Severe Foodborne Bacterial Infections Mimicking Multisystem Inflammatory Syndrome in Children Associated With COVID-19

To the Editors:

We read with interest the article by Campbell et al¹ entitled “Non-SARS-CoV-2 infections among patients evaluated for multisystem inflammatory syndrome in children associated with COVID-19 (MIS-C).” As underlined by the authors, the features of multisystem inflammatory syndrome (MIS-C) included in case definitions overlap with many infectious diseases.

The authors describe 39 children evaluated with suspected MIS-C, 11 of whom had nonsevere acute respiratory syndrome coronavirus-2 (SARS-CoV-2) infections. We also diagnosed 4 cases of foodborne bacterial infection, mimicking MIS-C attended in a tertiary hospital between September 2, 2020, and November 18, 2020. They were previously healthy males between 4 and 15 years old presenting with clinical and laboratory signs compatible with MIS-C (Table 1). All cases had fever and abdominal pain, 2 cases had vomiting and 1 associated diarrhea. None reported respiratory symptoms or skin lesions. All children presented with tachycardia, and 3 developed shock, requiring intravenous fluids. Additionally, 1 child required vasoactive support with intensive care unit admission.

TABLE 1. - Characteristics of Children With Foodborne Bacterial Infections Mimicking Multisystem Inflammatory Syndrome Associated With COVID-19 (MIS-C)

Case	1	2	3	4
Age (yr)/gender	4/male	15/male	12/male	8/male
Clinical findings
Fever (duration*)	Yes (6 d)	Yes (7 hr)	Yes (8 hr)	Yes (1 d)
Abdominal pain	Yes	Yes	Yes	Yes
Vomiting	Yes	Yes	No	No
Diarrhea	Yes	No	No	No
Tachycardia	Yes	Yes	Yes	Yes
Hypotension	No	Yes	Yes	Yes
Rash	No	No	No	No
Laboratory findings†
CRP (mg/dL)	9 (11.2)	4	5.4 (10.2)	7.6 (11.8)
Procalcitonin (mcg/L)	8.5 (8.5)	1.9	0.6 (1.3)	0.6 (0.6)
Leukocytes (cells/mm3)	4300 (6100)	16,600	16,600 (17,200)	12,000 (12,000)
Neutrophils (cells/mm3)	3200 (3250)	14,700	14,900 (16,300)	10,200 (10,200)
Lymphocytes(cells/mm3)	900 (900)	700	600 (500)	1300 (1300)
Platelets (cells/mm3)	145,000 (340,000)	217,000	213,000 (273,000)	261,000 (261,000)
APTT (s)	30.4 (30.4)	NP	29.2 (29.3)	36.4 (36.4)
Prothrombin time (s)	13.3 (13.3)	NP	19.9	14.8 (15.5)
d-dimer (ng/mL)	5618	NP	477 (497)	391
Ferritin (mcg/L)	2561	89	NP	154
Troponins (ng/L)	2.1	<1.6	<1.6	<1.6
NT-pro-BNP (ng/L)	82	110	475	69
IL-6 (pg/mL)	NP	138	732.2	15.6
Microbiology
Blood culture	Salmonella Typhi	No isolation	No isolation	No isolation
Stools culture	Salmonella Typhi	Campylobacter jejuni	Campylobacter coli	C. jejuni
SARS-CoV-2 PCR	Negative	Negative	Negative	Negative
SARS-CoV-2 IgG	Negative	Negative	Negative	Positive
COVID-19 contact	No	No	No	Yes
Imaging
Echocardiogram	Normal	NP	Normal	Normal
Abdominal ultrasound	Nonspecific diffuse gallbladder wall thickening	Right hemicolitis with terminal ileitis and mesenteric adenitis	Right ileocolitis and mesenteric adenitis	NP
Treatment
Fluid resuscitation	No	Yes	Yes	Yes
PICU admission	No	No	Yes	No
Inotropes	No	No	Yes (norepinephrine)	No
Respiratory support	No	No	Yes (standard nasal cannulas)	No
Antibiotic	Ceftriaxone	Ceftriaxone	Ceftriaxone	Cefotaxime
Methylprednisolone	Yes (2 mg/kg/d)	No	Yes (2 mg/kg/d)	No
Immunoglobulin	Yes (2 g/kg)	No	Yes (2 g/kg)	No
Acetylsalicylic acid	No	No	No	No

*Duration before consultation.

†Laboratory values represent the value at emergency department and maximum value during admission (inside parenthesis), except for lymphocytes, whose values represent the value at emergency department and the minimum value during admission. Some values were tested only once during admission.

APTT indicates activated partial thromboplastin time; COVID-19, coronavirus disease 2019; CRP, C-reactive protein; IL, interleukin; NP, not performed; NT-Pro-BNP, N-terminal prohormone of brain natriuretic peptide; PCR, polymerase chain reaction; PICU, pediatric intensive care unit.

Normal laboratory values: APTT (27–38 seconds), prothrombin time (10.5–13.5 seconds), d-dimer (0–250 ng/mL), ferritin (22–274 mcg/L), troponins (<34.2 ng/L), NT-pro-BNP (0–300 ng/L) and IL-6 (<4.3 pg/mL).

All had increased acute phase reactants and lymphopenia, and 3 had elevated d-dimer, with only 1 patient having a significant ferritin increase. All children had normal cardiac enzymes, electrocardiogram, echocardiogram and chest radiograph. Two patients had inflammatory findings on abdominal ultrasound, showing ileocolitis and mesenteric lymphadenopathy, whereas 1 child had gallbladder wall thickening.

Upon suspicion of MIS-C, all children were hospitalized following national MIS-C guidelines.² All received empirical antibiotics, and 2 started treatment with methylprednisolone and intravenous immunoglobulin for MIS-C. All 4 children tested negative for SARS-CoV-2 by polymerase chain reaction, and only 1 had SARS-CoV-2-positive IgG antibodies. After a broad diagnostic workup, gastrointestinal bacterial infection was diagnosed in all patients, and MIS-C was excluded. Salmonella typhi was isolated in blood culture in 1 case (autochthonous acquisition) and Campylobacter spp isolated in stool cultures in the remaining 3 cases (Campylobacter jejuni in 2 and Campylobacter coli in 1). All children had resolution of symptoms and improvement of inflammatory markers.

Compared with our report, no patient in Campbell’s study was diagnosed with a gastrointestinal bacterial infection, which may be related to different local epidemiologies. Another case series of 6 children received intravenous immunoglobulin as therapy for MIS-C despite not fulfilling MIS-C case definition and subsequently being diagnosed with murine typhus during an outbreak.³

Since gastrointestinal symptoms are common in MIS-C, differential diagnosis with other gastrointestinal conditions may be difficult, including acute appendicitis or gastrointestinal infections.⁴ Two of our cases had inflammatory signs on the abdominal ultrasound, including ileitis, which has been broadly described in MIS-C.⁵

Our case series show that foodborne bacterial infections should be ruled out when evaluating possible MIS-C, especially if gastrointestinal symptoms are predominant. Foodborne infections can also present with severe manifestations, including hemodynamic instability, which can resemble MIS-C. Despite the increasing concern about severe inflammatory conditions associated with SARS-CoV-2, we should continue assessing other infectious conditions within the differential diagnosis, obtaining appropriate samples and initiating timely empirical antibiotic treatment. Given the broad differential diagnosis of MIS-C, clinically stable patients with low suspicion of the disease could be admitted to the hospital under close monitorization before deciding to initiate immunomodulatory treatment.

ACKNOWLEDGMENT

The authors thank Cindy McCoig for kindly reviewing the article.