Neonatal jaundice and just so stories

Newborn child baby having a treatment for jaundice under ultraviolet light in incubator.

Writing on The Conversation, three UK scientists make an elementary error.

Their piece is entitled Jaundice in newborns could be an evolutionary safeguard against death from sepsis.

In newborn babies, jaundice is so common as to be termed physiological. It affects around 60% of term babies and around 80% of preterm babies in the first week of their lives. Clinicians need to monitor it carefully and sometimes treat it, since it can lead to conditions like acute bilirubin encephalopathy and kernicterus that can damage the infant’s brain and cause developmental problems.

But it now looks as though this jaundice is not merely one of the pitfalls of entering the world. New research just published in Scientific Reports, in which we have been involved, suggests that it is one of the gifts of evolution. Humans may develop jaundice as newborns to protect from something even more serious: sepsis.

The elementary mistake is invoking the the naturalistic fallacy, a logical fallacy that presumes that anything that exists in nature must be good.

[pullquote align=”right” cite=”” link=”” color=”” class=”” size=””]How would we determine if jaundice prevents sepsis? We would look at the relatively risk of sepsis in jaundiced babies vs. non-jaundiced babies.[/pullquote]

…[W]hy have humans not evolved to overcome this temporary bilirubin problem?

Why? Because evolution through natural selection does not produce perfection. A variety of often conflicting evolutionary pressures may result in a relatively high wastage rate.

For example, in nature many babies and mothers die because the baby is too big to pass through the maternal pelvis. The evolutionary pressure on maternal pelvis size is entirely independent of the evolutionary pressure on fetal head size. If natural selection produced perfection, human beings would have evolved some way for the mother’s body to communicate with the fetus’ body to constrain its size. That has not happened because natural selection is limited in what it can accomplish; it can only produce limited results with existing genetic material.

Moreover, evolution through natural selection leads to the survival of the fittest, NOT the survival of all. The fact that a given natural process has a relatively high death rate is entirely in keeping with that principle.

There is no evidence I’m aware of that shows that neonatal jaundice is beneficial in any way to a baby, but that hasn’t stopped these investigators from fantasizing otherwise.

One night he was looking after a baby boy who had sepsis, which is where the immune system goes into overdrive to protect against infection, potentially leading to severe inflammation, organ failure and death. This baby was profoundly unwell in intensive care, suffering from inflammation and a strikingly high bilirubin count that was only just being controlled with three phototherapy lamps. Usually this kind of difficult jaundice is caused by an immune reaction between mum’s and baby’s blood groups, but not in this case.

Richard began wondering if the bilirubin was directly linked to the infection, and if it was part of the baby’s body’s attempt to clear the sepsis (in this case the baby survived). He started thinking about the problem in evolutionary terms – if jaundice can harm the baby, what benefit does it offer to balance this?

The odds are high that the bilirubin is linked to the infection but not in the way that the investigators imagine. Sepsis can injure the liver, decreasing its ability to metabolize bilirubin. Sepsis induced jaundice occurs at all ages.

According to Clinical review: The liver in sepsis:

During sepsis, the liver plays a key role. It is implicated in the host response, participating in the clearance of the infectious agents/products. Sepsis also induces liver damage through hemodynamic alterations or through direct or indirect assault on the hepatocytes or through both. Accordingly, liver dysfunction induced by sepsis is recognized as one of the components that contribute to the severity of the disease.

In other words, there is no reason to believe that neonatal jaundice is protective against sepsis and no data that shows that neonatal jaundice prevents sepsis. No matter.

The results of this project have just been published. Our team have shown that even modest concentrations of bilirubin reduced by one third the growth of Gram-positive Streptococcus agalactiae. We also showed that bilirubin may alter substrate metabolism in the bacteria.

In short, it looks like the hypothesis is bearing out. We now need to do more work, probably in animal experiments of sepsis. This will enable us to think about whether clinicians should raise the accepted bilirubin threshold for babies at risk of sepsis – those born prematurely, for example.

No, it does NOT look like the hypothesis is bearing out.

That bilirubin kills some common bacteria is not unexpected. The whole problem with jaundice in the newborn is that bilirubin is cytotoxic; if it could kill newborn brain cells, it’s hardly surprising that it can kill bacteria, too. That doesn’t mean that excess bilirubin occurs to prevent bacterial sepsis.

Sadly, this is a “just so story.”

What’s a just so story? The term comes from a Rudyard Kipling book of the same name, filled with stories like “How the leopard got its spots.” It is:

an unverifiable narrative explanation for … a biological trait … The pejorative nature of the expression is an implicit criticism that reminds the hearer of the essentially fictional and unprovable nature of such an explanation.

This story could be titled “how the infant got its jaundice.”

Kipling’s just so stories were fairytales and most contemporary efforts to use just so stories to explain evolutionary phenomena are also fairytales.

As Steven J. Gould wrote:

…unfortunately a very large part of evolutionary theory and practice, natural selection has operated like the fundamentalist’s God–he who maketh all things… When evolutionists try to explain form and behavior, they also tell just-so stories–and the agent is natural selection.

But natural selection is not the only engine of evolution.

…[W]e now reject this rigid version of natural selection and grant a major role to other evolutionary agents (genetic drift, fixation of neutral mutations, for example). We must also recognise that many features arise indirectly as developmental consequences of other features direct subject to natural selection. Moreover, and perhaps most importantly, there are a multiple of potential selective explanations for each feature. There is no such thing in nature as a self-evident and unambiguous story.

How would we determine if jaundice prevents sepsis? We would look at the relatively risk of sepsis in jaundiced babies vs. non-jaundiced babies. Unless and until we can show that jaundice is protective, we have no business asserting that it is protective. We also have no business extrapolating from test tubes to human beings. No doubt bleach also kills the bacteria that cause neonatal sepsis, but that’s not a reason to start giving sick babies bleach.

The authors conclude:

It feels like we’re discovering something new about the physiology of newborn babies. It’s the excitement of being a clinician scientist: taking an idea from a real patient into the laboratory and testing then developing it to hopefully help future patients. When newborn babies develop jaundice in future, we’ll still need to treat it carefully. But quite possibly we will also be thankful that it’s protecting them from something potentially life-threatening.

No one has discovered anything about the physiology of newborn babies because no one looked at the physiology of newborn babies. They made up a just so story.

Just so stories are remarkably attractive; that’s why scientists must be very careful not to invoke them. They should be even more cautious about advancing therapeutic recommendations based on what at the moment is little more than wishful thinking.