3.1.1. Frequency

During the study period, 108 children were transfused out of 512 admissions in the pediatric department of Tengandogo University Hospital, representing a hospital transfusion frequency of 21.09%.

3.1.2. Sociodemographic Data

The mean age of transfused children was 32.87 ± 39.04 months, ranging from 1 day to 180 months. Children under five years accounted for 84.3% of cases. Males were slightly predominant (54.6%), with a sex ratio of 1.2. Regarding residence, 44.3% lived in rural areas, 36.8% in urban areas, and 18.9% in peri-urban areas. Among school-aged children, 62% were enrolled, including 22.2% at post-primary level.

The main clinical signs observed were pallor (84.5%), tachycardia (68.9%), dyspnea (50.5%), altered consciousness (20.4%), and hemorrhage (6.8%). Malaria was the most frequent underlying pathology, accounting for 62% of cases. Among the 19 deceased children, severe malaria with profound anemia was the primary diagnosis in 15 cases (78.9%). Other contributing comorbidities in this group included bacterial co-infection (3 cases), severe acute malnutrition (2 cases), and neonatal sepsis (2 cases).

The mean hemoglobin level before transfusion was 5.1 ± 2.8 g/dL, and 8.1 ± 2.9 g/dL after transfusion. The mean erythrocyte transfusion yield (ETY) was 1.01 ± 0.5. Before transfusion, severe microcytic normochromic anemia represented 47% of cases, while after transfusion, moderate microcytic normochromic anemia was found in 56.1% of cases. The distribution of children according to anemia type before transfusion is shown in Figure 1, and biological parameters before and after transfusion are presented in Table 1.

Figure 1. Distribution of children according to the type of anemia before transfusion.

Table 1.Biological parameters before and after transfusion.

Packed red blood cells (PRBCs) were the most frequently used blood product (93.5%). Most transfusions (71%) were performed within 24 hours of the request. The mean volume of PRBC transfused was 153.5 mL, with 63.4% of bags ranging between 100 and 300 mL. In neonatology, 44.4% of transfused volumes ranged between 25 and 60 mL. The mean transfusion rate was 1.9 mL/min, with extremes of 0.13 and 8 mL/min. The duration of transfusion was between 30 minutes and 1 hour in 40.6% of cases. The majority of children (88.4%) received a single transfusion. Figure 2 illustrates the delay between the request for blood products and the transfusion.

Figure 2. Delay between the request for labile blood products and transfusion.

Blood group B Rh⁺ was the most frequent among transfused children (32.4%), followed by group O Rh⁺ (29.6%). In neonates under 28 days (n = 18), group O Rh⁺ predominated (33.3%). Among mothers of children under four months (n = 23), blood group O Rh⁺ was also the most represented (34.8%), followed by groups O Rh⁻ and B Rh⁺ (21.7% each), A Rh⁺ (13%), then AB Rh⁺ and B Rh⁻ (4.3% each).

Per-transfusion incidents were observed in 2% of cases. Adverse effects occurred in 5.6% of children, mainly fever/chills (83.3%) and dark urine (16.7%). Hemovigilance forms were correctly completed in 64.1% of cases, but only 41.7% were returned to the blood bank. Blood request forms were compliant in 88.5% of cases.

After transfusion, mucosal coloration was normal in 60.8% of children, while 39.2% still presented pallor. Among children with persistent signs of decompensation, tachycardia was found in 70% of cases and dyspnea in 30%. Clinical outcomes were favorable in 59 children (54.6%), 29 (26.8%) were still hospitalized, 1 (0.9%) had discharged against medical advice, and 19 (17.5%) died.

The main limitations of this study lie in its short duration, monocentric nature, and the absence of multivariate analysis to identify predictive factors of poor outcomes. Furthermore, the absence of a hemovigilance committee in the department limited post-transfusion traceability. Additionally, the study was conducted from July to October, corresponding to the peak malaria transmission season in Burkina Faso, which may have increased the number of transfusions and the predominance of malariarelated anemia. This seasonal concentration limits the generalizability of the findings to other periods of the year. Despite these limitations, this study provided an overview of transfusion practices in our working context.

The high frequency of pediatric transfusions (21.09%) observed in this study is consistent with trends widely reported in sub-Saharan Africa, where the high prevalence of severe malaria-related anemia, delayed consultations, limited transfusion coverage in peripheral health facilities, and concentration of resources in urban centers make blood transfusion both frequent and indispensable in hospitals [1][4]-[7].

The predominance of children under five years (84.3%) among transfused patients may be explained by the high incidence of severe malaria in this age group, as confirmed by data from the Ministry of Health of Burkina Faso [5]. Male sex accounted for more than half of recipients (54.6%), a trend also observed in Porto-Novo (55.7%) [4] and Antananarivo (51.2%) [8]. This predominance may reflect increased vulnerability of boys to severe anemia, particularly in malaria contexts, as well as sociocultural factors favoring healthcare access for male children [4]. The majority of transfused children resided in rural areas (44.3%), reflecting difficulties in blood product supply in peripheral districts, leading to referrals to urban centers [5].

The most frequent clinical signs were pallor, tachycardia, and dyspnea, indicating anemia decompensation, often associated with malaria. These manifestations reflect tissue hypoxia secondary to reduced oxygen transport. Severe signs such as altered consciousness (20.4%) and hemorrhage (6.8%) were also observed. These results are comparable to those reported by Ba et al. in Bamako, where pallor and dyspnea were among the main clinical signs in transfused children [7]. Hemorrhage may be linked to emerging pathologies such as dengue, increasingly frequent in Burkina Faso [9], due to its ability to induce thrombocytopenia and increased vascular permeability.

The mean increase in hemoglobin after transfusion was satisfactory, rising from 5.1 g/dL to 8.1 g/dL. The relatively low ETY (1.01 ± 0.5) may reflect prolonged administration delays, hemolysis associated with infections, or factors affecting the quality of blood products, such as storage duration, coldchain interruptions, or variations in donor hematocrit. This hemoglobin gain meets clinical objectives, particularly exceeding the critical threshold of 7 g/dL. These values are close to those reported in Kinshasa (5.8 g/dL and 8.6 g/dL) [10], suggesting homogeneity of transfusion practices in African hospitals. Microcytic normochromic anemia was the most frequent form before transfusion (47%) and persisted in 56.1% of cases after transfusion. This profile suggests iron deficiency, justifying iron supplementation alongside transfusion in our context. It may also reflect inflammation related to infections.

PRBCs were the most prescribed blood product (93.5%), consistent with national recommendations and practices in most sub-Saharan countries for managing malaria-related anemia [3][7]. The transfused volume was generally adapted to children’s weight, with 63.4% of bags ranging between 100 and 300 mL. The mean transfusion rate was 1.9 mL/min, slightly below national guidelines recommending initiation between 2 and 3 mL/min [3].

The mention “vital emergency” appeared on 71.3% of blood request forms, yet 27.9% of transfusions were performed beyond 24 hours. This delay highlights persistent tensions in blood product supply, particularly during peak demand such as malaria season. This situation is corroborated by the ACTRAN report on transfusion chains in West Africa [6], which emphasizes structural and organizational constraints affecting transfusion responsiveness.

Group B Rh⁺ was the most frequent among transfused children (32.4%), followed by group O Rh⁺ (29.6%). In neonates under 28 days, group O Rh⁺ predominated (33.3%), as did in mothers of children under four months (34.8%). These distributions are similar to those reported by Ba et al. in Bamako, where group O Rh⁺ was also predominant among transfused children [7]. The predominance of groups B and O in West Africa reflects regional genetic distribution, where group O is historically the most frequent in sub-Saharan populations. The concordance between neonates and their mothers can be explained by hereditary transmission of ABO and Rh antigens, well documented in immunohematology literature [11].

Transfusion safety data show that reported incidents were infrequent but likely underreported, given the insufficient completion (64.1%) and return (41.7%) of hemovigilance forms. Immediate adverse effects occurred in 5.6% of children, mainly non-hemolytic febrile reactions (fever and chills). These results are consistent with those reported by Rakotoarisoa et al. in Madagascar, where 85.3% of adverse effects were febrile [9]. The absence of extended phenotyping and systematic screening for irregular antibodies may contribute to these reactions, favoring minor incompatibilities undetected during pre-transfusion checks. Establishing a CSTH in the department would be essential to improve traceability and systematic analysis of adverse events.

Post-transfusion outcomes were favorable in 54.6% of children, while 26.8% were still hospitalized at the end of the study. One child (0.9%) had absconded, and 19 (17.5%) died. The high mortality rate is concerning, markedly higher than that reported in Kinshasa, where Mbuyi et al. observed a mortality rate of 1.6% among transfused children [10]. This may be explained by the severity of initial clinical presentations (severe malaria, profound anemia), delays in transfusion supply, and the presence of severe comorbidities such as severe sepsis, severe acute malnutrition, and bacterial co-infection.