Dear Editor,

The paper by Sarkar et al[1] addresses a question of considerable practical and ethical importance for parents and clinicians caring for asphyxiated newborn infants. Which infants are so severely affected that they will not benefit from attempted neuroprotection with therapeutic hypothermia?[2]

However, it is not clear that their paper answers this question. They focus on risk of mortality, but do not provide any data on the mechanism of demise in dying infants. What proportion of infants died despite full attempts to save their lives, and what proportion died following withdrawal or withholding of life-sustaining treatment? This is critically important since in previous studies of infants with HIE the majority of deaths followed treatment limitation decisions.[3]

The problem is that if a significant proportion of deaths did follow treatment limitation decisions, the statistical association between an Apgar score of 0 and death may reflect a self-fulfilling prophecy.[4] Given the recommendation by the Neonatal Resuscitation Program and previous evidence of poor outcome in infants with such Apgar scores, it would not be surprising if an Apgar score of 0 at 10 minutes made clinicians more pessimistic about outcome and more likely to subsequently limit treatment.

Secondly, the authors suggest that infants surviving with Apgar scores of 0 at 10 minutes are "likely to have severe disability". They base this claim on the finding that there were no non-disabled survivors in the cohort. However, the absence of an event does not provide good evidence of its rarity in the presence of small sample sizes.[5]

The authors note in passing that their results differ from those of the NICHD sub-study.[6] But they understate the significance of those results. In that study there were 13 survivors with an Apgar score of 0 at 10 minutes, of whom 6 (46%) had mild or no disability.[6] Although that report did not distinguish between cooled and non-cooled infants, if the two reports are combined they suggest a 38% chance of favourable outcome among surviving infants, with a confidence interval of 18-61%.

Further follow-up data of infants with HIE treated with cooling may help clarify which infants do not benefit from therapeutic hypothermia. But at present it would be premature to deny infants treatment based solely on their Apgar score at 10 minutes of age.

Dr Dominic Wilkinson, Neonatologist, Nuffield Medical Research Fellow, The Ethox Centre, University of Oxford

REFERENCES

1. Sarkar S, Bhagat I, Dechert RE, Barks JD. Predicting death despite therapeutic hypothermia in infants with hypoxic-ischaemic encephalopathy. Arch Dis Child Fetal Neonatal Ed. 2010; [published Online First doi:10.1136/adc.2010.182725].

2. Wilkinson DJ. Cool heads: ethical issues associated with therapeutic hypothermia for newborns. Acta Paediatrica. 2008;98:217-20 doi: 10.1111/j.1651-2227.2008.01127.x [published Online First 28/11/2008].

3. Pierrat V, Haouari N, Liska A, et al. Prevalence, causes, and outcome at 2 years of age of newborn encephalopathy: population based study. Arch Dis Child Fetal Neonatal Ed. 2005;90:F257-61 doi: 10.1136/adc.2003.047985.

4. Wilkinson D. The self-fulfilling prophecy in intensive care. Theor Med Bioeth. 2009;30:401-10 doi: 10.1007/s11017-009-9120-6.

5. Eypasch E, Lefering R, Kum CK, Troidl H. Probability of adverse events that have not yet occurred: a statistical reminder. BMJ. 1995;311:619-20.

6. Laptook AR, Shankaran S, Ambalavanan N, et al. Outcome of term infants using apgar scores at 10 minutes following hypoxic-ischemic encephalopathy. Pediatrics. 2009;124:1619-26.

Conflict of Interest:

None declared