Clinical characteristics, laboratory findings, and tolerance acquisition in infants with cow's milk protein allergy in a private center in Lima, Peru for the period 2021–2022

1 INTRODUCTION

Food allergies (FA) represent a growing healthcare problem globally, including in developing countries, likely attributable to the adoption of Westernized lifestyles in these regions.¹ Additionally, epidemiological data on FA in these regions is limited due to low public awareness, inadequate physician education and training, limited access to healthcare and diagnostic tools, shortage of allergists, and overlap with other conditions such as malnutrition and childhood illnesses.^1-3 It is important to mention that overdiagnosis is also reported; for example, in Peru, 40.9% of parents reported FA in their children, as documented in a national reference center for allergic diseases.⁴

Cow's milk protein allergy (CMPA) is the most frequently diagnosed FA among children younger than 1 year. However, precise data in developing countries remains elusive, as several studies rely on self-reports, often lacking comprehensive assessments including family history, clinical characteristics, relationship with food exposure, utilization of FA tests, or conducting a double-blinded placebo-control food challenge (DBPCFC).⁵ Likewise, although using fecal calprotectin (FC) and eosinophil-derived neurotoxin (EDN) in evaluating intestinal inflammation and diagnosing non-IgE mediated CMPA is promising, their utilization remains limited, with studies showing variable results.^{6, 7}

Therefore, our study aims to describe the clinical features, laboratory findings, and outcomes of oral challenges and tolerance acquisition in Peruvian infants diagnosed with CMPA younger than 1-year-old.

2 METHODOLOGY

This single-center observational study was conducted in infants (<1 year) diagnosed with CMPA who attended a private allergic diseases center in Lima, Peru, in 2021. The study protocol received approval from the Ethics Committee of the Instituto Nacional de Salud del Niño, Breña, Peru.

3 RESULTS

A total of 64 patients, with a median age of 3 months (IQR: 1–5 months), were included in the study, with approximately 60% being male. Malnutrition was observed in 50 patients (78.1%), and C-section delivery occurred in 68.7% of cases. Before CMPA diagnosis, 33 patients (51.6%) were exclusively breastfed. Additionally, 56 patients (87.5%) had a family history of atopy, with 39% reporting a family history of asthma and 71.9% indicating a family history of allergic rhinitis. Further demographic data is presented in Table 1.

4 DISCUSSION

Our study describes infants under 1 year of age diagnosed with CMPA through stringent criteria. These infants were predominantly delivered via C-section, skewed toward males, with a notable proportion experiencing malnutrition at diagnosis. Additionally, they exhibited familial allergic respiratory diseases and a comparable distribution concerning diet type at diagnosis. This clinical profile is similar to other South American regions, as reported by Errazuriz et al., highlighting a predominantly male population with familial atopy history and a similar diet distribution, albeit with a healthier status in their cohort.¹² Similar observations were made in a prospective study conducted in a private center in Chile.¹³ However, in regions such as France, Kalach et al. reported similar clinical patterns but with a lower proportion of exclusive breastfeeding at diagnosis (33%, less than that in our study).¹⁴ Exclusive breastfeeding, slightly more prevalent in our CMPA patients' diagnosis (51.66% vs. 48.44%), has been shown to exert a protective effect against CMPA development compared with early infant formula use, as evidenced by a systematic review.¹⁵

The majority of infants in our study were diagnosed with CMPA within the first 5 months of life, primarily presenting with gastrointestinal symptoms, notably colic and diarrhea. These findings align with reports from other settings.^{12, 14} Moreover, several gastrointestinal manifestations observed in CMPA clinical presentations, such as regurgitation, colic, or constipation, may resemble functional conditions common in healthy infants. Hence, a focused clinical history and examination are crucial for distinguishing between physiological and pathological conditions such as CMPA. While all patients in our study with gastrointestinal symptoms underwent DBPCFC, facilitating confirmation of diagnosis may not always be feasible in every clinical scenario, underscoring the reliance on medical history for accurate diagnosis.¹⁶

Our study evaluated nutritional status at diagnosis, revealing a significant proportion (35.93%) with varying degrees of malnutrition, reflecting challenges posed by public health systems, poverty, and insufficient educational strategies in Latin American countries. Remarkably, 1 year after CMPA diagnosis, the prevalence of malnutrition considerably decreased to 17.18%, emphasizing the pivotal role of early CMPA detection.¹⁷

Exploration of IgE-mediated responses revealed six infants testing positive for cow's milk among the 64 infants (9%). Two of these six infants presented with anaphylaxis, characterized by predominant skin and respiratory involvement (hives and bronchospasm). Thus, the majority of our CMPA patients (91%) exhibited a non-IgE-mediated type, consistent with findings reported in multiple publications.^{18, 19} Studies in Latin America reporting IgE-mediated versus non-IgE-mediated CMPA frequency display variability. Arancibia et al. reported exclusively non-IgE-CMPA cases in their cohort (20 patients) using food challenges,¹³ whereas Mehaudy et al. documented a 27.6% frequency of IgE-mediated CMPA in Argentina.²⁰ Discrepancies may stem from differences in age at diagnosis, diagnosing specialists, patient demographics, and socioeconomic status. In our study, most patients were diagnosed before 5 months of age, referred by allergists, and hailed from socioeconomic backgrounds above the population average.

Few studies in Latin America have utilized DBPCFC for CMPA diagnosis. Errazuriz et al. conducted DBPCFC in 14% of their patients,¹² whereas Arancibia et al. performed it in all their patients, albeit employing open oral challenges.¹³ Goncalves et al. conducted single-blinded oral challenges in Brazil.²¹ In our study, DBPCFC was performed in 91% of patients and in all suspected non-IgE-mediated CMPA cases, yielding an 86% positivity rate. DBPCFC positivity rates vary widely in the medical literature,^{22, 23} likely influenced by factors such as duration of prior elimination diet, age at diagnosis, predominant CMPA symptomatology, and clinician's initial suspicion accuracy, among others.

The use of fecal biomarkers in diagnosing non-IgE-mediated CMPA is an ongoing area of research. Trillo et al. reported that an FC level below 138 μg/g could exclude non-IgE-mediated CMPA, suggesting its utility for prospective follow-up at 1- and 3-months postdiagnosis.²⁴ Similarly, Rycyk et al. found FC (88.9%) and EDN (84%) prevalence in 27 patients with non-IgE-mediated CMPA.⁶ However, Roca et al. did not find significant differences in FC or EDN levels among 30 patients with non-IgE–mediated CMPA, though not all patients underwent oral challenges.⁷ Our study evaluated all non-IgE–mediated CMPA patients for FC and EDN, revealing significant FC differences in cases presenting with regurgitation/vomiting. Significant EDN differences were also noted in patients experiencing recent symptoms (within 2 weeks) of diarrhea and regurgitation/vomiting. Furthermore, a significant median reduction in FC levels was observed after 1 year of observation. Larger-scale studies with specific designs are warranted to validate our findings.

Tolerance acquisition to cow's milk 1-year postdiagnosis mirrored findings from other regions.¹² In our analysis of patients utilizing nutritional replacement formulas, those using extensively hydrolyzed casein-based formulas exhibited greater tolerance acquisition (94%) and FC level reduction compared with those using other formulas (71%), though statistical significance was not reached (p values of 0.08 and 0.12, respectively). This trend, identified in our study, has been supported by previous clinical trials.²⁵

The strengths of our study include its comprehensive evaluation of infants under 1 year of age diagnosed with CMPA, including utilization of DBPCFC in all patients with non-IgE-mediated CMPA. Furthermore, our study contributes novel insights by exploring fecal biomarker levels in non-IgE-mediated CMPA, filling a crucial gap in the literature regarding diagnostic approaches in this subset of patients. Additionally, our study provides valuable data on nutritional status and tolerance acquisition, shedding light on the effectiveness of different formula types in managing CMPA. However, our study also possesses some limitations. Being a single-center observational study, our findings may not be generalizable to the broader population and the small sample size analyzed. In addition, our study explores fecal biomarkers in non-IgE-mediated CMPA, the interpretation of these biomarkers remains subject to ongoing research and validation, highlighting the need for further investigation in larger, prospective studies.

In conclusion, our study is among the few in Latin America that elucidates the clinical characteristics, laboratory tests, DBPCFC, and tolerance outcomes in infants under 1 year of age with CMPA. Furthermore, it is likely the first in the region to explore fecal biomarker levels in non-IgE-mediated CMPA diagnosed with DBPCFC. Our findings provide a foundation for future research investigating CMPA diagnosis and tolerance acquisition, considering the unique challenges faced in our region.

AUTHOR CONTRIBUTIONS

César Galván Calle: Conceptualization; data curation; formal analysis; funding acquisition; investigation; methodology; project administration; resources; software; supervision; validation; visualization; writing—original draft; writing—review and editing. Cecilia Díaz-Vásquez: Data curation; formal analysis; writing—original draft; writing—review and editing. Wilmer Córdova-Calderón: Data curation; formal analysis; writing—original draft; writing—review and editing. Juan Gómez de la Torre: Data curation; formal analysis; writing—original draft; writing—review and editing. Edgar Matos-Benavides: Data curation; formal analysis; writing—original draft; writing—review and editing. All authors contributed equally to the research process, data analysis and drafting of the article.