Status of drugs-of-abuse testing in urine under blind conditions: an AACC study.

We report results of a blind study designed to determine the accuracy of drugs-of-abuse testing in urine as done in 31 laboratories across the United States. The drugs studied were amphetamines, cannabinoids, cocaine, opiates, and phencyclidine. These laboratories confirmed all positive drug results with a different analytical method. Ten urine samples were sent to each laboratory, which resulted in 1486 trials. There were no false-positive results. The overall accuracy rate was 97%. Our study demonstrates that urine drug testing can be accurate when performed by qualified staff, using up-to-date screening and confirmation methods, appropriate quality-assurance measures, and a chain of custody.

Status of drugs-of-abuse testing in urine: An AACC study.

The study demonstrates that, in May 1987, testing urine for drugs of abuse can be accurate. All laboratories challenged in this study currently perform such testing under contract and are involved in monthly proficiency testing and in-service training (AACC's Surveys Plus and In-Service Training Program in Toxicology). The laboratories were challenged to detect drugs at the concentrations at which they accept business. We suspect that when results of studies of this kind have been reported previously, the laboratories may have been scored inaccurately because they used technology designed to detect higher concentrations of the drugs than were weighed into the study specimens. This points up the need for laboratories and clients to be specific about the threshold concentrations used to report positives and policies for reporting positives detected below those concentrations.

Lithium monitoring.

Lithium is the drug of choice for treatment of mania and for long-term maintenance to prevent both depressive and manic episodes in bipolar disorders. A thorough knowledge of lithium is necessary for its safe and effective use. It must be used with caution owing to its narrow therapeutic range and significant toxicities. Patients should be screened prior to initiating lithium therapy. The drug-drug interactions must be appreciated, and it is recommended that the lithium concentration, thyroid function, and renal function be monitored in addition to the clinical progress. Blood samples for lithium monitoring should be obtained 12 hours after the evening dose.

Evaluation of the Beckman Creatinine Analyzer 2.

The Beckman Creatinine Analyzer 2 was evaluated in our laboratory. This new instrument determines creatinine by a Jaffe rate method and utilizes an optical detection system which measures the rate of change in absorbance at 520 nm 25.6 seconds after sample introduction. The linearity of the instrument was verified up to creatinine concentrations of 24 mg/dl. Recovery averaged 92% from serum and 97% from urine. Comparison studies using serum specimens were made against a manual, non-kinetic Jaffe method and a Technicon SMA 12/60. Urine comparisons were made against the manual method. All comparisons indicate the Beckman analyzer yields lower creatinine values due to the minimization of interferences from non-creatinine chromogens. No significant change in these correlations was observed with elevated bilirubin concentrations (greater than 10 mg/dl) or with lipemic specimens (triglycerides up to 3390 mg/dl). Both within-day and day-to-day precision were shown to be within the stated manufacturer's specifications for both sera and urine specimens.

Calibration and monitoring of spectrometers and spectrophotometers.

We have delineated some of the factors affecting the performance of spectrometers and spectrophotometers in the clinical laboratory and have presented some of the methods for verifying that these instruments are functioning properly. At a minimum, every laboratory should perform periodic inspections of spectrometric functions to check wavelength calibration, linearity of detector response, and stray radiation. Only through such an inspection program can a laboratory ensure that these instruments are not contributing to inaccurate analytical results.

Quantitative high-performance liquid-chromatographic method for determining disopyramide (Norpace) in serum.

We report a high-performance liquid-chromatographic method for measuring disopyramide (Norpace, Searle) in serum. The drug is extracted from 0.5 ml of serum into chloroform containing the internal standard p-chlorodisopyramide, separated on a reversed-phase octadecylsilyl column at room temperature, and detected at 254 nm. The method is sensitive to 0.2 mg of disopyramide per liter, with a linear response to at least 10 mg/liter. Within-day precision (CV) for frozen serum pools is 8.3 (n = 21) and (n = 22) at mean concentrations of 2.6 and 5.9 mg/liter, respectively. Day-to-day precision (CV) is 9.4 (n = 80) and 9.3 (n = 29) at mean concentrations of 2.8 and 6.7 mg/liter, respectively. Recoveries for disopyramide in serum averaged 98% over the linear range when compared against an aqueous standard taken through the entire analytical procedure. The method is relatively specific, as evidenced by interference studies with over 85 drugs.

Colorimetric method for the quantitative determination of acetaminophen in serum.

A colorimetric method for quantitatively determining acetaminophen in serum is presented. The procedure involves preparing a protein-free filtrate by the addition of trichloroacetic acid, hydrolyzing the acetaminophen in the filtrate to p-aminophenol, and subsequent reaction of p-aminophenol with phenol and NH4OH to form an indophenol blue chromogen. The absorbance at 620 nm of this chromogen follows Beer's law up to acetaminophen concentrations of greater than 50 mg/1. The recovery of acetaminophen from serum is essentially 100% when compared with appropriately prepared standards. Serum acetaminophen concentrations as low as 1 mg/1 can be detected. Seventy drugs were tested for possible interference and none was found to interfere with the method. The coefficient of variation (day-to-day) is 4%. This new quantitative method provides the necessary sensitivity to quantitate therapeutic as well as toxic serum concentrations of acetaminophen.

Interferences with the starch-iodine assay for serum amylase activity, and effects of hyperlipemia.

Recent reports imply or claim that amylase activity is inhibited in hyperlipemic sera, because diluted samples showed greater activities when assayed by a starch-iodine method. We find that dilutions of both clear and lipemic samples give higher-than-expected activities when assayed by a starch-iodine method, an effect attributable to the variable effects of protein, turbidity, and triglycerides in the hyperlipemic samples. Thus the starch-iodine method is unsuitable for assessing the effects of hyperlipemic samples on amylase activity. To do so, we used an alternative method, in which soluble dyed amylopectin is used as the substrate. This method exhibits apparent zero-order kinetics, and we detected no interfering factors. Plots of sample volume (x) vs. activity (y) for clear and hyperlipemic (triglycerides up to 80 g/liter) sera gave straight lines with y-intercepts of zero. Evidently hypertriglyceridemia does not inhibit amylase activity.

A modified procedure for determining metanephrines in urine.

We have modified the procedure for determining total metanephrines in urine described by Gupta, Price and Keane(1) to shorten the time necessary for completion while allowing the technician time to perform other tests. The modification involves the extraction of interfering substances from urine. Instead of having to extract 3 individual tubes for each patient, we have developed an extraction that can be performed in a single tube per sample. This allows the number of tubes handled in the first portion of the procedure to be reduced by 66% and a reduction in analysis time of 25%. Day-to-day precision of the method is 8.7%. The modified procedure was compared to the original procedure by running 38 different patients by each method. The mean for the original and modified method were 1.1 mg/24h and 1.1 mg/24h respectively. The correlation coefficient was 0.9051. The t-test (p = 0.05) indicated that there was no significant difference in the means of the methods and the F-test (p = 0.05) indicated that there was no significant difference in the precision of the methods.

Convenient method for checking detector response of spectrophotometers at three wavelengths.

We describe the use of a water-soluble dye, available as green food coloring, to check the linearity of detector response of spectrometers and spectrophotometers at three wavelengths. The dye has absorbance maxima at 257 nm, 410 nm, and 630 nm. The advantages of using this solution for this important instrument function check are: one solution with absorption maxima in the opposite ends of the visible spectrum as well as in the ultraviolet region; safe; inexpensive; readily available; easy to use; absorbance not dependent on pH from 2.5 to 10.0 or on temperature from 4 to 56 degrees C.