Financial support:
This study did not avail any funding.
To the Editor:
Food protein-induced enterocolitis syndrome (FPIES), a
non-immunoglobulin E-dependent allergic disease, is diagnosed based on
clinical symptoms or positive results of an oral food challenge (OFC)
test.1,2 The lack of disease-specific biomarkers
render early diagnosis difficult. Solid food FPIES often develops 5–12
months after birth and may be misdiagnosed as various infectious
diseases that are common during that period. Therefore, it is necessary
to establish new biomarkers for FPIES diagnosis. In this study, we
measured serum thymus and activation-regulated chemokine
(TARC),3 and general biomarkers in the acute phase of
FPIES, infectious gastroenteritis, and sepsis to evaluate the usefulness
of each biomarker for FPIES diagnosis. We also examined the relationship
between disease severity and serum TARC levels in patients with FPIES.
This case-control study evaluated 20 episodes (9 emergency outpatient
visits and 11 positive oral food challenges) of 13 pediatric patients
with FPIES (8 egg yolk, 1 egg white, 1 whole egg, 1 wheat, 1 rice, and 1
short-neck clam) (FPIES group) and 40 age-matched patients with
infectious gastroenteritis or sepsis (control group) between April 2019
and March 2021 (Figure S1, Table
S1_Supporting Information). This
study was approved by the Ethics Review Committee of the Saitama Medical
Center, Jichi Medical University (approval number S19-077). Written
consent was obtained from the patient’s parents.
Patient background was evaluated. In patients with FPIES, the causative
antigen, age of onset, age at event, symptoms, and severity (SuppInfo)
of each episode were evaluated.
Serum TARC levels were quantified using a chemiluminescent enzyme
immunoassay via a HISCL® TARC Assay Kit (Sysmex, Hyogo, Japan). For
patients who visited the emergency department, serum TARC values were
measured at the time of outpatient visit. For OFC cases, serum TARC
values were measured before antigen ingestion and at 6 h, 24 h, and 1
week after ingestion. Peak serum TARC values were
evaluated. Since normal TARC
values differ by age, modified TARC values (serum TARC value − normal
mean for each age4) also were evaluated (SuppInfo).
Concurrently, complete blood count (CBC) including white blood cell,
neutrophil, eosinophil, and platelet, C-reactive protein (CRP), lactate
dehydrogenase, venous blood pH, HCO3-, and methemoglobin
(%) were measured. OFC was
conducted via the open, single-ingestion method (SuppInfo).
The patients with FPIES were five boys and eight girls. The age at onset
ranged 0.5–3.0 (median 0.7) years. In the FPIES group, the age for 20
episodes analyzed was 0.5–6.2 (median 1.1) years. No patient had eczema
at the time of the episode. Of these, 7 were severe, 8 were moderate,
and 5 was mild in severity (Table 1). The control group had 20 patients
with infectious gastroenteritis and 20 patients with sepsis aged
0.1–6.0 (median 1.5) years. There were no cases of eczema (Table
S1_SuppInfo).
Table 2 shows the blood test results of the FPIES and control groups.
The serum TARC (pg/mL) and modified TARC (pg/mL) levels were
significantly higher in the FPIES group than in the control group
(p < 0.001; Figure 1A). The eosinophil count was also
significantly higher in the FPIES group. There were no significant
differences regarding other parameters between the groups.
The results of the receiver operating characteristic analysis for each
parameter for FPIES diagnosis are shown in Figure 1B and Table
S3_SuppInfo. The area under the curve (AUC) of the serum TARC was 0.946
(95% confidence interval [CI], 0.871–1.000), with a 95%
sensitivity, 95% specificity, and a cutoff value of 1062 pg/mL (Youden
Index). The AUC of serum TARC and modified TARC values were higher than
those of other parameters.
In univariate analysis, serum TARC levels were a significant factor for
FPIES diagnosis, with an odds ratio (OR) of 29.5 (95% CI 4.3–200.0,p < 0.001) per 1000 pg/mL increase. The age-adjusted
TARC OR per 1000 pg/mL increase was 23.4 (95% CI 3.3–162.0, p =
0.001).
The median (range) serum TARC levels for patients with severe (n = 7),
moderate (n = 8), and mild (n = 5) FPIES were 6320 (2911–8147), 2020
(1084–4097), and 1087 (410–1678), respectively. Significant
differences were found between the severe, moderate, and mild groups
(p = 0.007, p = 0.037) (Figure 3C). The correlation
coefficients between the serum TARC value and number of vomiting
episodes and vomiting time were R = 0.570 (p = 0.008) andR = 0.699 (p < 0.001), respectively, showing a
moderate correlation (Figure S2).
Figure 1D shows the transition of the 11 OFC-positive TARC values. The
median (range) serum TARC (pg/mL) was 555 (227–1702) before ingestion,
1283 (410–3821) after 6 h, and 3748 (886–7816) after 24 h, increasing
significantly over time. After one week, the median (range) TARC
decreased to 961 (468–3359).
This study revealed that serum TARC levels are more useful than other
biomarkers, such as CBC, CRP, and methemoglobin levels for FPIES
diagnosis. We also found a correlation between the severity of FPIES and
serum TARC.
In a previous study comparing acute phase biomarkers in patients with
FPIES, infectious gastroenteritis, and sepsis, patients with FPIES had
higher white blood cell, eosinophil, platelet counts, and lower CRP
levels. However, the usefulness of these biomarkers for FPIES diagnosis
is unclear.5 We reported that neonates with FPIES had
significantly higher methemoglobin levels than those with other diseases
with vomiting. Those results were limited to the neonatal period with
less overlap of methemoglobin levels in both groups.6There are reports measuring various cytokines, although specific
recommendations have not been determined.1
TARC is a Th2 type chemokine produced by dendritic cells, endothelial
cells, keratinocytes, and fibroblasts.3 Although TARC
is used in diagnosing and assessing atopic dermatitis, some reports
associate it with food allergies.4,7,8 Recently, we
have evaluated serum TARC levels of the acute phase in patients with
FPIES and infectious gastroenteritis and reported that FPIES exhibits
high TARC levels.9 However, because the control group
was small and included patients with infectious gastroenteritis alone,
the predictive accuracy could not be clarified. Here, we clarified that
serum TARC value is useful for FPIES diagnosis.
A longitudinal evaluation of TARC levels showed an increase at 6 h
post-ingestion, with a further increase 24 h later. After one week,
serum TARC levels had decreased, although they had not normalized.
Therefore, TARC measurement after 24 h is recommended for the diagnosis
of FPIES. Furthermore, serum TARC levels in patients with FPIES were
positively correlated with symptom severity, vomiting frequency, and
vomiting duration. Since there were no test items that reflect FPIES
severity, we expect that the TARC value after vomiting may be used as an
objective evaluation item for FPIES severity.
This study had some limitations.
First, all participants had acute solid food FPIES. Patients with cow’s
milk FPIES or chronic FPIES were excluded. Further studies on these are
needed. Second, our sample size was small. Nonetheless, there was a
significant difference in the TARC levels between patients with FPIES
and the controls, indicating that TARC level elevation is specific to
FPIES.
In conclusion, post-emetic serum TARC levels are useful for FPIES
diagnosis. Early diagnosis may be possible by measuring TARC after
vomiting. It may also be useful for objective severity assessment.
Further larger-scale research is needed.
Keywords: biomarkers; food allergy; food protein–induced
enterocolitis syndrome; infectious gastroenteritis; pediatrics; sepsis