A letter in response to ‘Liquid chromatography tandem mass spectrometry: challenges in introducing published methods into the clinical laboratory’ by Khedr et al.

We would like to thank Khedr et al. for their interest in our paper and would agree with many of their points. We believe that they highlight some of the problems encountered when trying to reproduce mass spectrometry methods in different laboratories using different instruments. It isn’t always appreciated that the choice of mass spectrometer can have a significant effect on the method, and Khedr et al. make an excellent point when they found that oxalate was difficult to ionize using the Xevo TQ-S. We used an entry level Waters Xevo TQD instrument for our work which has different characteristics when compared to the Xevo TQ-S. We believe it would be difficult to use a Xevo TQ-S tandem mass spectrometer for the ‘dilute and shoot’ approach without a significant cleanup procedure as highlighted in our method. The TQ-S is a highly sensitive mass spectrometer and any impurities in urine which remain after insufficient sample purification may have a detrimental effect on long-term performance of the mass spectrometer.

With regard to the suboptimal product ion generation, we have found that the environment within the StepWave of the Xevo TQ-S is very harsh; for example, ionization of vitamin D using the Xevo TQ-S often results in loss of water. The harsh ionization of the small oxalate parent ion (m/z 88.9) is unlikely to result in a useable product ion as Khedr et al. saw in their preliminary studies.

We agree with the fact that oxalate shows significant ion suppression (even after significant cleanup); nevertheless, the use of a C₁₃ internal standard combined with ion exchange cleanup was found to compensate effectively for this. As mentioned by Khedr et al., we did not see ion suppression with citrate which is most likely due to differences in the sample preparation, chromatography and detection.

When looking at the matrix effect (ME), we used the quantitative approach which gives ME as a percentage to assess whether the method is acceptable or changes need to be made and we believe this provides more useful information than the qualitative assessment of ME alone. As stated in the manuscript, we found ion suppression in both analytes to be acceptable because internal standard compensates for any loss or enhancement. Due to optimization of sample cleanup and chromatography conditions from the start, we were able to use PBS-based calibrators which simplifies the process.

The combined oxalate and citrate assay has now been in use for over a year (as of December 2017) with acceptable EQA performance and reproducibility for both oxalate and citrate. In addition to this, we routinely monitor retention times and peak height to look for any unexpected changes which may affect assay performance; we have found our assay to be very robust over this time and we continue to use it for routine stone screens. In conclusion, we would encourage the use of our complete published method on an entry level instrument which should produce more favourable results. As we have stressed in the manuscript, we believe that sample specific issues exist with the enzymatic method for oxalate measurement and would actively encourage development of more LC-MS/MS assays for both oxalate and citrate.