Purpose: Sickle cell disease (SCD) is the most common hereditary hematologic disorder worldwide with greater than 90% of SCD births occurring in low resource settings including sub-Saharan Africa (SSA). The gold standard for diagnosing SCD is hemoglobin electrophoresis but it is often unaffordable or unavailable in low resource settings that have the highest burden of SCD. Instead, cheaper tests such as a peripheral blood smear (PBS) have been used to diagnose SCD. However, few studies have been conducted to assess the validity of a PBS as a diagnostic tool for SCD. The primary goal was to measure the sensitivity and specificity of a PBS for the diagnosis of SCD and the secondary goal was to evaluate the PBS characteristics of patients with SCD. Methods: A two-gate case–control study was conducted between January and March 2019, during which patients with SCD (Hb SS/Sβ0 thalassemia) and controls with anemia were enrolled with a ratio of 3 SCD patients to 1 non-SCD patient with anemia. A laboratory technician and hematologist, both of whom were blinded from each other and from whether the PBS was from an SCD or non-SCD participant reviewed the PBS. Using a standardized form, they assigned likelihood of SCD diagnosis as definite (presence of numerous sickled cells), probable (presence of other characteristics without multiple sickled cells) or negative based on review of each slide. The laboratory technician further reviewed each PBS for sickled cells, target cells, Howell-Jolly bodies, normoblasts and polychromasia. Results: A total of 154 participants were recruited and data analysis was run on 149 participants (102 SCD and 47 non-SCD) aged 1 to 18 years. The specificity was 100% (95% CI 94 - 100) for both reviewers and sensitivity was 88.2% and 70.5% for the laboratory technician and hematologist respectively, when ‘definite’ was utilized as a positive test. The specificity decreased to 87.2% and 97.6% and sensitivity increased to 98% and 95.8% for the laboratory technician and hematologist respectively, when ‘probable’ was included (Table 1). The interrater agreement for the laboratory technician and hematologist for positive SCD (definite and probable) versus negative SCD was high (96% agreement, kappa 0.897, standard error 0.085). Sickled cells were seen exclusively in PBS of SCD participants. Howell Jolly bodies and target cells were seen more frequently in PBS of SCD participants while polychromasia and normoblasts were commonly seen in PBS of SCD and non-SCD participants (Table 2). Conclusion: In a resource limited setting where the burden of SCD is high, the use of PBS to diagnose SCD is highly specific, arguing against the need for confirmatory testing. Furthermore, examination of PBS by experienced individuals looking for features of SCD can be a sensitive test for pediatric patients presenting with anemia.

Table 1

Review of Peripheral Smears

Table 1

Review of Peripheral Smears

Close modal
Table 2

Peripheral Smear Findings

Table 2

Peripheral Smear Findings

Close modal