Koshy 1988.

Methods	Randomised controlled trial.
Participants	72 pregnant women diagnosed with sickle cell anaemia presenting in early pregnancy without other medical disorders. Exclusion criteria: women previously on long‐term prophylactic transfusion beginning before the study; pregnant women with other haemoglobinopathies such as HbSC and HbSβ‐Thalassaemia. Setting: 6 hospitals in Chicago (secondary and university hospitals) and Johns Hopkins Hospital, Baltimore.
Interventions	Intervention: red cell transfusion at the beginning of the management of their pregnancy with the goal of maintaining Hb concentration at 10 g/dL and 11 g/dL, or the HCT near 0.33, and to reduce the HbS below 35% by simple transfusion or partial exchange transfusion. Immediately upon entry into the study, patients received 2 units of packed washed frozen red cells weekly for 3 weeks or until the above goals were reached. All blood transfused was obtained from volunteer donors and processed according to standard blood banking procedures (n = 36). Control: blood transfusion only for medical or obstetric indications. Haematologic indication for blood transfusion were a Hb concentration below 6 g/dL, a HCT below 18% and a reticulocyte count below 3%. All blood transfused was obtained from volunteer donors and processed according to standard blood banking procedures (n = 36).
Outcomes	Maternal death, severe maternal morbidity (acute chest syndrome, pulmonary embolism, congestive heart failure), perinatal death, pain crisis, total units of blood transfused, blood transfusion reaction.
Notes	The study was conducted over a period of 7 years and 2 months.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Method of random sequence generation not stated. Study was described as "controlled randomized prospective study…". Patients were random assigned to 1 of 2 treatments.
Allocation concealment (selection bias)	Unclear risk	No information about allocation concealment to permit judgement.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Although it is impracticable to blind participants and key study personnel to intervention, the knowledge of the intervention by the study personnel makes performance bias a high possibility.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Uncertain whether outcome assessors were blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	No missing data reported.
Selective reporting (reporting bias)	Unclear risk	Comparison of outcome measure in the 'Methods' and 'Results; sections of the report indicated no evidence of selective outcome reporting although there was a reference to occurrence of maternal mortality in the 'Discussion' section.
Other bias	Unclear risk	The significant difference in previous perinatal mortality (as one of the baseline characteristics) between intervention and control groups questions the effectiveness of the randomisation procedures.

Hb: haemoglobin
 HCT: haematocrit