-
- Albert Lopez
Greetings to all,
Background information; the test is dissolution (paddle at 75 rpm; medium is phosphate buffer pH = 5.8) and the analysis is performed by HPLC using the following conditions:
Column: 75 mm x 4.6 mm, LUNA 3m C8 (2)
Mobile Phase: MeOH:0.75% Acetic acid in 20:80 ratio
Detector: UV, 243 nm (our method requires that we use a Waters 2487 detector)
Flow Rate: 1mL/min
Injection Volume: 7 mL
Run Time: 4 min
Approx. RT 2 min
During end of July, we initiated a sample set, via Empower, in a Waters Alliance HPLC system. The run was started after verifying that all instrument settings, calibration and initial system suitability parameters were within specified limits. On the following day, the sample set was processed and we found that the %RSD of the overall standard for the ingredient active did not meet the acceptance criteria (of NMT 2.5%). The overall standard is system precision check that we perform throughout the entire run, which includes the initial 5 consecutive standards and all subsequent bracketting standards (see attached pdf file). We obtained an RSD result of 14.5%. Initial system suitability was 0.9%. The overall standard areas were verified and we observed that we had two different sets of areas for the same standard solution to our surprise! The approximate area counts for standard soluton injections observed from vial #2 thru vial #11 (corresponding to the stability or pre-run, initial system suitability, standard check, one samples lot, and bracketing std_01 was 6,600,000. In contrast, an area count of approximately 4,900,000 was obtained for the remaining standard solution vials (std_02 and std_03) and for the samples of the second lot as well! The lowest standard area for the ingredient active was 4776827 and the highest 6601386. The former are typical than latter.
This dramatic drop, rather than drift, in the peak areas, for both standard and sample injections within run, suggested that we have an instrument reproducibility problem occurring final stage of run. Our analyst and instrumentation personnel checked this HPLC but found nothing unusual (with autosampler performance). A reproducibility test was performed and no problems were detected (initial and overall sys suit were great). A re-measurement from same original vials was performed the following day and indicated that the correct solutions were used to fill these vials (area count of 4,900,000 were observed for all solutions). Therefore, at first glance we thought that problem was attributed to an itermittent faulty injection problem (i.e., injecting 28% more than what method require). Since this test did not meet the method suitability requirement, we felt that this overall event is possibly related to equipment malfunction however we don't have an exact root cause or potential logical reason(s) why this occurred. Note: we checked various things as follows - checked the history of the column used for the original analysis and no significant issue was observed that could be attributed to the out of requirement result. The lamp energy was also evaluated and the documented value (91 for the reference energy) was within the acceptable criteria as per our SOP. The wavelength was verified and the documented value was the indicated at the method specification. Sample set was verified and the correct volume was used.
Sorry could not attached pdf file for std and sample sequence. Thanks upfront to anyone who has potential explanation or recommendation.
Background information; the test is dissolution (paddle at 75 rpm; medium is phosphate buffer pH = 5.8) and the analysis is performed by HPLC using the following conditions:
Column: 75 mm x 4.6 mm, LUNA 3m C8 (2)
Mobile Phase: MeOH:0.75% Acetic acid in 20:80 ratio
Detector: UV, 243 nm (our method requires that we use a Waters 2487 detector)
Flow Rate: 1mL/min
Injection Volume: 7 mL
Run Time: 4 min
Approx. RT 2 min
During end of July, we initiated a sample set, via Empower, in a Waters Alliance HPLC system. The run was started after verifying that all instrument settings, calibration and initial system suitability parameters were within specified limits. On the following day, the sample set was processed and we found that the %RSD of the overall standard for the ingredient active did not meet the acceptance criteria (of NMT 2.5%). The overall standard is system precision check that we perform throughout the entire run, which includes the initial 5 consecutive standards and all subsequent bracketting standards (see attached pdf file). We obtained an RSD result of 14.5%. Initial system suitability was 0.9%. The overall standard areas were verified and we observed that we had two different sets of areas for the same standard solution to our surprise! The approximate area counts for standard soluton injections observed from vial #2 thru vial #11 (corresponding to the stability or pre-run, initial system suitability, standard check, one samples lot, and bracketing std_01 was 6,600,000. In contrast, an area count of approximately 4,900,000 was obtained for the remaining standard solution vials (std_02 and std_03) and for the samples of the second lot as well! The lowest standard area for the ingredient active was 4776827 and the highest 6601386. The former are typical than latter.
This dramatic drop, rather than drift, in the peak areas, for both standard and sample injections within run, suggested that we have an instrument reproducibility problem occurring final stage of run. Our analyst and instrumentation personnel checked this HPLC but found nothing unusual (with autosampler performance). A reproducibility test was performed and no problems were detected (initial and overall sys suit were great). A re-measurement from same original vials was performed the following day and indicated that the correct solutions were used to fill these vials (area count of 4,900,000 were observed for all solutions). Therefore, at first glance we thought that problem was attributed to an itermittent faulty injection problem (i.e., injecting 28% more than what method require). Since this test did not meet the method suitability requirement, we felt that this overall event is possibly related to equipment malfunction however we don't have an exact root cause or potential logical reason(s) why this occurred. Note: we checked various things as follows - checked the history of the column used for the original analysis and no significant issue was observed that could be attributed to the out of requirement result. The lamp energy was also evaluated and the documented value (91 for the reference energy) was within the acceptable criteria as per our SOP. The wavelength was verified and the documented value was the indicated at the method specification. Sample set was verified and the correct volume was used.
Sorry could not attached pdf file for std and sample sequence. Thanks upfront to anyone who has potential explanation or recommendation.
and did it by hand? 