Determination of ions in paediatric NPI bags using a microwave plasma atomic emission spectrometer: not so easy for the pharmacist!

Introduction
For the past three years, the routine determination of Na and K ions in NPI bags has been carried out by a microwave plasma atomic emission spectrometer, the MP AES 4210, which is more complex to handle than the flame photometer previously used.
Following the difficulties encountered by some users, we wanted to quantify and identify the nature of the problems. To do this, we set up a questionnaire to be completed by the pharmacist to monitor usage.

Material and method
We distributed a questionnaire to be completed daily by the pharmacist or intern for 25 days over a period of 5 weeks. It lists the number of samples to be analysed, those that were re-run, the number of bags re-done and whether the help of another person was necessary. A deviation of more than +/- 10% of the theoretical concentration of Na and K is considered as non-compliant (NC). After 2 NC results, the bag is destroyed.

Results
Four pharmacists and 1 newly trained intern completed the questionnaire. A total of 508 samples were analysed, of which 47 samples were reanalysed (9.25%) and 1 NPI (0.2%) was remanufactured for non-compliance. 35 of the 47 samples reanalysed (75%) were due to a technical error by the pharmacist (range error, forgot to calibrate the device). The intern needed help and called the referring pharmacist on 6 occasions, while a pharmacist called the referring pharmacist once.
Two breakdowns occurred during the study, after the start of the day’s production. 25 bags were destroyed due to a lack of analysis at a cost of €1250. The MP AES 4210 was not functional for 3 days, imposing a double control production. The first problem turned out to be a breakdown of the nitrogen generator. We had encountered the same problem a month earlier. The pharmacist and the biomedical staff were unable to restart the device. The help of the referring pharmacist was necessary. The second failure concerned the sampling probe, which was blocked. After analysis, it turned out that the preventive action of cleaning the probe every 6 months had not been carried out by the biomedical staff. Cleaning it solved the problem.

Discussion - Conclusion
Although the results provided by the MP AES are reliable, its use remains complex, particularly in the management of breakdowns. This leads to additional costs for the service and disorganisation of the team. It seems necessary to write quality documentation resources, allowing the user to be autonomous and more serene in the management of problems during the analysis of NP samples, in particular during intern changes. With a hindsight of 3 years, we have been able to identify and solve many problems, which will enable us to create a decision tree that is as exhaustive as possible according to the various difficulties encountered.

See le poster