Geospatial Mapping of Sickle Cell Disease in Northwest Tanzania: The Tanzania Sickle Surveillance Study (TS3)

Abstract

Tanzania ranks third in Africa for the estimated number of annual births with sickle cell disease, but these estimates are based on sparse data from small studies reported over the past 50 years. A recently completed surveillance study from Uganda documented substantial variation in the prevalence of sickle cell trait and disease across the country. Tanzania lacks a national newborn screening program, and no contemporary multi-regional screening of infants has been undertaken. We designed and conducted a prospective study to determine the prevalence of sickle cell trait and disease by region and district in northwest Tanzania, where the prevalence of sickle cell is thought to be highest. The study used existing public health infrastructure while building local capacity for accurate diagnosis of sickle cell disease. Secondary objectives included characterization of hemoglobin variants and exploration of associations between sickle cell trait, sickle cell disease, malaria, and HIV. The Tanzania Sickle Surveillance Study (TS3) is a prospective cross-sectional study of HIV-exposed infants born in 9 regions across the Lake Zone of northwest Tanzania. In Tanzania, the HIV early infant diagnosis (EID) program collects dried blood spots (DBS) from all children born to HIV-infected mothers. DBS are transported to a central laboratory for prompt detection of HIV vertical transmission. In northwest Tanzania, the DBS are transported to Bugando Medical Centre, a teaching and consultancy hospital in Mwanza, where they are tested for HIV and then stored on-site, and thus available for further testing. Isoelectric focusing (IEF) equipment was donated to Bugando Medical Centre along with reagents and supplies. Two laboratory staff were trained by a board certified hematologist, and then attended a two day seminar by the IEF manufacturer. One pediatrician completed a 2-month observership at Cincinnati Children's Hospital. All DBS samples were tested by IEF using appropriate controls. Completed gels were scored independently by two Tanzanian staff members as normal, disease, trait, variant, or uninterpretable. DBS samples scored as disease or variant were repeated for confirmation and preserved for later genotyping. Regular Skype calls were convened with US-based collaborators to improve quality and interpretation. HIV test results were obtained from the local EID program. Between February 2017 and May 2018, 232 IEF gels were completed by the local staff. After children >24 months of age were excluded to obtain a more accurate newborn prevalence, the median age of children tested was 52 days (IQR 41-93 days), and a total of 17,278 unique DBS samples were scored. The quality of laboratory testing was extremely high with only 20 samples scored as uninterpretable and 54 with missing results, and the primary analysis was performed on the 17,204 remaining samples. The overall prevalence of sickle cell trait and disease in the entire cohort was 20.3% and 1.2%, respectively, with a 0.1% prevalence of hemoglobin variants. This corresponds to an allelic frequency of 0.114 for the sickle gene mutation and demonstrates perfect Hardy-Weinberg equilibrium. No HbC or other common beta-globin variants were identified. Geospatial mapping revealed some variation across regions, with sickle trait ranging from 16.6% to 22.5% and disease ranging from 0.5% to 1.5%. Analysis of individual districts with >100 samples revealed wider geographic variability, with sickle trait ranging from 15.2% to 27.8% and disease ranging from 0.0% to 4.3%. Co-morbidity between HIV and sickle cell disease was analyzed to compare it with the effect on mortality previously observed in Uganda. The prevalence of sickle cell disease was the same among HIV-infected and HIV-negative children (1.2%), suggesting no difference in mortality. The prevalence of sickle cell trait and disease among infants born in northwest Tanzania is very high, exceeding 20% trait and 1.2% disease. All regions in the Lake Zone are affected possibly due to lack of immigration to the area and similar environmental exposures. Targeted newborn screening can be started in high prevalence districts, using existing public health infrastructure with minimal start-up cost and training. Future work will evaluate the correlation between historical malaria prevalence and sickle cell prevalence, and identify hemoglobin variants. Disclosures Ware: Addmedica: Research Funding; Novartis: Membership on an entity's Board of Directors or advisory committees; Bristol Myers Squibb: Research Funding; Agios: Other: advisory board; Global Blood Therapeutics: Other: advisory board; Biomedomics: Research Funding; Nova Laboratories: Consultancy.

Communicating sickle cell trait results after newborn screening: Approaches and implications to families

“…We have brought together strong accounts and recommendations from the interviews and discussions done with families on what are the key issues to consider when communicating SCT results. Families understanding of the results is of priority especially in places where the burden of the SCD is high (Ambrose, Makani, et al, 2018; Fanning, 2016; Ndeezi et al, 2016). Results from this study has shown that families understanding of the results provided is still challenging.…”

Perspectives on Building Sustainable Newborn Screening Programs for Sickle Cell Disease: Experience from Tanzania

“…Bugando Medical Centre (BMC) located in the Mwanza region serves as the zonal referral hospital for the lake and western zone of Tanzania, where the annual births of children with SCD are estimated to be one of the highest in the world [ 24 ]. A pilot NBS for SCD program implemented at BMC between August and September 2014 reported 1.4% of children with SCD out of 919 children screened and 19.7% children with traits [ 25 ]. Inclusion of BMC was motivated by both the prevalence of SCD in the region served by the hospital as well as their experience in implementing NBS for SCD [ 24 , 25 ].…”

“…A pilot NBS for SCD program implemented at BMC between August and September 2014 reported 1.4% of children with SCD out of 919 children screened and 19.7% children with traits [ 25 ]. Inclusion of BMC was motivated by both the prevalence of SCD in the region served by the hospital as well as their experience in implementing NBS for SCD [ 24 , 25 ]. Both sites have at least one day each week dedicated clinic for sickle cell patients.…”

Perspectives on building sustainable newborn screening programs for sickle cell disease: Experiences from Tanzania

“…Bugando Medical Centre (BMC) located in Mwanza region is serving as the zonal referral hospital for the lake and western zone of Tanzania where, the annual births of children with SCD is estimated to be one of the highest in the world [26]. A pilot NBS for SCD program implemented at BMC between August to September 2014, reported 1.4% of children with SCD out of 919 children screened and 19.7% children with trait [27]. Inclusion of BMC was motivated by both prevalence of SCD in the region served by the hospital as well as their experiences in implementing NBS for SCD [26, 27].…”

“…A pilot NBS for SCD program implemented at BMC between August to September 2014, reported 1.4% of children with SCD out of 919 children screened and 19.7% children with trait [27]. Inclusion of BMC was motivated by both prevalence of SCD in the region served by the hospital as well as their experiences in implementing NBS for SCD [26, 27]. Figure 1 (A) and (B) below are maps of Dar es Salaam and lake zone regions showing geographical locations of health facilities included in this study.…”