Low impact of SARS-CoV-2 infection among paediatric acute respiratory disease hospitalizations

Abstract

Objective

This study describes the characteristics of children requiring admission with an acute lower-respiratory disease (ALRD) during the SARS-CoV-2 pandemics.

Methods

Epidemiological, clinical, and microbiological data from patients with ALRD (pneumonia, bronchiolitis, bronchospasm) admitted to a reference paediatric hospital in Spain during the pandemic peak (week 11–20/2020) were prospectively analysed.

Results

110 patients were included. 7 were SARS-CoV-2(+) and they were older in comparison to SARS-CoV-2(-). Among SARS-CoV-2(+) patients, pneumonia was the main clinical diagnosis (6/7) and bronchospasm was absent. Only 1 of 29 infants diagnosed with bronchiolitis was SARS-CoV-2(+). Lower values of leucocytes, lymphocytes, neutrophils, and platelets and higher values of creatinine were found in SARS-CoV-2(+).

Human-rhinovirus/enterovirus was the main detection (11/32). There were not differences in PICU admission rates between SARS-CoV-2(+) and (-).

Conclusions

Most of the ALRD episodes identified during the pandemics were not related to SARS-CoV-2 infection. SARS-CoV-2 was mainly found causing pneumonia in older children.

Introduction

We read with interest the article by Pagani et al.¹, in which the authors report that 0 to 19-year-old children exhibit the lowest SARS-CoV-2 IgG seroprevalence among all the age-groups. In our centre, which is a 250-bed reference paediatric hospital located in Barcelona, 960 patients with fever and/or respiratory symptoms were tested for SARS-CoV-2 in the emergency department during the regional pandemic peak (week 11–20/2020) and only 56 of them were positive (6%). Thirty-one of them were admitted to hospital (Table S1, Supplementary data) and only 7 with an acute-lower respiratory disease (ALRD). Our centre captured most of regional paediatric hospitalizations due to the closure of paediatric services of general hospitals during the pandemics.

Lower respiratory tract infections are one of the leading causes of paediatric mortality and morbidity worldwide, and they also cause a high number of hospitalizations in well-developed countries.² Several reports have described that SARS-CoV-2 infection causes a much milder respiratory disease in children in comparison to adults,3, 4, 5, 6 but the real burden of SARS-CoV-2 infection in children requiring for hospital admission due to ALRD during the pandemics has not been addressed specifically. This study describes the clinical, epidemiological, and microbiological characteristics of children requiring admission with an ALRD during the first pandemic wave. Comparison of variables was made between patients with SARS-CoV-2 confirmed infection (SARS-CoV-2(+)) and those in whom SARS-CoV-2 was not detected (SARS-CoV-2(-)).

Data of patients < 18 year-old with ALRD (pneumonia, bronchiolitis, bronchospasm, or bronchopneumonia) requiring hospital admission were prospectively collected. The study was performed during the first pandemic peak in Spain (week 11–20/2020) after imposition of strict social contact measures (a state of emergency was declared on week 10/2020).⁷ Nasopharyngeal samples were collected from all children < 18 year-old with respiratory symptoms and/or fever in our emergency department. They were tested at admission using a real-time polymerase chain reaction assay for RNA detection of SARS-CoV-2. Influenza and respiratory syncytial viruses were also routinely tested using automated molecular assays until the end of these viral epidemics (week 16/2020). A real-time PCR for multiple pathogens was performed in those in whom respiratory specimens were available for re-test after routine microbiological diagnosis. Specific information about microbiological methods and definitions can be found at Supplementary data. This study was approved by the institutional ethical research committee.

411 patients were admitted during the study period, 125 (30%) with a diagnosis of ALRD. Informed consent was obtained from 110 (88%) and they were included in the study. Of them, 7 (6%) were SARS-CoV-2(+).

Median age of SARS-CoV-2(+) children was 16.9 year-old (interquartile range (IQR):11.7–17.7), being significantly higher in comparison to SARS-CoV-2(-) (3.5, IQR:0.9–7.5; p = 0.004). Only 2 patients had comorbidities (1 obesity, 1 leukaemia) in the SARS-CoV-2(+) group, whereas pre-existing respiratory conditions (recurrent viral-induced wheezing chiefly) and neurologic chronic conditions were not found despite being quite common among those SARS-CoV-2(-) (44% and 13%, respectively)). Only 3/7 SARS-CoV-2(+) patients had a household confirmed contact Table 1 .

Table 1.

Main epidemiologic, clinical, analytical, and microbiological characteristics of patients with an ALRD admitted during the pandemics in a tertiary care hospital in Catalonia.

	Total (n = 110)	SARS-CoV-2 (+) (n = 7)	SARS-CoV-2 (-) (n = 103)	p-value
Age (year-old)*	3.7 (0.9–8.4)	16.8 (11.7–17.6)	3.5 (0.9–7.5)	0.004
Sex (males)	55 (50%)	4 (51%)	51 (49%)	1
Previously healthy (n)	47 (43%)	5 (71%)	42 (41%)	0.236
- Pulmonary conditions	45 (41%)	0 (0%)	45 (44%)	0.040
- Neurologic condition	13 (12%)	0 (0%)	13 (13%)	1
- Cardiovascular conditions	4 (3.6%)	1 (14%)	3 (3%)	0.234
- Haematologic malignancy	2 (2%)	1 (14%)	1 (1%)	0.124
Ethnicity (Caucasian) (n)	64 (58%)	5 (71%)	59 (57%)	0.657
Household confirmed contacts (n with > 1 confirmed contact)	6 (5%)	3 (43%)	3 (2%)	0.003
Symptoms at hospital admission (n):
- Cough	105 (95%)	6 (86%)	99 (96%)	0.285
- Wet cough	61 (55%)	2 (29%)	59 (57%)	0.238
- Fever	80 (73%)	7 (100%)	73 (71%)	0.186
- Gastrointestinal	21 (19%)	2 (29%)	19 (18%)	0.618
- Exanthem	4 (4%)	1 (14%)	3 (3%)	0.236
Time-lag from the onset of symptoms to hospital admission (days)*	4 (2–7)	7 (4–9)	4 (2–7)	0.052
HbSat at admission (%)*	94 (92–96)	94 (91–96)	94 (92–96)	0.909
Chest-X-ray at admission (n/total in whom the test was performed)
- Normal	12 / 80	0 / 7	12 / 72	0.587
- Lobar pneumonia	22 / 80	4 / 7	18 / 72	0.089
- Interstitial pneumonia	39 / 80	2 / 7	37 / 72	0.431
- Pleural effusion	6 / 80	1 / 7	5 / 72	0.442
Required respiratory support during admission (n):
- NNCC	78 (71%)	4 (57%)	74 (72%)	0.413
- HFNO	14 (13%)	0 (0%)	14 (14%)	0.592
- NIV	7 (6%)	1 (14%)	6 (6%)	0.380
- MV	2 (2%)	1 (14%)	1 (1%)	0.125
Length of fever (days)*	4 (2–6)	5 (3–11)	4 (2–6)	0.231
Length of oxygen requirements (days)*	2 (1–4)	4 (2–11)	2 (1–4)	0.097
PICU admission (n)	11 (10%)	2 (29%)	9 (95)	0.145
Need for inotropes (n)	2 (2%)	2 (29%)	0 (0%)	0.004
Other viral pathogens (n/total tested)
- RSV	4 / 69	0 / 4	4 / 65	1
- Parainfluenza 1	1 / 32	1 / 4	0 / 28	0.125
- Parainfluenza 4	1 / 32	0 / 4	1 / 28	1
- Influenza A	3 / 70	0 / 4	3 / 66	1
- Influenza B	6 / 70	1 / 4	5 / 66	0.307
- Pre-pandemic coronaviruses	1 / 32	0 / 4	1 / 28	1
- HRV/EV	11 / 32	1 / 4	10 / 28	1
- Adenovirus	2 / 32	0 / 4	2 / 28	1
- Metapneumovirus	2 / 32	0 / 4	2 / 28	1
Clinical classification:
- Bronchiolitis	29 (26%)	1 (14%)	28 (27%)	0.520
- Bronchospasm/viral-induced wheezing	33 (30%)	0 (0%)	33 (32%)	0.134
- Viral pneumonia	23 (21%)	4 (57%)	19 (18%)	0.030
- Bacterial suspected pneumonia	25 (23%)	2 (29%)	23 (22%)	0.737
Analytical features at admission*:
- Haemoglobin (g/L)	11.7 (10.9–12.7)	12,4 (11.6–12.7)	11,7 (10.8–12.9)	0.417
- Leukocytes (cells x 109 /L)	11.6 (6.9–18.2)	5.5 (3.1–7.9)	13.8 (7.2–19.4)	0.001
- Lymphocytes (cells x 109 /L)	2.4 (10.5–3.7)	0.7 (0.4- 1.7)	2.8 (1.1–4.1)	0.007
- Neutrophiles (cells x 109 /L)	7.0 (3.9–12.5)	3.4 (1.5–5.8)	7.5 (4.3–12.6)	0.010
- Platelets (cells x 109 /L)	338 (243–448)	164 (127–262)	355 (264–468)	0.012
- C-reactive protein (mg/L)	37 (13–81)	48 (15–138)	36 (12–82)	0.662
- Procalcitonin (ng/mL)	0.24 (0.08–0.95)	0.06 (0.04–2.42)	0.25 (0.09–1.01)	0.098
- D-dimer (mg/L)	0.95 (0.67–1.91)	0.80 (0.49–4.22)	1.09 (0.76–2.58)	0.602
- Ferritin (µg/L)	183 (88–449)	350 (144–2580)	99 (85–269)	0.072
- Alanine transaminase (UI/L)	13 (8–20)	38 (17–89)	13 (8–17)	0.064
- Aspartate Aminotransferase (UI/L)	24 (18–38)	50 (21–57)	24 (18–33)	0.229
- Creatinine (mg/dL)	0.51 (0.42–0.64)	0.77 (0.67–0.93)	0.48 (0.40–0.59)	0.007
Hospital stay (days)*	3 (2–5)	11 (3–16)	3 (2–5)	0.024

^⁎median (interquartile-range)

Proportions between the groups (SARS-CoV-2(+) Vs SARS-CoV-2(-)) were compared using Pearson Chi-square o Fisher exact test. For continuous variables, the Mann-Whitney U test was performed.

NNCC: nasal cannula; HFNO: high-flow nasal cannula; NIV: non-invasive ventilation; IMV: invasive mechanical ventilation; RSV: respiratory syncytial virus; HRV: human rhinovirus; EV: enterovirus.

There were not significant differences in symptoms between SARS-CoV-2(+) and SARS-CoV-2(-) patients Table 1. Nevertheless, SARS-CoV-2(+) children required admission after 7 days (IRQ:4–9) since symptoms onset and this time-lag tended to be higher than in those SARS-CoV-2(-) (4, IQR:2–7; p = 0.052). Regarding physical examination, bronchospasm was not observed in SARS-CoV-2(+) (0/7 vs 77/103, p <0.001). Pneumonia was the main clinical diagnosis in SARS-CoV-2(+) children (6/7). Differences in chest-X-ray radiologic patterns were not observed between those SARS-CoV-2(+) and (-). Lower values of leucocytes, lymphocytes, neutrophils, and platelets and higher values of creatinine were found in SARS-CoV-2(+) Table 1. The only patient with bronchiolitis and SARS-CoV-2(+) was a new-born male who required oxygen for less than 24 h.

Seventy patients were tested for influenza infection, 69 for RSV and 32 for multiple viral pathogens. Viral codetection was observed in 1 of 4 SARS-COV-2(+) patients (1 with influenza B, human-rhinovirus/enterovirus (HRV/EV) and parainfluenza), while viral codetection was observed in 25 of 28 SARS-CoV-2(-) patients tested for multiple viral pathogens. Among SARS-CoV-2(-) patients, HRV/EV was the main viral detection (10/28). There were not significant differences in weekly rates of respiratory virus detections across the study period. Eleven patients, all in the SARS-CoV-2(-) group, had a confirmed bacterial pneumonia (6 Mycoplasma pneumoniae, 2 Streptococcus pneumoniae, 2 Gram-negative bacteria and 1 Staphylococcus aureus).

Despite there were not differences in PICU admission rates between SARS-CoV-2(+) and (-) children, two SARS-CoV-2(+) patients required inotropic support whereas none of SARS-CoV-2(-) required this treatment (p = 0.004). Hospital stay was longer (11 days (IQR:3–16) vs 3 (IQR:2–5); p = 0.024) and 1 patient died in the SARS-CoV-2(+) group. This was an 11 y-old boy with an influenza B coinfection and a graft-versus-host disease after an allogenic transplant due to an acute lymphocytic leukaemia. Table S2, Supplementary data.

Our results suggest that, despite conducting the study during the pandemic peak, those children who required hospital admission were infected more often by other respiratory microorganisms. Pneumonia was the main clinical diagnosis among SARS-CoV-2(+) children, and they showed indistinguishable clinical and radiological characteristics at hospital admission from those SARS-CoV-2(-), as observed by others.⁴ Nonetheless, the two groups showed different analytical features and SARS-CoV-2(+) patients were significantly older.

There are some cases reports of infants with bronchiolitis in whom SARS-CoV-2 was the only infection detected.⁸ Literature is scarce addressing this topic, but SARS-CoV-2 does not seem to be a main trigger of bronchiolitis.⁹ It caused only 1/29 episodes in our series. In the pre-COVID-19 era, viral coinfection in infants with bronchiolitis caused by coronaviruses was very common (85%).¹⁰ Since the pandemic did not coincide with the bronchiolitis season, the behaviour of the emergent virus in coinfection with other well-known triggers of bronchiolitis is uncertain.

The main limitation of this study was its observational design, as heterogeneous tests were performed in each patient depending on their clinical condition and the epidemiological context. Most patients with ALRD had fever (73/103), therefore a presumable viral infection could be assumed, however a low rate of patients was tested for multiple pathogens. On the other hand, a low number of SARS-CoV-2(+) respiratory patients was found.

To conclude, most of the ALRD episodes identified during the pandemics and under strict lockdown measures were not related to SARS-CoV-2 infection in this study. SARS-CoV-2 was mainly found causing pneumonia in older children, whereas the impact of SARS-CoV-2 infection among ALRD hospitalizations was low in young children.

Authorship

All authors have made substantial contributions to all the following: the conception and design of the study, the acquisition, analysis, and interpretation of data, drafting the article and revising it. The Kids-Corona Paediatric Hospitalist group carried out a very important task collecting data and taking care of patients.