Aims to describe the undiagnosed children in a UK study of progressive intellectual and neurological deterioration (PIND) and evaluate their investigations.

Methods Since March 1997 this study has used the British Paediatric Surveillance Unit to identify UK children with PIND. Clinical data are gathered from paediatricians by questionnaire, telephone interview or site visit. The PIND Study Expert Group of specialists in paediatric neurology, neurometabolic disease and neurogenetics independently reviews the cases.

Results By July 2016 4063 children with suspected PIND had been notified. There were 1958 finalised cases meeting the criteria for PIND; 1749 (89%) were diagnosed, a group of more than 190 different disorders. There were also 209 (11%) (111 male, 98 female) in whom investigations had failed to provide a diagnosis or who had died undiagnosed. Distribution by ethnicity was: White 104 (50%), Asian 85 (Pakistani 65, Indian 6, Bangladeshi 4, Asian unspecified 10) (41%), Black 5 (2%), other 7 (3%), not stated 8 (4%). There was a high rate of consanguinity: 90/209 (43%) over all (89% in the 65 Pakistani families). 72/209 (34%) had a similarly affected relative (57 of these were siblings).

The undiagnosed cases had usually been very thoroughly investigated with, where appropriate, blood and urine metabolic tests, CSF examination, plasma white cell enzymes, EEG, ERG and VER, ECG, echocardiogram, skin biopsy, muscle biopsy, karyotype, DNA studies, CT and MRI brain scans. For example 174/ 209 (83%) had at least one MRI brain scan. 152/209 (73%) had died. Only 10 were known to have had full post mortem examinations, 4 had limited autopsies. 94 did not have post mortems (44 cases: not known). This low autopsy rate was also found in the diagnosed PIND cases where most diagnoses were made during life.

Conclusion 11% of PIND children were undiagnosed despite intensive investigation. High rates of consanguinity (43%) and of a positive family history (34%) suggest that many had (currently unrecognised) neurometabolic diseases. Autopsies are rarely performed so storage of selected tissues for DNA and other analyses should be recommended in these undiagnosed cases.

Acknowledgment independent research funded by Department of Health Policy Research Programme [121/6443].