…And they were all yellow

3-day-old Christopher was born at term via normal vaginal delivery. There were no problems during or after the birth and he and his mother were discharged on day 2. The domiciliary midwife had been out to do their routine check this morning and thought he was looking a bit yellow and so has sent him in for assessment. You’re wondering how accurate this ‘eyeball’ test actually is…

Why do we care about neonatal jaundice?

Hyperbilirubinaemia, or jaundice, occurs in up to 60% of term infants and 80% of preemies in the first week of life. Jaundice that occurs within the first 24 hours of life is always pathological.

Bilirubin can be ‘split’ into the conjugated or unconjugated forms

For a reminder of some of the basics of bilirubin physiology, watch this video from Armando Hasudungan

Whilst there are few causes of conjugated hyperbilirubinaemia we need to consider – think structural abnormalities such as biliary atresia or choledochal cysts, neonatal hepatitis or certain metabolic conditions – there are many causes of unconjugated hyperbilirubinaemia. The majority of cases are benign such as physiological jaundice or breast milk jaundice but there are some pathological conditions we must know.

Step 1: Is the baby yellow?

Visual inspection during a thorough neonatal examination

• Jaundice often starts in the eyes, spreads to the face and then drops down to the trunk, legs and feet
• Remember it is harder to rule out jaundice in babies with darker skin, premature infants and those less than 36 hours old.

Step 2: How long have they been yellow for?

<24 hours: jaundice within 24 hours should always be assumed to be pathological. Often these are picked up through screening programmes in the UK and so are unlikely to present to ED.

>24 hours but <14 days if term (or <21 days if prem): likely physiological

>14 days if term (or >21 days if prem): prolonged jaundice – requires further investigation

Step 3: How yellow are they?

Quantify ‘how yellow’ the baby is with a transcutaneous bilirubinometer

Use the NICE treatment threshold graph (ensure corrected for gestation) to determine if hyperbilirubinaemia is safe or requires treatment.

• Generally: if term baby, level <250 = reassure, safety-net and home
• If level > 250 = consider admission

Step 4: Are they otherwise well?

Look for signs of underlying pathology (MAGGIE)

Step 5: Are they high-risk?

When unconjugated bilirubin crosses the blood-brain barrier, acute neurotoxicity ensues and often presents with lethargy, irritability and hypotonia. In severe cases, this leads to kernicterus which is characterised by yellow-staining of the basal ganglia on post-mortem. The baby may present as hypertonic, fitting or unconscious at this time.

Aside from a high bilirubin level, other risk factors for encephalopathy include:

• Jaundice within 24 hours
• ABO/Rh incompatibility
• Previous sibling requiring phototherapy
• Cephalhaematoma or significant birth trauma
• Family history of red blood cell defect
• Sepsis or acidosis (secondary to displacement of bilirubin from albumin)
• Prematurity (lower serum albumin concentrations)

Causes

Once you have assessed the child, you can start to think about possible diagnoses.

What is physiological jaundice?

This is a normal physiological response and should resolve by two weeks in a term baby or three weeks in a premature infant. It occurs because of the higher number of red cells in the neonate, their shorter lifespan, and the slower breakdown and metabolism of bilirubin.

What is breast milk jaundice?

Breast milk jaundice is also common and may last quite a few weeks. The benefits of continuing breast-feeding far outweigh the benefits of stopping. These patients often are asked to present to the emergency department. They key points in the history is that the jaundice normally begins at day three to five and goes beyond day ten, but more importantly the child appears well. At the time of writing, the prevailing theory is that it is due to a higher concentration of beta-glucuronidase in breast milk which, in turn, increases reabsorption of bilirubin.

These cases of prolonged jaundice do require some basic investigations to rule out any serious diseases. Whilst tests vary with patient population we routinely do:

Total and ‘split’ bilirubin

Full blood count and film

Reticulocyte count

Direct Coombs test

Thyroid function tests

What are the consequences of hyperbilirubinaemia?

Unconjugated bilirubin can cross the blood-brain barrier where it may lead to neurological toxicity.

Initial signs of acute bilirubin toxicity include lethargy, irritability and apnoea. They may then become hypotonic with a poor suck before progressing to hypertonia (opisthotonus), with a high pitched cry and eventually seizures and coma.

Acute toxicity, if untreated, may lead to chronic bilirubin toxicity with the pathological post-mortem finding of yellow staining of the basal ganglia. This is known as kernicterus.

Risk factors for severe jaundice include:

Jaundice within 24 hours of birth

ABO/Rh incompatibility

Previous sibling requiring phototherapy

Cephalhaematoma or significant birth trauma

Family history of red cell enzyme defect (G6PD) or membrane defect (spherocytosis)

How can you treat neonatal jaundice?

As with most things in medicine treatment of the underlying cause should take precedence so if there is evidence of sepsis (child looks unwell) then this should be treated aggressively.  

Compare the total bilirubin level with your local neonatal jaundice treatment graphs. If the child is under the treatment line for their age (in hours) then sunlight exposure is not going to help. If they require phototherapy then this needs to happen in a timely fashion.

More serious cases might require exchange transfusion.

How does phototherapy actually work?

Blue-green light water-solublephoto-polymerizes bilirubin into a more water-soluble form that can be more readily excreted. Light in the blue-green spectrum (~460nm) via conventional (rather than LEDs) is associated with a more rapid decrease in serum bilirubin levels and, thus, a shorter duration of phototherapy. This is especially important in a time of maternal and baby bonding.

How accurate is visual assessment?

Whilst there is evidence to suggest that visual examination is enough to rule out neonatal jaundice, it is highly unreliable when it comes to estimating levels of bilirubin.

Visual inspection, ideally, should take place under bright, natural light. It first appears in the sclera (at around 35-40 micromol/l) and then face (70-100 mmols/l) before dropping, like paint poured on the head, down the trunk to the legs and feet. Kramer’s rule has been used to ‘guesstimate’ levels of jaundice but shouldn’t be relied upon to make clinical decisions.  For those with less experience an icterometer can be used to estimate levels of jaundice. Imagine it as strip of plastic with a series of colours similar to that used to match paint, that is held to the babies skin.

It can be much harder to rule out jaundice in babies with darker skin, premature infants and those less than 36 hours old.

How about transcutaneous measurement?

Skin bilirubinometers are really only useful for babies over 35 weeks of gestation and older than 24 hours of age. It is worth considering using them if it is going to take a long time to get lab results back and should the level be greater than 250 micromol/litre formal testing should take place. An added benefit is that they may reduce the number of heel pricks or venepunctures a child has to go through.

The gold standard is total serum bilirubin with venous and capillary samples being equivalent.

Case resolution

Little Christopher certainly looks jaundiced with yellow sclera and jaundiced skin down to the mid thigh. Using Kramer’s rule you estimate him to have a serum bilirubin in the range of 200-250 mmol/l and so decide to take a formal sample. After some non-nutritive suckling you perform a venipuncture and send a sample off to the lab. It comes back as 212 and so you decide to send him home with close follow-up.

References

National Institute for Health and Clinical Excellence. Neonatal jaundice. CG98. London: National Institute for Health and Clinical Excellence; 2010

Woodgate P, Jardine LA. Neonatal jaundice. BMJ: Clinical Evidence. 2011; 09:319

It’s also worth taking a look at some of the guidelines from Victoria and Queensland.