Send letter to journal:
Re: Survival in cerebral palsy

Dear Editor

Re: Effects of cognitive, motor, and sensory disabilities on survival in cerebral palsy. ADC 2002;86:84-90

We appreciate Rosenbloom’s commentary on our article and would like to respond to some of the criticisms that he raises.

It is important to accept that the Mersey cerebral palsy register was established over two decades ago to be used as an epidemiological tool to examine trends in prevalence in cerebral palsy. At that time neonatal intensive care was in its infancy yet there was no routine data source that would allow monitoring of trends in the prevalence of cerebral palsy. The massive developments that have taken place in neonatal intensive care with its concomitant improvement in survival of ever smaller infants alone has justified the establishment of the register. It is easy, with the aid of a retrospectoscope, to make the criticism that data should have been collected as to whether the visual disability associated with cerebral palsy was due to retinopathy of prematurity or to cerebral cortical damage. The database was not established to examine the effects of sensory, cognitive and physical disability on survival. This has been a by -product, albeit a very useful one for paediatricians and lawyers. We only suggested that severe visual disability may be a marker of the severity of cerebral impairment. To what extent is the ophthalmic pathology of the visual impairment relevant in examining the probability of survival? The results we presented in Table 4 show that severe visual disability is associated with the number of severe motor and cognitive disabilities. Concerning retinopathy of prematurity (ROP), Table 3 shows that 44% of early impairment cerebral palsy (EICP) with birthweight greater than 2500g are without normal vision. Over half the late impairment cerebral palsy children (LICP) are without normal vision. This surely cannot be attributable to ROP. We have calculated the survival of EICP infants of birthweight >2500g; 45.9% of those without normal vision live to age 30 years compared with 92.1% of those with moderate or no visual disability. This pattern holds in a multifactorial analysis. Clearly ROP does not explain the substantially poorer survival of people without normal vision.

We are criticised for not emphasising the need for further evaluation of the individuals with sensory impairment. To carry out such further evaluation would be a major and very expensive undertaking. It would entail the assessment of individuals from two counties born between 1966 and 1989. Ethical permission to establish the register was originally granted on condition that we did not approach the parents or the child. Any further assessment by us, therefore, would be ethically unacceptable. Even if such additional assessment was possible, what improvement would result in the estimates of the probability of survival?

Rosenbloom suggests that it would be helpful to sub-divide those with severe cognitive impairment into severe and profound sub-groups based on a dividing line of an IQ of <=20. He made this criticism when commenting on a previous paper from the North East England CP register (Hutton, Colver, Mackie Arch Dis Child 2000;83:468-474) to which a response was made. We accept that this sub-division may reveal significant differences in the probability of survival. It is because this criticism has been made previously, that we went to Rosenbloom with a sub-set of the children with severe cognitive disability who were from Liverpool. Many of these children were Rosenbloom’s patients. Over two years ago we requested that he sub-divided them into profound and severe sub-groups and we asked for precise criteria to be specified that would enable such a sub-division to be made. We are still awaiting a response. Specifying an IQ criterion of <=20 appears reasonable but it has its problems. IQ measurement at any level raises questions as to what it is actually measuring and the validity of the measurement. In children with severe cerebral impairment, the validity of IQ measures may be further compromised by the concurrent sensory, physical, speech and other problems.

Finally, the suggestion is made that statisticians, epidemiologists and clinicians should work together more coherently and productively. Rosenbloom was involved with the Mersey Region CP register when the database was established because many of the children being placed on to the register were from his Regional unit. He was a co-author on the first three scientific articles that came out of the database, all published in Archives of Disease in Childhood. We are unaware of any suggestions that he made as to what data should or should not be collected.

Yours sincerely,

J.L.Hutton
P.O.D.Pharoah