INTRODUCTION

Older adults and those with chronic comorbidities have been severely affected by COVID-19 ¹. The involvement was mostly mild in children ², with the exception of cases of pediatric multisystem inflammatory syndrome (SIM-C) associated with COVID-19 ³.

Several variants emerged during the pandemic. In Peru, the lambda and gamma variants (during the second wave) were associated with higher mortality and clinical involvement in adult and pediatric patients compared to the original variant ⁴. Then, the third wave of COVID-19 appeared in Peru at the beginning of 2022, in which the predominant variant was omicron (B.1.1.529) and its lineages ⁴. Several countries reported a high number of hospitalized children with omicron initially ^5-8; however, later studies showed a lower percentage of hospitalizations and less severity compared to the delta variant ^9-11. These reports differ, and no publications on omicron were found in Latin American pediatric centers. It is important to have data from Latin America, since what is described in other regions may not reflect what happens in countries such as Peru; due to their socioeconomic characteristics, health system, vaccination coverage, and strict public restrictions against COVID-19 ¹².

Therefore, this study aimed to describe the characteristics of pediatric patients hospitalized with COVID-19 during the third wave (omicron variant), at the Edgardo Rebagliati Martins National Hospital (HNERM) in Lima, Peru.

THE STUDY

A retrospective cohort was carried out. Patients older than 28 days and younger than 14 years, hospitalized in the COVID-19 area (common ward and the HNERM intensive care unit [ICU]) were included. Patients with COVID-19 who became infected inside or outside hospital were also included.

In Peru, the first wave of COVID-19 occurred from March to December 2020, the second wave from January to June 2021, which was mostly attributed to the lambda variant ⁴, and the third wave, between the first week of January until the first days of April 2022 ^4,13 with predominance of the omicron variant ⁴. We collected data from patients hospitalized throughout the third wave at HNERM.

The diagnosis of SARS-CoV-2 infection was made by antigen test or reverse transcriptase-polymerase chain reaction (RT-PCR). Genomic sequencing was not performed. We used the criteria from the Centers for Disease Control and Prevention (CDC) for the diagnosis of SIM-C ¹⁴. Data were collected from patients’ medical records (physical and electronic), from hospital admission to discharge or death.

The following variables were collected: a) epidemic wave of COVID-19, age, sex, vaccination against COVID-19, origin of the patient (emergency or intrahospital), test for diagnosis of SARS-CoV-2, comorbidities; b) reason for hospitalization, upper (croup, pharyngitis, and cold) and lower (pneumonia, wheezing, and bronchiolitis) respiratory findings based on medical diagnosis, systemic, gastrointestinal, neurological, and mucocutaneous symptoms, dysuria, cervical adenopathies, and days with symptoms before diagnosis; c) pneumonic bacterial superinfection (medical diagnosis), coronary aneurysm (by echocardiography), admission to the intensive care unit (ICU), days in ICU, requirement for invasive mechanical ventilation (IMV), days on IMV, use of high flow nasal cannula (HFNC), inotrope/vasopressor support, macrophage activation syndrome (MAS) ¹⁵, hospital days, death; and d) administered medications.

Data were stored in a Microsoft Excel ® file. Two researchers reviewed and debugged the final database. Statistical analysis was performed with STATA version 16 (StataCorp LP, College Station, Texas, United States). Numerical variables were reported using the mean and standard deviation or median and interquartile range (IQR) after evaluating the normal distribution of the data, which was performed by visual inspection of the histogram as well as the Shapiro-Wilk test. Categorical variables were reported by using absolute and relative frequencies.

This project was approved by the HNERM Ethics Committee (approval code: 832-2022-069). The confidentiality of the patients included in the study was preserved. The registration code for health research projects (PRISA) of this study is EI00000002918.

FINDINGS

We found that 488 pediatric patients were hospitalized in the COVID-19 area of the HNERM from the beginning of the first wave to the end of the third wave. During the first wave (March to December 2020), 206 patients were hospitalized (42.2% of all who were hospitalized during the pandemic), 100 (20.5%) during the second wave (January to June 2021), 60 (12.3%) during the second and third wave, and 122 (25%) during the third wave (January to April 2022).

The number of hospitalizations increased rapidly during the third wave; 95 (77.9%) out of 122 patients were hospitalized only during the first month of this wave. Hospitalizations during the first months of the other waves were significantly less; one patient (0.5%) during the first month of the first wave and 11 (11.0%) during the first month of the second wave (Figure 1).

Figure 1 Pediatric patients hospitalized with COVID-19 per month at the Edgardo Rebagliati Martins National Hospital, 2020-2022 (n=488). 

DISCUSSION

This study showed a rapid increase in patients hospitalized due to COVID-19 during the third wave compared to previous waves, who mostly progressed favorably (6.6% in the ICU, 4.9% in IMV, 5.7 % used vasoactive agents, and 1.6% died). Also, the number of hospitalizations was greater during the third wave than during the second wave.

The rapid increase in hospitalizations was probably due to the high number of cases that occurred during the third wave of the pandemic in Peru, which may be due to the fact that the omicron variant is very contagious and spreads quickly ⁴. In addition, other factors could have contributed, such as comorbidities (because HNERM is a national referral hospital), the lack of vaccination in children (at that moment, only people older than 12 years old were to be vaccinated), relaxation of biosecurity measures (less social distancing, non-mandatory use of masks in public places, and fewer restrictions to access them), and the probable reservoir of uninfected pediatric patients.

Consistently, other studies have reported a large increase in pediatric hospitalizations ^5-8, but studies that analyzed the delta variant reported fewer hospitalizations: 1.76% ⁹, 2.2% ¹⁰, and 10% ¹⁶ in the United States, and 2.3% ¹¹ in Spain; as well as fewer emergency admissions, ICU admissions and IMV requirements ^9-11.

Most hospitalized patients were male, which is similar to what was reported by other studies ^5,9,11,17; on the other hand, most were school-age children, which differs with other studies that reported a higher frequency of hospitalizations in children under one year of age ^5,9,11. We found that the median age was similar according to the clinical involvement (respiratory and SIM-C). Regarding comorbidities, most previous research reports patients without chronic diseases ^9,10,17; studies similar to ours report a predominance of these as well ⁵. Those patients with moderate to severe respiratory involvement had a higher frequency of comorbidities. However, despite the fact that half of our patients had some comorbidity and none completed the vaccination schedule for COVID-19, most progressed favorably, similar to what has been described in other countries ^5,11,17.

Respiratory conditions were the main reason for hospitalization, similar to what has been reported previously ^11,16. Most studies report fever, gastrointestinal and respiratory symptoms ^5,8,11, which is similar to our findings. Previous studies have also reported neurological involvement (7-20%) due to omicron ^5,11,16,17, even in patients with no history of epilepsy. In our series, 27 (23.5%) patients presented neurological symptoms, but eight had a previous diagnosis of epilepsy.

Wheezing occurred in 13 (11.3%) patients; however, six had a diagnosis of asthma. Few studies have reported bronchospasm ¹¹ due to omicron, therefore, more studies are required to assess whether this variant induces wheezing. Croup symptoms were also evidenced in seven (6.1%) patients. Croup symptoms in 12% have been reported in previous studies ¹⁶, as well as moderate and severe involvement ^17,18, making it important to consider omicron as a differential diagnosis. On the other hand, the number of cases with SIM-C was lower than that found in our hospital during the first and second waves ^3,19.

Ivermectin and hydroxychloroquine were not used in any case, which coincides with the evidence available from the study period ^20,21 as well as with the evidence-based clinical practice guideline of the Institute for Technology Assessment in Health and Research (IETSI) of the social security of Peru (EsSalud) for the management of COVID-19 in children ²². However, broad empirical use of antibiotics was found (43.4%). This could be due to the suspicion of superinfection, which was low and probably the reason why only 7% received complete antibiotic treatment. The mostly used medication were corticosteroids (27.1%), although it is unknown whether these were used for any other indication unrelated to COVID-19.

On the other hand, unfavorable outcomes were not frequent, when compared to what was described during the first wave in HNERM ¹⁹. With omicron, hospital stay was shorter (median 6 vs. 10 days for respiratory involvement), ICU admissions were less (6.6% vs. 13%), vasopressors were less used (5.7% vs. 13%), fewer patients died (1.6% vs. 4%), and less IMV was required (4.9% vs. 14%). The favorable progression of patients infected with the omicron variant coincides with studies from the United States and Europe ^9-11, which report less severity compared to delta; and also coincides with what was described in Peru (at the population level) with lower mortality in pediatric patients in the third wave ⁴. However, severe cases were reported, and greater severity has been reported by studies that compared other influenza and parainfluenza viruses ¹⁷.

The main strength of our study is that these results are representative of the largest EsSalud hospital, as well as the fact that the data was reviewed and filtered before being analyzed. However, being a retrospective study, some medical records may contain errors. Likewise, our results could differ from other hospitals in Lima. In addition, patients with nosocomial infection were included, but these represent only 8.9% of the total number of participants. Another limitation is that these cases occurred during the third wave in Peru, in which the omicron variant (B.1.1.529) predominated nationwide, therefore genomic sequencing was not performed ⁴.

In conclusion, hospitalizations due to COVID-19 increased rapidly during the third wave in the HNERM, when compared to previous waves; most patients were admitted due to respiratory conditions, and most progressed favorably. Antibiotics and corticosteroids were the most used drugs. Genomic and clinical surveillance of SARS-CoV-2 variants is necessary, as well as encouraging follow-up to assess post-COVID-19 syndrome in children.