Case Commentary

Dominic Barton is a little unlucky – he is born early, has GBS sepsis, his ductus fails to close spontaneously and has biliary atresia. 

Treated optimally, this is all he has and he does well. All routes through the case lead to an alive child at the end, but with varying amounts of morbidity. However, this case illustrates how, by making other decisions, Dominic can develop subglottic stenosis, a DVT, MRSA and have a worse outcome from his biliary atresia. Having two or more major deviations from ‘ideal’ increases anxiety in the parents, although sometimes even doing the right thing is not what the parents will want.

The subglottic stenosis will lead to croup when he encounters a virus later in his first year of life. The DVT leads to treatment with heparin – a discomfort in itself, but a potential cause of bleeding. The MRSA will become a problem as he requires surgery for his liver condition. And early intervention for bliary atresia crucial in improving outcome.

Beyond this, inaccurate assessment of the situation can lead to needless parental anxiety and, in this case, failure of breast feeding.

The case illustrates not just neonatal decision making, but also illustrates several common forms of medical error.

Decision step 1: Fixation, Playing the odds, and Communication

The first decision point allows a choice between antibiotics, surfactant and steroids. This is in the context of a slightly premature infant with developing respiratory distress. The differential diagnosis is wide – respiratory distress syndrome, pneumothorax, pneumonia, septicaemia, coarctation of the aorta and metabolic disorders are only a few of the more likely causes. 

One of the errors likely at this stage is that players will see that the infant is premature and so RDS is most likely – playing the odds. This has its merits as long as an important alternate diagnosis is not missed. In this case neonatal infection is quite likely and very serious if not treated. If antibiotics are delayed in the case, the infant has a more severe septic episode and also requires inotropes. This in turn requires a central line, leading to a DVT and prolonged heparin course, bringing further discomfort.  

Players might overvalue the relevance of his prematurity, showing fixation on this diagnosis. The chest radiograph would be consistent with RDS, although it would be consistent with congenital pneumonia too. He apparently improves from a respiratory point of view after ventilation. If fixated on RDS they might then ignore his poor condition at birth, temperature instability and lowish blood pressure.  

Communication is also key to this step. Although not crucial for making the correct decision at the first step, the handover neglects to mention that the mother has a fever in labour. This, preterm labour, prolonged rupture of membrane and a poor condition of the infant are the factors suggesting neonatal septicaemia or infection may be important. Unfortunately tests are not helpful to confirm or eliminate infection in this age group. The white blood count is typically high at birth without infection; a low white count is more suggestive of infection. Blood cultures have a low specificity especially early in a bacteraemia. This explains why the cultures are negative if antibiotics are given at the first step.

Decision step 2: Ignorance, Timidity, Triage, Insufficient skill, Poor teamworking, Fixation 

The second decision point offers an echocardiogram (the following day), antibiotics and diuretics as options. To identify the correct option, a player would need to identify that the examination and radiograph is typical of a patent ductus arteriosus, and that there are no features suggesting sepsis. This should lead to diuretics being used.

Although an echocardiogram would further confirm that the infant is in heart failure (as well as demonstrating the cause) this is not needed before starting the diuretics.

Antibiotics are not unreasonable in this scenario, but a cause beyond sepsis is much more likely – players fixated on sepsis as a cause may choose this option. 

Ignorance and insufficient skill are similar errors, one being more technical and one more relating to learnt information. Being unable to read a chest radiograph, examine a patient, undertake a procedure successfully are all examples of lack of skill and can be taught. Not knowing about typical clinical features of pulmonary oedema, the significance of an enlarged liver and that a ductus arteriosus might be the cause of these problems is ignorance. Again this can be learnt. 

It is of course not expected that a health care professional will know everything. They should be aware of common issues affecting them in their area of practise….and be aware of when to ask someone else, and this is one aspect of teamworking. In this case the consultant would have been well aware of the condition and how to read the radiograph, so could have been contacted. This is not an explicit option**.

Understanding the level of severity of a situation is an important skill for healthcare workers. This is an important facet of triage and informs how quickly an action needs to be taken and which actions take priority over another. In this case, the infant is deteriorating and something needs to be done if possible immediately rather than waiting for the echocardiogram the next day. Antibiotics can also be given immediately, but only work if an infection is responsible. In this case, giving antibiotics is not unreasonable, but does not address the pulmonary oedema and heart failure. Part of the role of a professional is taking control of a situation, and not doing so, or waiting without an end point or purpose is an example of timidity.  

Decision Step 3: Sloth, Ignorance, Triage, System error 

At this step, players must choose between discharging the patient, which the parents are keen for the player to do, and keeping the infant in the hospital until the cause is better understood. Players must also choose how many tests they want to do. Although there is the temptation to do a large number of tests in virtual and real scenarios, this is always unwelcomed by the patient and does not necessarily lead to better diagnosis. However not doing what is needed suggests sloth.

One error that will be made here is ignorance. To make a good choice at step 3, players must understand that jaundice developing at 3 or more weeks is abnormal. They should also note the features of obstructive jaundice (pale stools and urine darker than the normal baby colour which is almost colourless). This should distinguish the infants condition from the much more common and benign physiological jaundice and breast milk jaundice.

Good triage is also key to this step. Health care professionals typically want to be kind to their patients, even if this is not in their interests. Late onset jaundice is an important problem that needs to be addressed even if it extends an inpatient stay.

Lastly, complicated referral mechanisms that make accessing the right person at the right time are system errors. Systems should be arranged to facilitate good practise and make it difficult to cause harm.