a comprehensive Chinese case series [⁶], diarrhea was the main GI symptom, although often accompanied by nausea and vomiting. Nausea as a single GI symptom was present in only 10 of 651 analyzed patients, but these patients exhibited typical COVID-19 symptoms such as fever and/or cough. Our patient did not present with any fever or cough on admission, despite these very widespread changes in both the lungs and a disease course of already 1 week at this time point. Fever and exertional dyspnea developed only a few days after admission; however, cough was not present at any time.

Pancreatitis has been reported in two case reports [^13,14] but the reported patients either developed pancreatitis in the late course of the disease [¹⁴] or showed clear COVID-typical symptoms such as fever and/or cough as presenting symptoms of the disease [¹³]. In our case, a COVID-pancreatitis is likely because other etiologies for pancreatitis such as alcohol or bile duct stones could be excluded.

Interestingly, the patient initially showed stable (nausea) or even improved clinical symptoms (abdominal pain, low fever for only 24 h) over 5 days after diagnosis of COVID-19, before he suddenly developed respiratory distress leading to the necessity of intubation. The main gastrointestinal symptoms could not be evaluated further after worsening of the clinical condition because the patient required sedation for mechanical ventilation. Nevertheless, we attribute the massive nausea of the patient and also the mild pancreatitis to COVID-19 because there could not be found any other reason for these symptoms and they resolved completely after the patient had recovered from the severe lung disease.

This presented atypical clinical picture led to a delayed diagnosis of COVID-19 and to a high threat of further infections out of and in the hospital. Analysis and testing of all contact persons in the primary hospital revealed no cases of SARS-CoV-2-infection. Probably, the lack of cough reduced the spreading of the virus from the patient to the caring personnel of the hospital.

The disease chameleon COVID-19 seems to mimic a high number of other diseases with occasionally very atypical symptoms. In times of a pandemic event with a high number of COVID-19 patients, even those atypical symptoms should lead to PCR testing or pulmonary imaging in order to exclude (or confirm) COVID-19 as origin. Because COVID-19 often goes along with unremarkable changes in auscultation and sometimes also conventional radiographs and CT imaging of the lung have the best sensitivity for diagnosis of COVID-19 [¹⁵].

The endoscopic investigations in this patient (gastroscopy and endosonography) were done only with a surgical mask as protection for the investigators because at this time point, the positivity for SARS-CoV-2 was not known. Although no FFP2-masks were used, no infections took place. This fact may support the notion that also surgical masks provide some protection against SARS-CoV-2 or similar viruses, as already has been implied by other studies [^16,17].

Acknowledgements

C.G.D. and A.B. cared for the patient initially, collected all data and wrote the manuscript. D.H. contributed the radiological investigations and the interpretation of the radiologic findings. J.B. and G.M. cared for the patient, provided data from the clinical course and contributed to writing the manuscript.

Conflicts of interest

There are no conflicts of interest.