Prevalence and diagnostic dynamics of iron deficiency anaemia in adults with sickle cell anaemia attending Muhimbili National Hospital

Abstract

Background: The prevalence of iron deficiency anaemia (IDA) and sickle cell disease (SCD) is high in Tanzania. Evidence from other countries, suggests that the prevalence of IDA in SCD is low. There is limited information available on the prevalence of IDA in SCD in Tanzania. This is compounded by the challenges in the diagnosis of IDA in SCA.

Aim: To determine the prevalence, diagnostic models and factors associated with IDA in adults with SCD at Muhimbili National Hospital (MNH).

Methods: This was a descriptive cross-sectional hospital based study involving steady-state adult individuals with SCD who attended the SCD Clinic at MNH, between May and December 2014. A structured questionnaire was used to collect socio-demographic and clinical information. Venous blood samples were taken for blood count and biochemical iron studies. Bone marrow iron status was determined amongst a subset selected using low mean cell volume (MCV) and mean cell haemoglobin (MCH). Six diagnostic models for IDA in SCD were proposed and compared with bone marrow iron status. Chi-square test was used for categorical variables, and student t-test for independent sample means. Logistic regression was used to analyse factors associated with IDA in SCA.

Results: A total of 170 individuals with SCA aged 18 to 51 years (65.3% females) were enrolled. Prevalence of IDA ranged between 16.5% - 48% using the seven diagnostic models. The sensitivity and specificity for diagnosis of IDA using transferrin saturation (TSat) was 47.6% and 81% respectively, whilst for serum ferritin were 71.4% for both parameters. The highest specificity, 95.2% was observed using TSat+Ferritin and TSat+ferritin+MCV models. However overall performance in the receiver operating curves (ROC) curves was greater for serum ferritin than any other tests (AUROC =0.765, p=0.003, 95% CI 0.615-0.915). Low MCV was found to be the predictor of IDA in SCA. Age, sex or history of blood transfusion was not found to be associated with IDA.

Conclusion: The prevalence of IDA in adults with SCD is high. Different models showed a variation in detection and accuracy of diagnosis of IDA. Serum ferritin below 100µg/L was found to be the best test for diagnosis of IDA. Further research is needed to determine the feasibility and cost of determining iron status in SCD.