Dear Editor

We are astonished by Wiwanitkit's inaccurate remarks regarding the number of subjects studied and analytical methodology in our study [1]. The data set was of more than sufficient size to be subjected to rigorous statistical analyses. If Wiwanitkit had bothered to even casually read the articles he cites [1,2], he should have immediately realised that firstly the thrust of Lenters-Westra's and Slingerland's paper was to highlight poor analytical performance of some HbA1c point-of- care (not laboratory based) instruments and secondly that laboratory HbA1c analyser used in our study (Tosoh G7, cation-exchange HPLC) was a reference method for measurement of HbA1c in study of Lenters-Westra and Slingerland [1,2]. In addition our HbA1c assay, and indeed all our assays, are subject to stringent internal quality control procedures and external quality assurance. We had hoped our article would stimulate meaningful debate rather than inane comment.

Dr Taruna Likhari, Professor Rousseau Gama

References

1. Likhari T, Gama R. Ethnic differences in glycated haemoglobin between white subjects and those of South Asian origin with normal glucose tolerance. J Clin Pathol 2010; 63:278-280

2. Lenters-Westra E, Slingerland RJ. Six of eight hemoglobin A1c point-of- care instruments do not meet the general accepted analytical performance criteria. Clin Chem 2010;56:44-52

Conflict of Interest:

None declared

Dear Editor,

I read the recent publication by Likhari et al with a great interest [1]. Likhari et al concluded that "In subjects with similar fasting and postprandial glycaemia on OGTT, those of South Asian origin have higher HbA(1c) levels than white subjects. It is speculated that the higher glycaemia-independent HBA1c levels in people of South Asian origin could possibly contribute to their increase cardiovascular risk [1]." I have some additional concerns on this work. First, the rather few subjects in this work reduce the value of the research. It is questionable whether the number of subjects in this work is statisticall acceptable or not. Second, the concern on the performance of the analyzers for HBA1C should be raised. In a recent publication, only a few analyzers are acceptable for precision and accuracy [2].

1. Likhari T, Gama R. Ethnic differences in glycated haemoglobin between white subjects and those of South Asian origin with normal glucose tolerance. J Clin Pathol 2010; 63:278-280

2. Lenters-Westra E, Slingerland RJ. Six of eight hemoglobin A1c point-of- care instruments do not meet the general accepted analytical performance criteria. Clin Chem 2010;56:44-52

Conflict of Interest:

None declared