Etiologies and Prognostic Factors of Dyspnea in Infants at the University Hospital Center (CHU) of Bouaké (Ivory Coast) ()
1. Introduction
Dyspnea is defined as difficult and labored breathing. It can be accompanied by a change in the respiratory rate and affect only one of the breathing times. Its origin is a cardiac or respiratory origin [1] . Its frequency was high in infants [2] [3] [4] [5] . However, there are few data on the etiologies of dyspnea in general and especially those relating to infants. The causes of these dyspneas are related to 2 mechanisms: obstructive dyspnea and non-obstructive dyspnea [6] . In our country, there are few works on dyspnea in infants, hence the interest for us to carry out this survey. The objective of our study was to identify the epidemiological characteristics, etiologies and evolutionary aspects of dyspnea in infants hospitalized in our department and to make recommendations with the aim to improve their management.
2. Patients and Methods
Our study took place in a tertiary level hospital, the CHU of Bouaké, a city located in the Center-North of Côte d’Ivoire, 350 km from the economic capital, Abidjan. The Bouaké University Hospital is the reference center and last resort in the Center-North zone of Côte d’Ivoire. It welcomes patients coming from this area and especially those from the municipality of Bouaké. This CHU has several departments, including that of Pediatrics, where the study took place, which included a neonatal block, 3 hospitalization rooms, a therapeutic nutrition unit, an awakening room and a continuous monitoring unit. The latter, made up of 8 beds, received children admitted to the pediatric ward for vital distress and had an occupancy rate of 200%. Our study population consisted of infants admitted to pediatric emergencies and hospitalized for dyspnea. Immediate management consisted of hospitalization in a continuous monitoring unit, nasopharyngeal clearing (either by aspiration or by washing with physiological serum), administration of oxygen (2 L/min by nasal prongs) to maintain saturation greater than 94% and a hydroelectrolytic contribution. Etiological treatment was initiated in parallel and regular monitoring was instituted to ensure improvement in vital parameters. Included in the study were all infants aged 29 days to 23 months, admitted to the continuous monitoring unit (USC) for dyspnea. During the study period, we identified 193 infants. We did not retain 34 hospitalized for severe malaria and 07 with incomplete patient records. Our sample was therefore 152 infants. Some infants came from regional hospitals and general hospitals located at a distance of a few dozen to several hundred km from the city of Bouaké.
This was a descriptive and analytical retrospective study of infants hospitalized for dyspnea over a period of one year from January 1 to December 31, 2020. Our survey was carried out on databases collected from individual patient records. Data collection was based on a pre-established survey form. The parameters of the study were sex, age, month of admission, origin, time to admission, mode of breastfeeding, vaccination status, existence of underlying pathology, diagnosis at hospitalization, the length of hospitalization and the evolutionary modalities. We hospitalized infants with dyspnea on admission. We evaluated their ventilation, which is based on 4 parameters: respiratory rate, work of breathing, tidal volume and oxygenation. Indeed, tachypnea varies with age. It is >60/min before the age of 3 months, >50/min between 3 and 6 months and >40/min beyond; the evaluation of the work of breathing is based on the search for signs of struggle (thoraco-abdominal rocking, indrawing, flapping of the wings of the nose, xiphoid funnel, expiratory grunting); the tidal volume is evaluated by thoracic expansion and auscultation; oxygenation is assessed by staining the integuments and mucous membranes as well as oxygen saturation (SaO2) [6] . The diagnosis of pulmonary infection was suggested on the association of fever, polypnoea and/or localized or diffuse crackles on auscultation [7] . The positive diagnosis of bronchiolitis was evoked in the presence of more or less febrile slowing expiratory dyspnoea associated with cough and more or less extensive wheeze auscultation, as well as crackles in the event of associated alveolitis [8] . We considered as hypotrophic, that is to say low weight for gestational age, newborns with a birth weight lower than the tenth percentile [9] [10] . We recommend, in accordance with the recommendations, exclusive breastfeeding for 6 months, dietary diversification between 6 and 12 months and a totally diversified diet from the age of one [11] . In practice, we consider poor dietary behavior any deviation from one of these recommendations. The confidentiality of patient records was ensured by assigning an anonymity number during data collection. There is no conflict of interest. Data analysis and processing were possible using Word, Excel and EPI info version 7 software. The relative risk was calculated and presented with its limits in the 95% confidence interval and the significance threshold was set at p < 0.05
3. Results
At the end of the study, we retained 152 infants out of the 193 hospitalized for dyspnea in the pediatric department. Girls were 65 (42.76%) and boys 87 (57.24), giving a sex ratio of 1.34. Infants under 3 months accounted for 32.24% and those under 6 months 57.24%; the different age groups are described in Table 1. The median age observed was 4 months with an interquartile range of 2 to 9 months. The median time to admission was 4 days with an interquartile range of 2 to 7 days. This admission time was less than 3 days in 29.61% and less than 7 days in 63.16%. Patients came from home, urban health centers, regional hospitals and general hospitals (Table 1). Breastfeeding was exclusive in 46.71% and other modes of breastfeeding in 53.29% (Table 1). Vaccines from the expanded immunization program (EPI) were up to date in 76.32%. No pathological history was found in 75.67%. The main antecedents at risk identified were malnutrition, hypotrophy at birth, interventricular communication. The pathologies observed were low acute respiratory diseases in 90.79%, ENT in 04.60% and cardiac in 03.95%. The distribution of diagnoses has been specified in Table 2. We found, during the period from September to December 2020, 49 cases (67.65%) of bronchiolitis, 22 cases (53.66%) of acute pneumonia Community (CAP). The frequencies of dyspnea and these conditions are shown in Figure 1. The median duration of hospitalization in our department was 4 days with an interquartile range of 2 to 7 days. The infants discharged from the cured department numbered 122, or 80.26%, and those who died represented 15.13%; seven infants were transferred to the Pneumo-phtisiology department (PPH) for pleural drainage and also came out cured. The deaths concerned 9 girls or 39.13% and 14 boys or 60.87%. The median time to death recorded was 1.63 days with an interquartile range of 0.43 days (10 h) to 3 days. The median age of infants who died was 4 months with an interquartile range of 1.95 to 4.85 months. Infants under 3 months and under 6 months who died accounted for 39.13% and 82.61%
![]()
Table 1. Distribution of epidemiological parameters in children hospitalized for dyspnea.
*Others: bronchiolitis (2), premature (2), trisomic facies (2), HIV mother (2), pneumonia (1), Prune Belly (1), **others: nursery (3), rural health center (1), clinic (1), sisters’ hospital (1), surgical emergencies (1) and resuscitation (1).
![]()
Table 2. Distribution of diagnoses in infants hospitalized for dyspnea.
*: Purulent pleurisy: 05 (03.29%), Pyo-pneumothorax: 03 (01.97%), Hydro-pneumothorax: 01 (00.66%).
![]()
Figure 1. Distribution of the frequencies of dyspnea, acute bronchiolitis and Community-acquired pneumonia in infants hospitalized for dyspnea.
respectively. The distribution of these deaths according to the pathologies is illustrated in Figure 2. Acute bronchiolitis represented 52.17% of deaths with a lethality of 14.12%. The patient hospitalized for the primary tuberculosis infection died. We recorded 2 deaths or 33.33% among the cases of heart disease. CAP lethality was 17.07%. We noted certain factors related to death which were among others, age less than 6 months, history with risk, outdated EPI vaccines with a statistically significant difference. The analysis of these factors has been transcribed in Table 3.
![]()
Table 3. Risk factors for death in infants hospitalized for dyspnea.
*: Dominant, artificial, mixed.
![]()
Figure 2. Distribution of numbers and deaths of infants hospitalized for dyspnea according to diagnoses.
4. Discussion
During 2020, 193 infants were hospitalized with dyspnea, but 152 were retained. The limitations of the study were related to the difficulties relating to the assessment of the severity criteria of the various conditions in question. Indeed, the elements of gravity, such as age varied from one pathology to another. For example, bronchiolitis and pneumonia are considered severe at ages less than 6 weeks and 6 months, respectively. In addition to these elements, due to the retrospective nature of our study, the impact of environmental factors [12] [13] [14] could not be assessed. In addition, the identification of the germs responsible for acute respiratory infections could not be carried out. Nevertheless, despite these limitations, we have made observations which have given rise to some comments.
Epidemiologically, we observed a variation in cases depending on the month with peak admission frequencies, in particular in the months of February, October and December 2020 during which there were 17, 28 and 18 cases respectively. These numbers would reflect the seasonality likely associated with respiratory infections. Indeed, we noticed 2 frequency peaks concerning bronchiolitis, particularly in October and December. Seck, in Dakar during a retrospective study about acute bronchiolitis in 2017, noted two frequency peaks during the cool months of the year (February, March and April) and during the rainy season (July, August and September) [15] . In France, the period of bronchiolitis begins in October and ends at the end of winter [12] [16] . Kané, in Mali in 2017, also noted peaks in the frequency of pneumonia in the months of September (12.9%) and October (10.8%) [17] . All these results showed the influence of the seasons on the occurrence of acute respiratory infections with frequency variations depending on the geographical area.
We observed, like some authors, a slight male predominance (sex-ratio = 1.34) [18] . Infants under 6 months represented more than half of the cases and those under 3 months nearly a third of our study population. These figures could be explained by the fact that infectious viral rhinitis is manifested by nasal obstruction which can interfere with breathing and feeding in young infants [19] , and therefore justify their hospitalization. Most infants arrived at our pediatric department after 3 days; which testified either self-medication, or their passage through health centers before being referred to our service (center of last resort of the health pyramid in the center-north zone of Côte d’ Ivoire).
At the diagnostic level, acute lower respiratory conditions accounted for more than 90% of our workforce. They were dominated by infections which accounted for more than 85%. They concerned, in order of frequency, bronchiolitis, CAP and complications of CAP (0.92%). These infections are a major cause of morbidity in children with an incidence of 240‰ in infants under 1 year old [18] . The other conditions, around 8%, concerned congenital heart disease and ENT pathologies (Table 2). The latter, apart from laryngeal pathologies, were mostly referred directly to the ENT department, thus justifying their low proportion in our study. In addition, nasopharyngitis, angina and otitis were in the majority of cases treated on an outpatient basis. Congenital heart disease, although the most frequent congenital malformations, has an overall incidence of 5 to 8‰ [20] [21] .
At the evolutionary level, the median length of stay for our patients was 4 days. Indeed, patients were discharged with outpatient treatment generally 48 hours after the stability of the clinical state. Hospitalization stays varied according to the pathologies concerned and were longer in the event of the occurrence of complications or the persistence of clinical signs which would sometimes express therapeutic ineffectiveness, sometimes witnessing a diagnostic delay. The unfavorable evolution was the cause of death in 15.13% of cases. Over 75% of deaths were in infants under 5 months and the majority of these deaths occurred within 48 hours of admission. We have also seen very high lethality. The lethality of acute bronchiolitis (14.12%) remained much higher than those reported in the series of Seck in Dakar (1.6%) and Doumbia (1.8%) [2] [15] . The figures that we observed explained the difficulties encountered in infants, and could be linked to the severity of the clinical picture (34.78% of deaths occurred in the first 24 hours), to the quality of surveillance (an occupancy rate of 200%), the failure of the technical platform or the delay in treatment due to the median time for admission to our department, which was 4 days. On the other hand, we did not note any death relating to laryngeal affections. The generalization of vaccinations against Haemophilus influenzae serotype b and measles has completely changed the epidemiology and severity of laryngitis [22] . Deaths in our study were influenced by factors including age less than 6 months, history with risk (undernutrition, birth hypotrophy, and ventricular septal defect), and incomplete EPI vaccination. Indeed, infants under 6 months of age run a 3 times higher risk of die, because this age group is considered to be a serious factor in certain pathologies in question, in this case CAP [14] . Furthermore, it has been reported that most deaths caused by bronchiolitis occur in infants aged less than 6 months, who have a history of prematurity or underlying cardiological pathology [23] . Infants whose EPI vaccination was not up to date had more than 2 times the risk of dying than the others. In fact, the impact of vaccinations has been described, mainly that of conjugate vaccines against Haemophilus influenzae b (Hib) and streptococcus pneumoniae (7-valent, then 13-valent PCV), which have considerably reduced incidence and severity of childhood CAP [24] . The death rate of infants with a history of risk was twice as high as that of those without a history, due to their greater exposure to the risk of presenting serious forms. Sagbo in Benin also reported that malnutrition and vaccination status were factors associated with deaths related to respiratory distress in infants [25] .
These figures and risk factors for death found in our study highlight the problems our services have in achieving the Sustainable Development Goals (SDGs) with regard to child health, i.e. eliminating preventable deaths among these children. Reducing mortality therefore appears to be a priority. To do this, it would be necessary to improve the conditions of care for children by equipping the various health structures with essential materials and qualified human resources, by developing decision-making algorithms relating to the clinical pictures encountered and setting up patient referral procedures. When an infant is admitted to the emergency department for dyspnea, clinical analysis is essential to assess the signs of severity in order to propose appropriate measures [7] . The population should also be educated to go to health centers as soon as possible in order to avoid the occurrence of complications that can lead to death, and follow-up should be planned when patients have a history with risk. It would be necessary to insist on preventive measures, in particular hand hygiene measures, avoidance of places at risk of contamination (confined public places) [12] and above all correct vaccination for age, in particular for Hib and PCV-13 [24] .
5. Conclusion
Infant dyspnea remains a concern in our department. They are caused by pulmonary, cardiac and ENT pathologies and are responsible for many deaths in pediatric hospitalization. Reducing mortality requires the development of dyspnea management procedures, continuous staff training and the strengthening of the technical platform, and above all the adequate treatment of cases at different levels of the health pyramid and the early referral of serious cases.