Urine Testing

Injection of unprocessed urine samples could cause massive interferences, matrix effects, or damage to the highly sensitive instruments that are used for analysis. Therefore, an extraction method is required for urine samples. During this process, the internal standards are added, and the samples are diluted to produce the best results with minimal matrix effects.

The extraction method begins with the addition of equal volumes of the urine sample and the internal standard methanol solution, which causes protein precipitation. After thorough mixing and centrifuging to separate any solid particulates, the liquid (170 microliters) is diluted in HPLC grade deionized water (800 microliters), the supernatant is transferred into a glass HPLC autosampler vial, and 8 µL of the sample extract is injected onto the HPLC column.

While there are other accepted methods of determining these values such as dipstick testing, the use of the Indiko^TM system allows for more precise values that provide a more detailed analysis of the patient samples. Abnormal levels indicate some form of impurity or contamination in the sample that was introduced during collection of the sample or at any point before extraction.

An “inconsistent” value indicates that the sample provided was either not human urine but a different liquid such as artificial urine or was intestinally modified to mask drug use, and results provided may not reflect the patient’s true levels. Abnormal or inconsistent urine samples are analyzed as normal, but results will likely not reflect the true concentrations of drug analytes for that patient. When abnormal or inconsistent levels are detected in samples, they are flagged for clinicians so that LC-MS/MS results can be interpreted with any disparities in mind.

Development of the LC-MS/MS method and assay required monitoring of accuracy, precision, and sensitivity to ensure the results are reliable. Matrix effects, meaning the interaction of several different compounds in the ionization interface at the exact same time that can lead to suppression or enhancement of the ionization process, are a common potential problem of testing for multiple unknowns in one assay. Electrospray ionization tends to show more matrix effects than atomic pressure chemical ionization, due to their slightly different ionization methods.8, 11 Comparing peak intensities of drug added to extracted matrix versus pure solvents can provide insight into ionization effects and is an important aspect in developing a method. Additionally, a post-column infusion method could be applied to assess matrix effects. For more details, please refer to references8, 11. This involves the addition of a standard or blank to identify which regions of the ion chromatogram are likely to experience matrix effects.11 To minimize matrix effects many parameters can be modified and adjusted. Smaller injection volumes and flow rates, dilution of the sample, the use of an internal standard, utilizing a column with a smaller diameter, more efficient and specific extraction procedures, or better HPLC separation from interfering substances may reduce or compensate for these effects.8, 11 

Another common issue with analysis of samples with high concentrations of analytes is the carryover of drug substances to the LC-MS/MS system. In most cases the source of such carry-over is the autosampler or the chromatography column. To minimize carry-over, the autosampler undergoes several wash steps of the injection needle, syringe and switching valve. Additionally, methanol blanks are injected in between each sample to elute off any drugs from the previous injection that may have remained bound to the column, as a way to clean out the column. 

An important step in method development are stability studies which determine the stability of each analyte included in the assay in the urine matrix. To test stability of the instrument’s autosampler, analysis is run on samples immediately after extraction and after 48 hours in the autosampler sample tray kept at 4°C. The stability after freeze-thawing at -80°C in three cycles compared to initial analysis is also required in case re-analysis of stored samples is requested.12, 13 Additionally, long-term stability of the samples is important to determine how analytes degrade in the urine matrix over time. The validation of this assay showed that after 48 hours in the autosampler and three cycles in the -80°C freezer, all analytes were detected within the acceptance criteria.12 After three months of storage at -80°C, only two of the 136 analytes tested were significantly degraded while all others were within the criteria.12

A system of quality assurance and quality control is necessary when validating an assay for use on clinical samples. Quality assurance includes monitoring the entire process from preparing samples to reporting results including documentation and monitoring the personnel’s qualifications, skills, and training recors.¹⁴ The development of standard operating procedures, cross-validations, and external proficiency testing such as provided by the College of American Pathologists (www.cap.org) are part of proactive quality control and assurance methods. Quality control refers to the continual evaluation of uncertainty of results and the calibration of instruments.14 Calibrators and quality controls are analyzed with every batch of samples, and contain known concentrations of each analyte in the same urine matrix. This is necessary for the assessment, control, and documentation of the accuracy and precision of the assay.

Due to high levels of variability between the urine matrix from individuals, controls extracted from “blank” or clean urine is important for the assessment of the quality of the assay.¹¹ A passable method will yield reproducible results between calibrators, quality controls, and samples from individuals with inconsistent matrices. New calibrators and quality controls are prepared every 48 hours for the best results and are analyzed daily to keep track of the instruments’ performance. The calculation of concentrations compared to known concentrations is a main component of quality control and is the most important system in place for providing proof that the results provided are valid on a daily basis.

An important part of quality assurance is the analysis of the “trueness” of calculated values for samples. This is assessed regularly by College of American Pathologists proficiency testing. Precision is also monitored for each set of quality controls by calculating their coefficient of variation (CV%). The CV% for quality controls should be within 15% of true concentration to ensure the sensitivity is being maintained.¹³ The accuracy and precision of the assay are monitored by determining the CV% within each run and in between separate runs, as well as long-term over several months and even years.

Specific parameters for the quality control of a clinical LC-MS/MS assay depends on the scope of the assay and signal to noise ratios. Applicable guidelines provided by the United States Food and Drug Administration and Clinical Laboratory Standards Institute (www.clsi.org) have been followed and adapted to meet the requirements of the United States Clinical Laboratory Improvement Amendments and College of American Pathologists.^{12, 13}

The purpose of a calibration curve is to ensure the concentrations being calculated by the LC-MS/MS system are precise and accurate. Accurate curves have concentrations close to the true values. The calculated concentrations of calibrators should be within ±15% of the exact concentrations being used for the curve, while the lower limit of quantification (LLOQ, listed in Table 3 for each analyte) should be within ±20%.13 To ensure linearity of the calibration curve, the correlation coefficient (r) of the fit should be at least 0.995.¹⁵ 

Processing a calibration curve of ten concentrations from zero to 1000 ng/mL, including a blank that lacks internal standard, for each batch of samples ensures the assay is performing correctly and the results could be deemed accurate. As concentration among drugs and their metabolites in urine vary widely, the clinically relevant confirmatory ranges for each analyte (Table 3) dictate which calibrator concentrations are included in the analysis, and those outside of the range are not required to pass for that specific analyte.

Quality controls are used to further evaluate the precision, accuracy, and stability of the assay. At least two thirds of all the quality controls’ calculated concentrations should be within ±15% of the nominal concentrations for a passable batch of data. A set of eight quality controls with concentrations of 3, 6, 15, 37.5, 75, 200, 375, and 750 ng/mL are processed with each batch of samples to ensure the results are accurate.¹²