Pneumonia is a major cause of morbidity and mortality in children under five years of age [1] . By 2015, 5.9 million children under five years of age have died from pneumonia, including 70% in sub-Saharan Africa and Southeast Asia [1] [2] [3] [4] . In Africa, pneumonia accounts for 25% - 33% of hospitalizations [5] [6] [7] . HIV/AIDS infection has increased its morbidity due to the well-established relationship between pneumonia and HIV infection [8] .

Pneumonia can be prevented by vaccination and its mortality reduced with the implementation of the WHO strategy [1] for the management of Acute Respiratory Infections (ARI) in children under five years of age. This is based on early clinical screening and rational use of antibiotic therapy in primary and peripheral hospitals.

In the Congo, the prevalence of pneumonia is not known, nor there is a system for the epidemiological surveillance of respiratory diseases in children. However, respiratory emergencies are the second leading cause of hospitalization in children under 5 years of age, according to the study by Moyen et al. [9] .

This first Congolese study aimed to determine the frequency of pneumonia and to identify the causative germs and factors associated with death in children under 5 years of age in Brazzaville.

This was a prospective cohort study conducted between January and June 2016 at the University Hospital of Brazzaville (UHB) in the Republic of Congo. The population of Brazzaville was estimated at 1,373,382 inhabitants in 2015, the climate is tropical with an average annual temperature of 25.5˚C.

Children aged 1 to 59 months admitted to the pediatric wards during the study period, meeting the diagnostic criteria for pneumonia, were included. Parental consent was required and obtained to be eligible. The diagnostic criteria of the pneumonia selected were those defined by the WHO [10] : pneumonia is said to have no sign of severity when breathing is rapid (≥60 cycles per minute in children less than 2 months, ≥50 cycles per minute between 2 and 11 months and ≥40 cycles per minute between 1 and 5 years) associated with crackling rats on auscultation. Pneumonia is said to be severe when, at the signs of pneumonia without signs of gravity, a draw is associated. Pneumonia is said to be very severe when severe pneumonia is associated with central cyanosis, inability to drink or nurse, incoercible vomiting, convulsions, lethargy or loss of consciousness, severe respiratory distress.

Children with chronic cough were not included in the haunting of tuberculosis.

The variables analyzed were:

Ø Epidemiological:

- Those related to parents: age, educational level, socio-economic level, number of adults sleeping in the same room as children under 5 years of age, use of oil lamp for lighting houses, fumigant tortillons used for mosquito control, parental smoking;

- Child related: age in months, sex, consultation time, antibiotics received prior to hospitalization, vaccination status for pneumococcal antigens and Haemophilus influenzae b, feeding for the first six months, the existence of a particular terrain such as HIV/AIDS, sickle cell disease, heart disease, diabetes.

Ø Clinic:

The state of consciousness, temperature, color, respiratory rate, moisture state, nutritional status, signs of pulmonary auscultation, percutaneous oxygen saturation and the existence of hepatomegaly and/or splenomegaly.

Ø Paraclinic:

The radiograph of the thorax, interpreted by two paediatricians, with confirmation in each case by a radiologist, a hemogram, C-reactive protein, retroviral serology (HIV) and blood culture whose technique was as follows: After sampling of blood on culture medium, an incubation in the oven at 37˚C. In the BacT-Alert 3D automaton was carried out for five days. Beyond that, the bacteria detected were contaminants. The inoculum was seeded in the following culture media: cooked blood agar supplemented with poly vitex for the growth of Streptococcus pneumoniae and Haemophilus Influenzae, fresh (human) blood culture agar for streptococcus culture, Chapman medium for culture of staphylococci. Microscopic analysis was performed using Gram staining, identification and antibiogram were performed by the Compact 2 raptor. Quality control was performed using the reference strains recommended by CA-SFM/EUCAST for: Streptococcus pneumoniae ATCC 49616, Staphylococcus aureus ATCC 29213 and Klebsiella pneumoniae ATCC 700603.

Ø Therapeutic and Evolutionary

Treatment during hospitalization, length of hospital stays, observed complications, and type of progression (favorable outcome, progression to death or complications).

- The socio-economic level was assessed on the basis of the food poverty line defined by the National Congolese Center for Statistics and Economic Studies (CNSEE) [11] : it is said to be low when the family spends less than 523 FCFA (African Financial Community Franc/CFA. 1 Dollar = 550 CFA Francs =) per person per day, average between 523 and 1685 CFA francs and raised between 1685 and 5055 CFA francs.

- Hypoxemia was retained when the oxygen saturation was less than 90%.

- Nutritional status was assessed on the basis of WHO growth standards [12] .

- The typology of the radiological images was made by referring to the WHO standardized criteria [13] .

- For HIV serology, presumptive diagnosis in infants under 18 months of age was retained when the infant met the following criteria: signs suggestive of HIV infection [14] , the positivity of its serology and that of his mother [14] .

- The cough was called chronic for a duration greater than or equal to 1 month.

- Vaccination was up-to-date when the child received 3 doses of both vaccine serotypes (PCV13 and HI b).

- Polypnea was defined as fast breathing (≥60 cycles per minute in children less than 2 months of age, ≥50 cycles per minute between 2 and 11 months and ≥40 cycles per minute between 1 and 5 years of age.

- Hypothermia was defined as a temperature below 35˚C.

- Promiscuity has been defined by the existence of more than three persons other than the child under 5 years of age.

The study protocol obtained the approval of the Committee of Ethics of Health Sciences Research: N 041/MRSIT/IRSSA/CERSSA.

Data was entered using CS Pro and SPSS^® 16. Percentages, averages and standard deviations were calculated. In order to assess the influence of age on certain variables and to identify the determinants of pneumonia-related mortality, an analytical study was carried out with relative risk (RR) and confidence interval (CI) to 95%. In order to identify the factors associated with mortality, some variables were adjusted on others to eliminate the confounding factors, thus allowing the calculation of the adjusted RR * by the Logit method. The statistical influence observed between two variables was assessed by the Pearson chi-square test, the significance level of the comparisons was <0.05.

1) Sociodemographic characteristics of parents (Table 1)

2) Epidemiological Aspects

- Frequency, age and sex

During the study period, 1934 children under 5 years were hospitalized, 237 of whom were pneumonia (12.3%). Pneumonia was the third leading cause of hospitalization after malaria and bronchiolitis. There were 133 boys (56%) and 104 girls (44%), with an average of 15.9 ± 13.8 months extremes (1 and 59 months).

- Type of breastfeeding, vaccination profile and field

Exclusive breastfeeding during the first six months of life was performed in 112 children (47.3%), pneumococcal vaccination (PCV13) and anti-Haemophilus influenzae b in 191 (80.6%) and 303 (85, 6%) children. There were 15 (6.3%) children born prematurely, 13 (5.5%) sickle-cell anemia and 11 (4.6%) with congenital heart disease.

- Living conditions of children: living conditions are recorded in Table 2.