“Treatment of Ebola Haemorrhagic Fever with Blood Transfusions from Convalescent Patients”

The aims of this study were to determine whether a blood transfusion from a convalescent patient can be used to treat patients with Ebola effectively. Another objective was to determine whether the death rate is reduced when this measure is taken. The study was carried out during the 1995 Ebola outbreak in Kikwit, Democratic Republic of Congo. Eight female patients, with Ebola, between the ages of 12 and 54 were treated with blood from five male convalescent patients.

The study was a small scale clinical trial, conducted to assess a new area of treatment during an outbreak. The small sample size may reflect the lack of blood available. There were no controls – the intervention was simply given to 8 patients and their outcomes assessed. The intervention was simply a blood transfusion from a convalescent patient – the blood contained no Ebola antigen, but contained IgG and IgM Ebola antibodies. This was determined using ELISA testing, which is an accurate test. There was no randomisation or blinding in the trial, but given the nature of it, this wasn’t really appropriate anyway.

The outcome measure was to see whether the patients treated with the transfusions survived. As this is a very simple measure, no statistical tests were used to analyse results. As there were no controls, the fatality rate for the treated patients (found to be 12.5% – 7 out of 8 patients survived) was compared with the rate for the whole outbreak. This was around 80%, so overall the study found that patients who received the intervention had significantly better outcomes.

There are several potential sources of bias in the study. The patients who received transfusions were all ill towards the end of the outbreak, and by this stage, the general care for Ebola patients had improved, excluding the blood transfusions. Healthcare workers had better knowledge and practiced methods, and had access to antibiotics and antimalarial drugs. This means that patients who were ill towards the outbreak would have been more likely to recover anyway, and the study population’s survival may not be down to the transfusions. It is also not clear in the paper how the eight patients were selected. This may mean that the organisers could have chosen patients who were more likely to get better anyway, patients with milder symptoms for example, leading to a misrepresentation of the intervention’s effectiveness.

This study addresses the aims and objectives well, and uses appropriate methods to work towards a conclusion. It has a small sample size however, which means we can’t say for certain whether the method will work on a large scale. Additionally, in the discussion, the paper refers to other patients out with the study who received transfusions with much less success, many of them dying. This indicates that the method may not be as effective as the study results indicate.

We also can’t say which aspect of the transfusion made the difference: antibodies against Ebola is the most likely reason, but it could also be that transfused blood contained more clotting factors to help stem bleeding, or that the blood transfusion was effective for prevention of shock

The initial signs from this study are good, showing that treating patients with blood transfusions is successful. However, further research will be needed to test this further and ascertain that it is an effective measure. Limitations with study mean that we can’t take this as cast iron evidence that it is an effective treatment, and while opening up areas for further research, it has not definitely concluded anything.