Discussion

Unfortunately, the available information about these three deaths is limited, which is regrettable as the death of a child in a drug study is a major event. We cannot be sure about the exact cause of these deaths: in the candesartan study the authors indicate that the treatment was a possible cause, while the treatment was not regarded of importance in the valsartan study.

One can argue, from the limited information provided, that at least two of these deaths are likely to have been partly caused by the ARB treatment. These two children died during an episode of gastroenteritis and one of the children was also nephrotic. They were therefore in clinical situations where dehydration and intravascular fluid depletion typically occur. The combination of dehydration, nephrotic syndrome and treatment with an ARB probably contributed to acute kidney injury, fluid overload, possibly hyperkalaemia and death.

The child in the candesartan study had, before the fatal episode, already been shown to be very sensitive both to dehydration and to ARB treatment. This had previously resulted in a major increase in serum creatinine. The autopsy findings also support death from renal failure rather than from dehydration and showed pulmonary and brain oedema.

Glomerular capillary pressure largely reflects efferent arteriolar vasoconstriction maintained by angiotensin II.8 Blocking the renin–angiotensin axis with an ARB or an ACE inhibitor (ACEi) therefore dilates the efferent arteriole, and reduces glomerular capillary pressure and the glomerular filtration rate. This is more pronounced in neonates and small children with their lower systemic blood pressure. Patients with intravascular fluid depletion are also more susceptible to these effects, as are patients with reduced glomerular capillary pressure from renal artery stenosis.

It is therefore likely that preschool children are more at risk for these side effects than older children as they experience episodes of gastroenteritis more often than older children. These infections are also typically more severe with a higher mortality in the younger age group. These children also have a lower glomerular perfusion pressure which makes their kidney function more vulnerable to a reduction in circulating blood volume.

The reduced glomerular capillary pressure is the mechanism behind the well-known side effect where drugs blocking the renin–angiotensin system can increase serum creatinine with a concurrent rise in serum potassium. No data on serum creatinine or glomerular filtration rate (GFR) are given in the valsartan paper. However, one child did have to stop the study early because of “renal function impairment”. In the candesartan study, the change in eGFR (calculated GFR) was −5.4±20 ml/min/1.73 m2. Recalculating that figure assuming that 66% of all children were within ±1 SD, indicates that 17% of all children in the study experienced a decline in eGFR greater than 25 ml/min/1.73 m2, which is a substantial reduction in GFR. This could cause further deterioration in a dehydrated child.

One major problem that has been highlighted in previous publications is the lack of availability of full information about patients in drug studies sponsored by the pharmaceutical industry.9 It is also of note that the death of the child in the candesartan study was initially, in abstracts and oral presentations, reported as not related to the treatment but was described in the final paper as “acute-on chronic renal failure provoked by incidental diarrhoea-induced hypovolaemia during chronic ARB treatment”.7 10

A recent paper on children undergoing cardiac surgery showed that treatment with an ACEi conferred a high risk (21%) of acute kidney injury postoperatively, in particular if the children were also treated with furosemide.11

These new studies in young children might stimulate increased use of ARBs. They do, however, show that treatment with ARBs in young children has certain risks and that there is a need for careful monitoring of the children when using these drugs. This is particularly important in children with markedly impaired kidney function or ongoing nephrotic syndrome. ARB treatment should also be discontinued in situations where there is a substantial risk of intravascular fluid depletion.

A more thorough analysis of all data from the above mentioned cases would increase our knowledge of how to safely use ARBs in children below 5 years of age. The European Medicines Agency has expressed concern about the side effects but have no further information. The Food and Drug Administration have approved candesartan for the use in children from the age of 1 year.