A pilot study of manual chronic partial exchange transfusion in children with sickle disease

Abstract

Objective

Red cell exchange transfusion is frequently used in the management of patients with sickle cell disease (SCD) either electively or chronically to maintain hemoglobin S (HbS) <30%. The purpose of this retrospective study was to evaluate the results of manual chronic partial exchange transfusion (MCPET) on level of Hb and HbS, on iron load and on the need for chelation, on risk of immunization, monitoring transfusion-transmitted viral infection, and clinical outcome.

Methods

We reviewed the long-term effect of MCPET in 10 children (six men and four women) with SCD and evaluated the iron balance during a median follow-up of 20 months (range: 6–36) in which 248 exchanges were performed.

Results

The pre-exchange median Hb value was 9.5 g/dl (range: 7.7–10.9 g/dl) and the median post-exchange value was 9.4 g/dl (range: 8.4–11.1 g/dl).The majority of patients reached an HbS of <50% with a median HbS value of 40.04% (range: 30–54). At start of the MCPET program, the median ferritin was 439 ng/ml (range: 80–1704 ng/ml). In the final evaluation, the median value of ferritin was 531 ng/ml (range: 84–3840 ng/ml). The annual calculated iron balance was 0. 28 ± 0.08 mg/kg/day. MCPET was well tolerated, and adverse effects were limited.

Discussion

MCPET in children with SCD is safe to prevent iron overload, and is effective and easy to use in our cohort.

Conclusion

Indication for chronic exchange blood transfusion is essential for patients with SCD with recurrent and frequent crises who do not respond to hydroxyurea. However, there is no consensual study for the period at which chronic transfusion can safely be stopped and further research in large population of patients with SCD will need to clarify this question.

Keywords:

• Sickle cell disease
• Children
• Red cell exchange
• Manual chronic partial exchange transfusion
• Transfusion risks
• Belgium

Introduction

Sickle cell disease (SCD) is a common inherited hemoglobin disease due to a single mutation that substitutes valine for glutamic acid in the beta-globin subunit. The pathogenesis of SCD is centered on the sequence of events that occur between polymerization of deoxyhemoglobin (hemoglobin S, Hb S) and causes distortion of red blood cells, anemia, vaso-occlusions, and organ dysfunction. SCD is a new health problem in Belgium within the immigration of the African and Afro-Caribbean population.^1–3 Life expectancy is reduced to less than 50 years with morbidity and mortality resulting from accelerated hemolysis of red cells anemia, infections, vaso-occlusions in the microvasculature, large vessel thrombosis, and chronic organ damage.⁴

Red cell exchange transfusion is an effective but probably underutilized therapy for chronic complications of SCD.^5,6 In a red cell exchange, patient's red cells are removed and replaced by exogenous normal red cells resulting in improved delivery of oxygen to hypoxic tissues. The exchange prevents the participation of sickle cells in new vaso-occlusive events, reduces hemolytic complications, and provides added oxygen carrying while decreasing the blood viscosity.⁷ The aim in exchange transfusion is to reduce the percentage of HbS (usually to <30%) without volemic overload and with reduced risk of high iron burden.

In our Unit, since 2000, manual partial exchange transfusion is preferred to classical chronic transfusions mainly to prevent iron overload.^8,9 The purpose of this retrospective study was to evaluate the results of manual chronic partial exchange transfusion (MCPET) on level of Hb and HbS, on iron load and on the need for chelation, on risk of immunization, monitoring transfusion-transmitted viral infection, and clinical outcome.

Materials and methods

Patients

The medical charts of children with SCD treated at Hôpital Universitaire des Enfants Reine Fabiola, Brussels, Belgium, with chronic exchange transfusions were reviewed. Data on the indication for the treatment, age at start of MCPET, sex, venous accesses, hemoglobin level before and during MCPET, HbS prior and during MCPET, volume exchanged, and ferritin were recorded, as well as serologies for HIV and hepatitis viruses. The clinical outcome in terms of (i) new hospital visit and admission for pain crisis; (ii) new clinical central nervous system event; (iii) acute chest syndrome (ACS) event; (iv) evolution of leg ulcer; and (v) the need for chelation therapy was assessed.

Transfusion protocol and chelation

Patients on MCPET were treated in our one-day clinic, and Hb level, residual percentage of HbS, and ferritin were measured prior to each exchange. Hb level was also measured at the end of the exchange. Values are given as median and range. Our targets were a post-transfusion Hb S level under 30% with a stable Hb level between 9 and 10.5 g/dl. The volume exchanged is calculated taking into account the Hb level and the last HbS percentage. Manual partial exchange transfusions were performed by removing 30–40 ml/kg of blood from the patient and administering 15–25 ml/kg of donor red blood cells and 15–25 ml/kg of normal saline with an interval of 3–5 weeks between two exchanges. All patients had a full red cell phenotype performed and the received blood matched for ABO, Rhesus, Kell, and Duffy antigen systems. Our policy is to start chelation therapy when ferritin is above 1000 ng/ml.

Venous access

Venous access was usually via peripheral veins. Only patients with poor peripheral venous access necessitated a central venous line.

Data management

Data are represented as mean ± standard deviation when the distribution was normal and median with range when the distribution was not normal. The comparison of continuous variables was calculated using Student's T test. Statistical significance level was set at P = 0.05.

Results

Characteristics of the study patient

A total of 248 MCPETs were performed for a period of 4 years. Ten children and young adults (six men and four women) between ages of 4 and 20 years were assessed on a population of 163 patients with SCD who were treated by MCPET. Their geographical area of origin was Central Africa for seven, Western Africa for one, mixed marriage between Central and Western Africa for one and multiple mixed origin for one. Regarding the SCD subtypes, all patients were Hb-SS. The median age at start of treatment was 12.5 years (range: 3.0–19.0 years).

Indication for chronic transfusion

In our group of patients, indications for partial exchange transfusion were stroke (3), increasing cerebral blood flow velocities (1), frequent ACS (3), despite hydroxyurea treatment, leg ulcer (1), previous severe hepatic cholestasis (1), and pulmonary hypertension (1). Eight patients were on hydroxyurea prior to start of MCPET.

Transfusion data

Venous access was via peripheral veins in seven patients. Three patients had poor peripheral venous access and necessitated insertion of a central venous line.

The mean Hb value before MCPET was 9.5 g/dl (range: 7.7–10.9 g/dl) and the mean Hb level on the MCPET program was 9.4 g/dl (range: 8.4–11.1 g/dl) (P > 0.05). Before start of the MCPET program, the median HbS level was 77.7% (range: 38–63). After first three PETs, the mean HbS dropped to 40.4% (range: 30–56.2%) and remained stable with time (P < 0.05). In all 10 patients, a post-exchange HbS was equal or lower than 30%, post-exchange was achieved only in one pateint, eight patients had an HbS level between 30 and 50%, and one patient remained with Hb S >50%. Individual's values are given in Table 1. For technical reasons, the volume exchanges for old children were limited.

Table 1. Characteristics of the 10 patients treated by manual chronic partial exchange transfusion

The median ferritin level was 439 ng/ml (range: 80–1704 ng/ml) with two patients having already more than 1000 ng/ml because of numerous previous transfusions. In the final evaluation, the median value of ferritin was 531 ng/ml (range: 84–3840 ng/ml). Among two patients in whom the ferritin level was >1000 ng/ml at the start, chelation could be stopped in one patient. In the group of patients without chelation therapy before MCPET, the mean level of ferritin increased from 359 to 541 ng/ml (P > 0.05), and only one patient was started with chelation therapy. The annual calculated iron balance in all 10 patients was 0.28 ± 0.08 mg/kg/day. Individual data are given in Table 1.

Tolerance

These 248 manual red blood cell exchanges performed during a period of 4 years were well tolerated with few adverse reactions. Three patients developed malaise with mild hypotension during an exchange which was thought to be secondary to rapid exchange and to changes in circulating volume. No patient developed hypertension, fever, and rigors or episodes of hyperviscosity during or after the procedure or alloimmunization. No autoimmune hemolytic anemia was documented.

Transfusion-transmitted viral infection

All children remained negative for HIV and hepatitis C virus infections. All have been previously vaccinated against hepatitis B.

Clinical outcome

All patients who have undergone exchange showed clinical improvement with no hospital visit and admissions for severe pain crisis. Clinical situation after exchange characterized by stabilization of cerebral vascular stroke and over and control of ACS complications. The healing of leg ulcers present in one patient was achieved after 1 year of MCPET. In this patient, this effect is not correlated with ferritin levels (Table 1).

Discussion

MCPET is preferable to single transfusion for many reasons: iron overload is better prevented, the hematocrit and Hb values remain constant, and volume overload is prevented. Nevertheless, MCPET is more difficult to implement because of the need of acceptable venous access. In our experience, MCPET was feasible without central venous access in 7 of 10 patients. Hemoglobin level remained within the expected range. Despite the HbS level under 30% achieved only in one patient, the clinical outcome was good with no recurrence of ACS or stroke, stable cerebral blood flow velocities and no new SCD-related event. The relief of the symptoms or the stabilization of the disease was also described in many other series even if the goal of HbS level <30% was rarely achieved.^10–12 Manual exchanges provide needed oxygen carrying capacity while reducing the overall viscosity of blood which promotes the physiology of sickling. Replacement of sickle cells by normal cells can help to prevent further vaso-occlusive process, although pre-existing vaso-occlusion may not be reversed and may lead to maintenance of low HbS levels.

Our data showed that iron balance was good (ferritin < 1000 ng/ml) in the majority of patients with minor iron overload assessed by serum ferritin value when compared with their data before the start of MCPET. Savage et al.¹³ reported similar observation in their series. Two patients who needed iron chelation had already high ferritin before the start of MCPET. This situation is due to chronic transfusion before the start of MCPET, and this was also observed by Kwiatkowski et al.¹⁴ in their recent report. As it was applied, MCPET was safe to prevent middle-term iron overload and the need of chelation therapy in most of patients. Even if chelation therapy is proven to be successful to prevent liver and cardiac impairment due to iron overload, desferrioxamine, deferasirox, or deferiprone can be associated with side effects or poor compliance that prevents its efficacy. Finally, treatment failure is common because of non-compliance, difficulty of administration, and numerous other side effects.^14,15 Poor compliance seems to be a major problem in SCD.¹⁶ This study suggests that this procedure controls the risk of iron overload and reduces the need of chelation therapy when compared with the conventional transfusion program.^17–19

Although serum ferritin is the most frequently used test to estimate iron overload, it has some particular limitations in SCD. If measured in steady state, several studies show a significant correlation between total transfused units by simple top-up and serum ferritin. At high body iron loads, increasing proportions of serum ferritin are derived from leakage of iron-rich unglycosylated ferritin from hepatocytes and are associated with increasing liver enzyme leakage. Owing to these limitations in monitoring iron overload in SCD, the use of serum ferritin alone is particularly problematic and magnetic resonance imaging monitoring is recommended.^20,21

In the red cell exchange, exogenous normal red cells replace the patient's sickle cell red. This procedure rapidly decreases the participation of sickle cell in new vaso-occlusive events, the rate of hemolysis complications, and lead to a reduction in net blood requirements compared with simple transfusion. In addition, normal transfusion in patients with SCD increases the risk of high blood viscosity.⁷ In a previous study, Porter and Huehns¹⁹ showed that patients who had received exchange transfusions were less likely to be iron overload (ferritin increment per unit of blood = 9.9 ± 3.8 µg/l) than patients who had received an equivalent number of units by conventional transfusion (ferritin increment per unit of blood transfused = 25.1 ± 2.42 µg/l).

During the period of study, automated exchange transfusion is not available and exchange transfusion was being performed manually. This method is not so well controlled as automated exchange may result in significant changes in circulating volume. In our study, this procedure was safe, well tolerated, and effective, but was time consuming. We did not observe any problem related to hyperviscosity after an exchange. Only three patients experienced symptoms related to hypovolemia. Owing to of this side effect, the way in which MCPET is conducted has been modified with emphasized hourly volume balance. All patients experienced a clear clinical improvement and none decided to stop the MCPET. The only MCPET stopped was in one patient after the complete healing of his chronic leg ulcer but there were no effect correlates with ferritin level (medical decision). The same observation was described by other researchers as well.^22–25 All patients were aware of the transfusion-related risks. In the Belgian sickle cell population, transfusion carries the risk of alloimmunization to red cell because the blood donors are of Caucasian origin and the patients from African origin.²⁶ In our study, no patient developed red cell antibodies after 6 months to 4 years of this procedure. However, the retrospective nature of our study and the short duration of MCPET did not permit to determine the risk and the mean period of apparition of alloimmunization. No contamination by HIV, hepatitis B, or hepatitis C was observed.

Indication for chronic exchange blood transfusion is essential for sickle cell patients with recurrent and frequent crises and who do not respond to hydroxyurea or those at risk for stroke. However, there is no consensual study for the period at which chronic transfusion can safely be stopped and further research in large population of patients with SCD will need to clarify this question.

Acknowledgments

We thank Drs Benedetta Chiodini, Chiraz Ghaddab, Pauline Lecerf, and Nathalie Tram who involved in the collection of samples; all the nurses of Salle 60 of the Department of Hemato-oncology at HUDERF for the support given during the study.

Disclaimer statements

Contributors MNA, PQL, CH, MNM, and AF conceived and designed the study protocol. MNA, PQL, SH, NA, MNM, CD, and AF carried out the clinical assessment. MNA, MNM, and AF are involved in the analysis and interpretation of these data. AF and MNA drafted the manuscript. AF, LPQ, SH, NA, and CD critically revised the manuscript for intellectual content. All authors read and approved the final manuscript. MNA and AF are guarantors of the paper.

Funding None.

Conflicts of interest None.

Ethics approval The medical charts of sickle cell patients were anonymous, reviewed retrospectively. Ethical approval is not required for this study.