RR:C19 Evidence Scale rating by reviewer:
Potentially informative. The main claims made are not strongly justified by the methods and data, but may yield some insight. The results and conclusions of the study may resemble those from the hypothetical ideal study, but there is substantial room for doubt. Decision-makers should consider this evidence only with a thorough understanding of its weaknesses, alongside other evidence and theory. Decision-makers should not consider this actionable, unless the weaknesses are clearly understood and there is other theory and evidence to further support it.
***************************************
Review: Due to the devastating global effects of COVID-19, vaccines were developed under pressure and approved and administered at speeds never seen before. On approval, only short-term adverse events were known. Speedy vaccination resulted in rapid declines in severity and mortality, even as new variants appeared. However, in some public sectors, this rapid approval led to scepticism fueled in part by the rapid spread of adverse events transmitted by mass media and social networks alike, resulting in further vaccine hesitance. In that scenario, understanding the long-term safety profile of those vaccines became critical.
This preprint presents hematologic disorders occurring in temporal association with vaccination against SARS-CoV-2 in a large Korean cohort of 1.9 million recently vaccinated individuals; as controls, the investigators used a group of 0.3 million individuals who had not been vaccinated. The authors evaluated new-onset nutritional anaemia, hemolytic anaemia, aplastic anaemia, coagulation defects, and neutropenia, but not other hematologic conditions, observing small but significant increases in the incidence of hematologic conditions among vaccinated individuals. However, the evidence is weak and the analysis is prone to several sources of bias.
I will use nutritional anaemias as an example: in this study, older individuals had higher vaccine rates than younger ones, making the comparison of vaccinated and non-vaccinated complex. Being female, older, having a higher comorbidity index, or a lower health insurance coverage, were all associated with a new diagnosis of nutritional anaemia, a condition that is more common in females (in part due to menstrual losses) and older persons. Moreover, people with lower health insurance coverage may have lower income, hence lower availability of nutrients, resulting in a higher likelihood of nutritional anaemia. Interestingly, an increased likelihood of de novo nutritional anaemias was observed as soon as one week after vaccine administration. This makes little physiological sense, as bodily stores of iron, vitamin B12 and folates last for months to years, making an abrupt loss -and the appearance of anemia- unlikely.
Thrombocytopenia, aplastic anaemia, and hemolytic anaemia were observed more commonly among vaccinated than non-vaccinated individuals. Those conditions in isolation or combined can be immunologically mediated; for instance, some (ChAdOx1 nCoV-19 and Ad26.COV2.S) but not all adenoviral vaccines are known to increase the trigger anti-platelet factor 4 antibodies, resulting in vaccine-induced immune thrombosis with thrombocytopenia (VITT), an infrequent but sometimes fatal adverse event. Moreover, these conditions were associated with older age and being female, circumstances known to increase the risk of those conditions regardless of vaccination status. Hence, the observation of those potentially immune-mediated conditions with all the evaluated vaccines -including mRNA-based ones that have not been associated in other studies with immune hematologic conditions- needs to be further addressed.
While the study raises interesting questions, its limitations make it difficult to conclude. In the meantime, with billions of vaccines already applied globally, it is reassuring that very few hematologic side effects have been observed in other cohorts.
Creative Commons Attribution 4.0 International License (CC-BY 4.0)
