Epidemiological, Diagnostic, Therapeutic, and Prognostic Profile of Non-Tuberculous Community-Acquired Purulent Pleurisy in Children at the Bouaké University Hospital Center, 2017-2021 ()
1. Introduction
Purulent pleurisies, also known as empyemas, are characterized by the accumulation of a thick, creamy liquid effusion, or a turbid to clear fluid laden with variably altered polymorphonuclear leukocytes typical of pus, between the two layers of the pleura [1] [2] . These conditions are a major complication of community-acquired acute pneumonia in children. Byington et al. report that 28% of childhood pneumonia leads to purulent pleurisy [3] . Although the current global incidence is not well-documented, purulent pleurisies have been significant public health challenges in developed countries over the past two decades [4] - [10] . The World Health Organization (WHO) has identified it as the second leading cause of hospitalization and pediatric deaths after malaria in developing countries [11] . Specifically in Côte d’Ivoire, and particularly in Bouaké, there is a scarcity of data regarding purulent pleurisies in children [12] [13] . Although diagnosing purulent pleurisy can be straightforward based on the macroscopic and/or cytological characteristics of the pleural puncture fluid, identifying the causative pathogens is challenging in resource-limited settings like Côte d’Ivoire. This difficulty is due to pre-analysis antibiotic treatments, inadequate technical resources, and the prohibitive cost of paraclinical evaluations for many patients. The aim of this study was to describe the epidemiological, diagnostic, therapeutic, and evolutionary aspects of purulent pleurisies in a pediatric context, contributing to data essential for developing strategies to reduce child morbidity and mortality at the Bouaké University Hospital Center (CHU).
2. Patients and Methods
This study was a descriptive retrospective analysis utilizing the records of children admitted to the CHU of Bouaké from January 2017 to December 2021. Situated in central Côte d’Ivoire, 347 km from Abidjan, the CHU of Bouaké stands as the sole tertiary health care facility serving the central, north, and west regions of the country. The study encompassed children from 1 month to 15 years old diagnosed with purulent pleurisy. The diagnosis was confirmed through radiological imaging and the biological examination of pleural puncture fluid, primarily of an exudative type (protein > 30 g/l), marked by the presence of polymorphonuclear neutrophils, either intact or altered [1] [2] . Exclusion criteria included being younger than one month, lacking a pulmonary radiograph, or having a bacterial superinfection of a tuberculous pleurisy, pleurisy not confirmed radiologically and biologically, as well as cases of pleurisy not hospitalized. The study employed comprehensive sampling of all eligible children’s medical records from the specified period. We retrieved filing numbers from the admission/discharge records of the pediatric, pneumophthisiology, and internal medicine departments for children aged 01 month to 15 years with lower respiratory tract infections. Using these numbers, we then compiled the relevant patient records that met our inclusion criteria. Data were gathered using a validated, pre-established survey form, capturing epidemiological (frequency/prevalence, age, sex, residence), diagnostic (clinical, radiological, bacteriological), specific treatment (antibiotic therapy, local treatment), and outcome (recovery, death, other) information. Pleurisy was categorized by its extent: 1) Low abundance, indicated by radiological evidence of fluid in the costo-diaphragmatic recess(es); 2) Medium abundance, occupying the lower two-thirds of the lung field; 3) High abundance, extending over more than two-thirds of the lung field [14] [15] . Ethical approval was granted by the scientific medical commission of the CHU of Bouaké. Confidentiality was maintained through anonymous numbering. Record review and data collection were performed at the storage sites. Data processing and analysis were conducted using Epi Info software version 7. Qualitative variables were represented as proportions, while quantitative variables were analyzed using mean and standard deviation.
3. Results
3.1. Epidemiological Profile
In the period from January 2017 to December 2021, the study at CHU of Bouaké’s pneumophthisiology and pediatrics departments identified 124 cases of purulent pleurisy in children, out of 687 children with lower respiratory tract infections and a total of 15,505 pediatric hospitalizations for various conditions. Purulent pleurisies thus constituted 18% of lower respiratory tract infections and accounted for 0.8% of all pediatric hospitalizations. The annual distribution was as follows: 6 cases (4.8%) in 2017, 8 cases (6.5%) in 2018, 70 cases (56.5%) in 2019, 19 cases (15.3%) in 2020, and 21 cases (16.9%) in 2021. Male children formed the majority of cases, accounting for 67%, with a sex ratio of 2.1. The age range of these patients was from 1 month to 168 months, with an average age of 33 months and a median age of 11 months. Children aged ≤ 60 months represented 81% of cases, with a significant proportion (69%) being between 1 to 24 months old. Regarding the geographic origin, 56% of these children were from outside the city of Bouaké, including 52% from different health regions.
3.2. Diagnostic Profile
· Clinical Aspects (Table 1)
The vaccination history indicated that 39% (n = 48) of the children were not
![]()
Table 1. Clinical and paraclinical manifestations.
up-to-date with their routine immunizations. Among the patients, 37% (n = 46) had received the 13-valent pneumococcal and Haemophilus influenzae type b vaccines as part of the expanded immunization program. The medical history in the past three months included skin rash in six (5%), exposure to passive smoking in twelve (10%), and pneumonia in three (Table 1) patients. Prior to hospitalization, self-medication practices involved antibiotics in 39 (31%), antipyretics in 19 (15%), corticosteroids in 15 (12.1%), traditional remedies in 14 (11.3%), and antimalarials in 12 (9.7%) patients. observed were malnutrition in five (4%) and HIV infection in four (3%) patients. The symptom duration before the first medical consultation ranged from 2 to 60 days, with 44% presenting within less than 7 days, 33% between 8 and 14 days, and 23% after more than 14 days. The primary symptoms at admission were dyspnea (85.5%) and fever (43.5%). The predominant physical findings in the pleuropulmonary examination included pleural effusion syndrome (86.3%), O2 saturation below 95% in ambient air (76.6%), and respiratory distress (68.5%). Other physical examination findings were cutaneous-mucosal pallor (63.7%), deterioration of the general condition (60.5%), and dermatoses (20.2%).
· Radiological Findings (Table 1)
Pulmonary radiographs at the time of admission primarily showed right-sided pleurisy (62.1%) with a significant accumulation of fluid (78.2%). Additional radiographic findings included Hydropneumothorax (37.1%), alveolo-interstitial opacities (8.1% of cases), and abscess formations (1.6%).
· Bacteriological Results (Figure 1, Table 2)
Out of the 124 patients, 64 (52%) had documented results of pleural fluid
![]()
Figure 1. Prevalence of bacteria isolated from pleural fluids.
![]()
Table 2. Antibiotic sensitivity and resistance profile of isolated Staphylococcus aureus.
Note: SXT stands for Trimethoprim-sulfamethoxazole. Oxacillin was not tested; cefoxitin was used as a substitute.
culture. Among these, 30 (46.9%) exhibited positive pleural fluid cultures, identifying a total of 32 bacteria. Staphylococcus aureus was the predominant bacterium, accounting for 75% of these cases (Figure 1). Antibiotic sensitivity testing on 24 S. aureus strains showed complete sensitivity to Cefoxitin (100%) (Table 2).
3.3. Treatment and Outcome Profile (Table 3, Table 4)
Empirical antibiotic treatment was administered to all 124 patients. Oxacillin (53.1%) and amoxicillin-clavulanic acid (53.1%) were the commonly used beta-lactams, often in combination with other antibiotics such as gentamicin (64.1%) and/or metronidazole (21.9%), resulting in dual therapy (42%) or triple therapy (33%). Single-drug antibiotic therapy was applied in 25% of the cases (Table 3). Pleural drainage was executed in 68.5% of patients, while iterative evacuation punctures were employed in 33.1% of children due to the unavailability of drainage kits. Among the 106 patients with anemia, 41 (39%) received blood transfusions. The outcomes for the 124 patients included a favorable progression in 73.4%, a mortality rate of 18.5%, and surgical transfer for 4 (3.2%) patients. The hospital stay was ≤7 days for 88 (71%) patients, averaging 6 days, with a range from 1 to 13 days
4. Discussion
4.1. Epidemiological Perspectives
Between January 2017 and December 2021, the CHU of Bouaké’s pediatric department recorded 124 cases of purulent pleurisy out of how many patients observed during the study. These accounted for 0.8% of hospital admissions and 18% of lower respiratory tract infections among children aged one month to years. The recruitment methodology for purulent pleurisy cases in this study may underrepresent the actual incidence in the CHU of Bouaké. The analyzed
![]()
Table 3. Antibiotics prescribed for purulent pleurisies.
![]()
Table 4. Local treatments and patient outcomes.
Note: A child could undergo one or several of these local care interventions.
data solely encompassed biologically confirmed purulent pleurisy in hospitalized patients, potentially excluding less severe cases managed on an outpatient basis. Additionally, before 2019, the CHU of Bouaké’s biology laboratory faced challenges with frequent shortages of biological materials, only improving post-2019 following an Ivorian-German partnership. The years 2020 and 2021 also saw reduced healthcare facility attendance due to the SRAS-Cov 2 pandemic [16] , possibly influencing the lower pleurisy cases recorded in 2017, 2018, 2020, and 2021 compared to 2019. Regular monitoring of hospital incidence rates is vital to accurately track current annual trends of children’s purulent pleurisy. Similar hospitalization rates have been observed in other African studies [17] [18] . The average age of the patients was 33 months, with a median of 11 months. Children aged ≤ 60 months comprised 81% of the cases, with 69% between 1 and 24 months. This vulnerability of young infants, noted by authors like Ndiaye, Thiam, and Andriatahihirintsoa [17] [18] [19] , is likely due to their developing immune systems [19] [20] . A notable male predominance was observed, with a sex ratio (M/F) of 2.1%, consistent with other studies reporting ratios from 1.2 to 2.7 [3] [8] [10] [17] [18] [19] [21] [22] [23] . This male susceptibility, though not fully explained, is recognized in pediatric healthcare [24] [25] . Regarding residence, most children (52%) hailed from other health regions of Côte d’Ivoire, indicating that any policy aimed at addressing infantile purulent pleurisies should encompass these broader regions.
While our study was not designed as an analytical one, we did observe several risk factors for purulent pleurisy in children, as highlighted in previous literature [3] [6] [10] [18] [26] - [30] . Notable among these were inadequate pneumococcal vaccination coverage (37%), self-administered antibiotic treatment (31%), and malnutrition (4%). Our findings did not include the use of non-steroidal anti-inflammatory drugs (NSAIDs). However, it’s noteworthy that corticosteroid therapy was administered to 12.5% of patients prior to their hospitalization.
4.2. Diagnostic Profile
· Clinical and Radiological Characteristics
While fever is typically a prevalent symptom in studies of purulent pleurisy [17] [37] , it was present in only 43.5% of cases in our study. This contrasts with the more severe symptoms noted upon admission, such as respiratory distress (68.5%) and extensive pleurisy (78.2%). Over half (55.7%) of the patients in our study experienced a delay in seeking medical attention, a common occurrence reported in African research [17] [21] [32] . This delay and the gradual worsening of pleuropulmonary symptoms could be partially attributed to self-medication practices involving antibiotics, antipyretics, and corticosteroids.
Radiologically, most studies indicate a predilection for purulent pleurisies to occur on the right side, as was the case in 62.1% of our study’s cases [17] [19] [21] [23] . This tendency may be due to the anatomical orientation of the right main bronchus, which is obliquely inclined nearly vertically, making the right lung more susceptible to infections, pollutants, and consequently to fluid effusions. In contrast, the left bronchus has a more horizontal oblique angle [33] [34] . The presence of pneumothorax, seen in 37.1% of our study’s patients, is often associated with staphylococcal infections [19] [35] .
· Microbiological Considerations
Bacteriological diagnosis of purulent pleurisies is crucial for guiding antibiotic treatment and for epidemiological surveillance of bacterial ecology. In our research, 52% of patients had documented pleural fluid culture results. The rate of bacterial identification from these cultures was 46.9%. Lower positivity rates, between 25% and 42%, have been observed in studies conducted in Burkina Faso, India, Madagascar, and Senegal [17] [18] [21] [31] . The lower rate could be attributed to prior antibiotic treatments and the challenges faced by microbiology laboratories in developing countries, such as the lack of appropriate culture media and the delay in culturing the biological samples, which can lead to bacterial death. An alternative approach, as demonstrated in Garba’s study [36] , is the use of latex tests for detecting soluble antigens. In their research, while all pleural fluid cultures were negative, the latex test detected Streptococcus pneumoniae in 46% and Staphylococcus aureus in 4% of cases. Akoua also reported the interest of latex tests applied to biological fluids [37] . In our study, Staphylococcus aureus was the most frequently isolated bacterium, accounting for 75% of cases, differing from other studies where pneumococcus is typically predominant [17] [36] . The prevalence of Staphylococcus aureus might be influenced by the warm, humid climate and inadequate skin hygiene among children and/or mothers, which facilitates the spread of skin and mucosal infections caused by Staphylococcus aureus, eventually leading to pleuropulmonary infections [38] . Other factors potentially influencing the low incidence of pneumococcus in our findings include routine empirical antibiotic therapy for pulmonary conditions, delayed sample processing, inadequate laboratory equipment, and the protective effects of pneumococcal and Haemophilus influenzae type b vaccines [17] [27] [39] [40] .
Additionally, our study did not identify any cases of Haemophilus influenzae.
4.3. Treatment Profile
Antibiotic strategies for purulent pleurisies should be adapted to local bacterial ecologies and antibiotic sensitivity profiles [41] . In our cohort, the predominant bacterium, Staphylococcus aureus (75%), was methicillin-sensitive, unlike the methicillin-resistant strains reported by Thiam et al. [17] . All our patients received empirical antibiotic therapy, mainly oxacillin (53.2%) or amoxicillin-clavulanic acid (53.2%), combined with other antibiotics like gentamicin (64.1%) and/or metronidazole (21.8%), administered as either dual (42%) or triple therapy (33%). The use of aminoglycosides has been questioned by some due to their limited efficacy in acidic conditions and poor penetration in experimental empyema [42] . The selection of oxacillin or amoxicillin-clavulanic acid in our study aligns with the wild-type nature of Staphylococcus aureus identified. Pleural drainage, the recommended urgent intervention for moderate to large pleurisies [43] , was performed in 68.5% of patients. However, the immediate unavailability of drainage kits led to delays or the inability to drain the pleural fluid in some cases, a challenge also observed by Garba et al. in Niamey [36] . Consequently, 33% of children underwent repeated evacuation punctures. Early physiotherapy, targeting the reduction of functional sequelae [44] , was implemented in 80% of the patients. Fibrinolysis and thoracoscopy, however, were not conducted in our study.
4.4. Outcome Profile
The course of the disease was favorable in 73.4% of the cases. Nonetheless, a significant mortality rate of 18.5% was observed, higher than the rates reported by Thiam in Dakar, Alao in Cotonou, and Garba in Niamey [17] [32] [36] . Literature identifies several factors that contribute to increased mortality, such as delayed initiation of pleural evacuation [45] , prolonged time before hospitalization exceeding 7 days, young age, malnutrition, and severe anemia [21] , all of which were observed in our study. Furthermore, mortality rates are heavily influenced by underlying comorbidities, with potential mortality rates reaching up to 40% in cases of immunosuppression [46] . In our study, the only notable comorbidities were malnutrition (4%) and HIV infection (3%). Mortality also appears to be linked to the microbial cause of the infection. According to the bacteriological data from the MIST-1 study [47] , infections caused by Gram-negative bacilli, Staphylococcus aureus, or polymicrobial aerobic germs have higher mortality rates. Conversely, infections caused by streptococcus and anaerobes tend to have a better prognosis. In our cohort, Staphylococcus aureus was the predominant pathogen (73%). Therefore, addressing these prognostic factors is crucial for improving outcomes in children with purulent pleurisy.
5. Conclusion
This study highlights that non-tuberculous purulent pleurisies continue to be a significant concern in pediatric hospitalizations at the CHU of Bouaké. Infants are predominantly affected, with notable incidences of home self-medication and delays in seeking healthcare. Methicillin-sensitive Staphylococcus aureus emerges as the primary etiological agent, despite challenges in pathogen identification. The observed mortality rate was alarmingly high. Implementing microbiological monitoring and establishing a comprehensive management policy for infantile purulent pleurisies are imperative steps towards reducing their morbidity and mortality.