The STR system in the human phenylalanine hydroxylase gene: true fragment length obtained with fluorescent labelled PCR primers

We present a simple, fast, non-radioactive method for the analysis of the polymorphic short tandem repeat (STR) system in the human phenylalanine hydroxylase gene. Previously, sizing of the STR marker involved radiolabelling of PCR amplified fragments and resolution on denaturing polyacrylamide gels using M13 sequencing ladder as a standard. However, this method consistently gave sizes 2 bp longer than the known sequence. The fluorescent method presented here employs internal lane standards and enables accurate sizing of the fragments. To avoid confusion, we suggest that the true fragment lengths are used as reference values in the future. The analysis of STR alleles is valuable for population genetic studies and for targeted mutation screening in phenylketonuria (PKU). It can replace RFLP-based haplotype analysis for carrier detection, and we report its use for prenatal diagnosis in a Northern Irish family with PKU. The analysis of 250 Northern Irish chromosomes, including 128 PKU alleles, showed no significant difference between normal and PKU alleles, with fragment lengths of 238 and 242 bp most common in both groups.     

Molecular analysis of PKU in Ireland

Classical phenylketonuria (PKU: McKusick No. 261600) is caused by mutations occurring at the phenylalanine hydroxylase (PAH) locus on chromosome 12 and has a prevalence in Ireland of 1 in 4500. We examined 304 independent alleles from 350 patients for the presence of six mutations and have characterized VNTR alleles within the minisatellite region 3' to the PAH gene in patients carrying the most prevalent mutation. R408W was the most common mutation found, with a relative frequency of 42%. All other mutations had relative frequencies of <10%. VNTR analysis showed that the R408W mutation is associated with the VNTR-8 allele in the Irish population, indicating that R408W is associated with RFLP haplotype 1. This differs from that reported from eastern Europe where R408W is associated with RFLP haplotype 2/VNTR-3; an observation which has led several groups to propose a Balto-Slavic origin for this mutation. These results support the hypothesis of a second, independent founding event for the R408W mutation on an RFLP haplotype 1 VNTR-8 chromsome background in the Irish/Celtic population.     

Fructosamine and glycated haemoglobin in the assessment of long term glycaemic control in diabetes

Fructosamine and glycated haemoglobin were measured simultaneously in 147 children with diabetes. If glycated haemoglobin is considered as the 'gold standard' for long term glycaemic control, then fructosamine is a poor indicator of actual glycated haemoglobin values, with wide 95% confidence (fiducial) limits. This shows that it is impossible to accurately predict glycated haemoglobin concentrations and therefore, by implication, longer term glycaemic control, from measurements of fructosamine. As the major studies on the prevention of microvascular complications in diabetes have used glycated haemoglobin levels to assess glycaemic control, it is suggested that this measurement should be used in all children with diabetes in preference to the measurement of fructosamine.     

Catch-up growth up to ten years of age in children born very preterm or with very low birth weight

Background  

Improved survival due to advances in neonatal care has brought issues such as postnatal growth and development more to the focus of our attention. Most studies report stunting in children born very preterm and/or small for gestational age. In this article we study the growth pattern of these children and aim to identify factors associated with postnatal catch-up growth.     

World health system performance revisited: the impact of varying the relative importance of health system goals

Background

In 2002, the World Health Organization published a health system performance ranking for 191 member countries. The ranking was based on five indicators, with fixed weights common to all countries.

Methods

We investigate the feasibility and desirability of using mathematical programming techniques that allow weights to vary across countries to reflect their varying circumstances and objectives.

Results

By global distributional measures, scores and ranks are found to be not very sensitive to changes in weights, although differences can be large for individual countries.

Conclusions

Building the flexibility of variable weights into calculation of the performance index is a useful way to respond to the debates and criticisms appearing since publication of the ranking.