Effect of outliers and nonhealthy individuals on reference interval estimation.

BACKGROUND: Improvement in reference interval estimation using a new outlier detection technique, even with a physician-determined healthy sample, is examined. The effect of including physician-determined nonhealthy individuals in the sample is evaluated. METHODS: Traditional data transformation coupled with robust and exploratory outlier detection methodology were used in conjunction with various reference interval determination techniques. A simulation study was used to examine the effects of outliers on known reference intervals. Physician-defined healthy groups with and without nonhealthy individuals were compared on real data. RESULTS: With 5% outliers in simulated samples, the described outlier detection techniques had narrower reference intervals. Application of the technique to real data provided reference intervals that were, on average, 10% narrower than those obtained when outlier detection was not used. Only 1.6% of the samples were identified as outliers and removed from reference interval determination in both the healthy and combined samples. CONCLUSIONS: Even in healthy samples, outliers may exist. Combining traditional and robust statistical techniques provide a good method of identifying outliers in a reference interval setting. Laboratories in general do not have a well-defined healthy group from which to compute reference intervals. The effect of nonhealthy individuals in the computation increases reference interval width by approximately 10%. However, there is a large deviation among analytes.

Standards of laboratory practice: theophylline and caffeine monitoring. National Academy of Clinical Biochemistry.

Theophylline is used in the treatment of asthma and chronic obstructive pulmonary disease. The use of theophylline has declined with the advent of potent steroid inhalants. Because of the therapeutic index of this drug, monitoring of theophylline concentrations in plasma is essential. Monitoring should be done on trough specimens after steady-state has been reached. Non-steady-state concentrations may be indicated in selected situations. Caffeine is used to treat apnea of the newborn because of its low toxicity. Monitoring is often by clinical effect. Monitoring of serum concentrations should be performed in cases where there is no clinical response or if there is suspected toxicity.

A robust approach to reference interval estimation and evaluation.

We propose a new methodology for the estimation of reference intervals for data sets with small numbers of observations or for those with substantial numbers of outliers. We propose a prediction interval that uses robust estimates of location and scale. The SAS software can be readily modified to do these calculations. We compared four reference interval procedures (nonparametric, transformed, robust with a nonparametric lower limit, and transformed robust) for sample sizes of 20, 40, 60, 80, 100, and 120 from chi 2 distributions of 1, 4, 7, and 10 df. chi 2 distributions were chosen because they simulate the skewness of distributions often found in clinical chemistry populations. We used the root mean square error as the measure of performance and used computer simulation to calculate this measure. The robust estimator showed the best performance for small sample sizes. As the sample size increased, the performance values converged. The robust method for calculating upper reference interval values yields reasonable results. In two examples using real data for haptoglobin and glucose, the robust estimator provides slightly smaller upper reference limits than the other procedures. Lastly, the robust estimator was compared with the other procedures in a population where 5% of the values were multiplied by a factor of 5. The reference intervals were calculated with and without outlier detection. In this case, the robust approach consistently yielded upper reference interval values that were closer to those of the true underlying distributions. We propose that robust statistical analysis can be of great use for determinations of reference intervals from limited or possibly unreliable data.

Clinical chemistry and post-liver-transplant monitoring.

Liver transplantation is an accepted therapy for end-stage liver disease. After allografting, a variety of clinical problems may require laboratory involvement for accurate and timely diagnosis and intervention. Critical factors in the choice of a laboratory test menu to support a transplant program include turnaround times that support clinical decisionmaking, real diagnostic value, and real value for money. Particular clinical problems, whose early presentation must be anticipated, include graft ischemia, primary nonfunction, and hepatic artery thrombosis. Acute rejection is common at 5-10 days posttransplantation, the principal target being the biliary tree. Longer-term problems are associated with the therapeutic drug measurement of cyclosporin A and, increasingly, tacrolimus (FK506); the side effects of immunosuppressant therapy also require monitoring. A successful liver transplant program can be adequately supported with a simple battery of automated tests that are cheap, fast, and available at all times.

Pharmacokinetics of zidovudine phosphorylation in peripheral blood mononuclear cells from patients infected with human immunodeficiency virus.

As part of an effort towards optimization of dosing of zidovudine (ZDV), formation and elimination of total phosphorylated ZDV (ZDVPt) in peripheral blood mononuclear cells were examined in 21 asymptomatic human immunodeficiency virus-infected patients during their first 24 weeks of therapy (AIDS Clinical Trials Group Protocol 161). Intracellular concentrations of ZDVPt were measured with a previously described and validated radioimmunoassay technique. Although ZDV phosphorylation occurred readily upon initiation of therapy, it declined with time; the area under the concentration-time curve (AUC) at week 4 (mean +/- standard deviation, 3.41 +/- 0.93 pmol.h/10(6) cells) was significantly greater than that at week 24 (2.19 +/- 1.10 pmol.h/10(6) cells). Plasma ZDV AUC did not change with time and did not correlate with ZDVPt AUC. In dose-response experiments (20 to 100 mg orally), phosphorylation did not proportionally increase with increasing plasma ZDV concentrations. Similarly, compared with a single dose, two doses of ZDV over an 8-h period resulted in little ZDVPt increase in cells relative to increase in plasma ZDV concentrations. The half-life of intracellular ZDVPt was twice that of plasma ZDV (4 versus 2 h), suggesting that an every-8-h dosing regimen is justifiable. These findings suggest that metabolism of ZDV to its active intracellular forms may be saturable in some patients, is poorly correlated with plasma concentrations, and diminishes over time. These findings have implications for future development and management of anti-human immunodeficiency virus nucleoside therapy.

Correlates of zidovudine phosphorylation with markers of HIV disease progression and drug toxicity.

OBJECTIVE: To determine the relationships between in vivo zidovudine (ZDV) phosphorylation in cells from HIV-infected patients and markers associated with disease progression and drug toxicity. DESIGN: A pharmacokinetic study of ZDV metabolism sponsored by the AIDS Clinical Trials Group (protocol 161). Plasma and intracellular pharmacokinetics following a 100 mg oral dose of ZDV were determined at weeks 4 and 24 of initial therapy in adult patients. Plasma concentrations and phosphorylated ZDV were determined by radioimmunoassay, and area under the concentration-time curves (AUC) were compared with clinical data collected during the pharmacokinetic study. SETTING: An outpatient setting at the University of Cincinnati AIDS Treatment Center, Cincinnati, Ohio, USA. PATIENTS: HIV-infected adults with CD4+ lymphocyte counts 200-500 x 10(6) cells/l with no prior history of anti-HIV therapy and no active infections requiring systemic therapy. Of 30 patients enrolled, 21 were evaluable. INTERVENTIONS: None. MAIN OUTCOME MEASURES: AUC of plasma ZDV and intracellular total phosphorylated ZDV were compared with change from baseline of the following surrogate markers: CD4+ lymphocyte count, %CD4+ lymphocytes, CD4+/CD8+ cell ratio, serum beta 2-microglobulin, serum neopterin, neutrophils, red cell count, and hemoglobin. RESULTS: No correlations between plasma AUC and markers of therapeutic response were observed. However, significant positive correlations were observed between the AUC of total phosphorylated ZDV and changes in the %CD4+ lymphocytes and CD4+/CD8+ lymphocyte ratio; a negative correlation was observed with change in hemoglobin. Patients who responded to ZDV therapy, as measured by these variables, demonstrated significantly higher intracellular AUC (> 3 pmol x h/10(6) cells) than those who did not (approximately 2 pmol x h/10(6) cells). CONCLUSIONS: The ability of HIV-infected patients to phosphorylate ZDV correlates with changes in markers associated with drug effect and toxicity. Potential individualization of therapy through monitoring of total phosphorylated ZDV in patients therefore warrants further exploration.

Electrochemical homogeneous enzyme immunoassay of theophylline in hemolyzed, icteric, and lipemic samples.

We demonstrate here an electrochemical homogeneous enzyme immunoassay for theophylline, which can be performed in hemolyzed, lipemic, and icteric samples. The assay used an unmodified Syva EMIT theophylline kit. One of the enzymatic reaction products, NADH, reacted with 2,6-dichloroindophenol (DCIP) to reduce DCIP to DCIPH2, which was detected electrochemically with flow-injection analysis. The inter- and intraassay coefficients of variation of this manual technique were < 9% at theophylline concentrations of 14 to 34 mg/L. The CVs were 9-15% at low concentrations (6.3 mg/L), which is below the therapeutic range. Analytical recoveries were 91-97% for normal serum and 92-111% for hemolyzed, icteric, or lipemic sera. The measured concentrations (y) were compared with those obtained by the fluorescence polarization immunoassay (x); a scatter plot of the results showed a linear relationship of y = 1.00 x - 0.57 mg/L (r = 0.966, Sy/x = 1.51). This alternative way to measure the serum concentration of theophylline overcomes the shortcomings of spectrophotometric methods, by which it is difficult to measure theophylline in severely hemolyzed, icteric, or lipemic sera.

Clinical correlations of cyclosporine-specific and -nonspecific assays in stable renal transplants, acute rejection, and cyclosporine nephrotoxicity.

Accurate and early diagnosis of the cause of renal transplant dysfunction is important in successful patient management. Controversy exists as to whether a cyclosporine-specific or -nonspecific method is more predictive of clinical events. In an attempt to answer this question, all episodes of acute renal dysfunction were reviewed in 322 stable renal transplant recipients over a 20-month period. To diagnose the cause of each episode of renal dysfunction, an analysis was made of patient demographics; weight; serum creatinine; cyclosporine dose; cyclosporine level, using a specific method--high-performance liquid chromatography (HPLC)--and a nonspecific method--fluorescent polarization immunoassay (FPIA); changes in cyclosporine dose; renal biopsy; and response to any therapeutic intervention. There were 138 patients, who developed 279 episodes of renal dysfunction. Causes of renal dysfunction were cyclosporine-related (n = 103), acute rejection (n = 63), extracellular fluid volume depletion (n = 27), other (n = 59), and unknown (n = 27). The mean HPLC cyclosporine level was significantly different in patients with acute cyclosporine toxicity (p < 0.001) and patients with acute rejection (p < 0.001) when compared to those with stable renal function; the mean FPIA cyclosporine levels were not significantly different between the three groups. However, a larger percentage of patients with rejection were subtherapeutic when measured by HPLC, while a higher proportion of patients with nephrotoxicity were above the therapeutic range measured by FPIA.(ABSTRACT TRUNCATED AT 250 WORDS)

Monoethylglycinexylide formation in assessing pediatric donor liver function.

Lidocaine metabolism to monoethylglycinexylide (MEGX) has been described as a novel method to assess liver function in adult transplant donors and recipients. While this assay appears to offer a number of advantages over existing liver function tests, limited work has been done to evaluate its potential in the pediatric population. This study evaluated MEGX formation in potential pediatric liver donors (n = 35) and a control group of children (n = 16). The mean MEGX formation was significantly higher in pediatric donors than in the control group (156 +/- 62 vs 106 +/- 33 ng/ml, p < 0.05). No correlation with age, total bilirubin, liver transaminases, or alkaline phosphatase could be made within each group. Significant differences in MEGX levels were noted when each group was compared to its adult counterpart. Both pediatric donors and controls had greater mean MEGX formation than has been reported for adult donors and controls (156 +/- 62 vs 127 +/- 61 ng/ml, p < 0.05 and 106 +/- 33 vs 72 +/- 36 ng/ml, p < 0.05, respectively). Drugs that alter lidocaine pharmacokinetics and their potential influence on MEGX formation were evaluated in the pediatric donor group. Donors exposed to hepatic enzyme-inducing drugs had a higher mean MEGX formation (187 +/- 60 vs 146 +/- 63 ng/ml). No significant differences were noted between donors receiving and not receiving vasopressors. In conclusion, the significant differences between pediatric and adult MEGX formation should be noted when establishing reference or normal ranges for this diagnostic test. Furthermore, concomitant drug therapy may significantly alter MEGX formation.

Pharmacokinetics of zidovudine phosphorylation in patients infected with the human immunodeficiency virus.

The relationship between zidovudine phosphorylation inside mononuclear cells and plasma zidovudine pharmacokinetic was assessed in six subjects. Plasma and intracellular concentrations were measured by radioimmunoassay over an 8-h period after administration of 100 or 200 mg of zidovudine. Plasma pharmacokinetics followed expected patterns, with considerable interpatient variability in area under the concentration-versus-time curve (AUC), and a terminal half-life of 1.5 h. Intracellular AUC was even more variable than plasma AUC, but the data suggested a crude linear relationship between these parameters. The intracellular half-life of 3.5 h was consistently longer than the plasma half-life, and varied little between patients. The prolonged intracellular half-life suggested that total phosphorylated zidovudine, as measured by the method described, is not greatly dominated by the 5'-monophosphate as predicted from the in vitro studies reported in the literature. Plasma concentrations of zidovudine have shown little correlation with clinical effect. Study of the relationship between phosphorylated zidovudine and clinical outcome could lead to a more effective management of therapy.

The medical heritage concept: a model for assuring comparable laboratory results in long-term longitudinal studies.

The success of a three or four decade health monitoring program depends upon the constancy and stability of the laboratory measurement data. This paper describes a prospective model which has been implemented as part of a 30-year health care program developed for the benefit of a particular group of residents who have lived for more than two years within five miles of an industrial plant which processed uranium metal and contaminated the surrounding area. Effective long-term laboratory monitoring (20 to 40 years) requires (1) a stable analytical base to establish comparability of all data collected, (2) innovative computer based record keeping, and (3) two selected reference populations which reflect method bias and widespread population change bias. To meet this need for comparability, the long-term Medical Heritage comparability concept was developed. This is an approach to the determination, storage, and retrieval of the laboratory data obtained on each specific participant in such a manner that all of that participant's data are internally comparable and continually traceable to definitive and/or reference methods developed by the National Institute for Standards and Technology and the Centers for Disease Control and the National Committee for Clinical Laboratory Standards. Systematic long-term documentation differentiates the Medical Heritage 30 to 40 year concept from the current short-term two- to three-year data continuity systems. As a model for long-term medical monitoring, the Medical Heritage comparability concept gives validity to individual patient care monitoring decisions.

Electrochemical determination of azidothymidine in human whole blood.

An electrochemical method based on differential pulse voltammetry is presented for the determination of AZT in whole blood of fasted subjects. A protein-free supernatant of whole blood is prepared using HClO4 precipitation followed by neutralization with phosphate buffer. The AZT is reduced at a hanging mercury drop electrode. The linear dynamic range of standards in buffer is from the detection limit of 4.1 nM to 206.5 microM (1.1 to 55,200 ng/ml). However, in spiked blood samples the linear dynamic range is from 0.029 to 0.29 microM (7.75 to 77.5 ng/ml). The whole blood assay yields a recovery of 92.30 +/- 5.92% compared to the standard solution assay. After a 30-min preparation time, each sample can be analyzed in 10 min by a manual procedure.

Regulation of the human immune response to ragweed pollen by immunotherapy. A controlled trial comparing the effect of immunosuppressive peptic fragments of short ragweed with standard treatment.

A new allergenic preparation consisting of peptic fragments of short ragweed has been tested for its clinical effectiveness. Such enzymatically derived fragments have been shown in prior murine studies to retain the T epitopes of the original allergen but to have a severe reduction in the number of B epitopes. Three groups of ragweed hayfever patients were placed on pre-seasonal immunotherapy. One group received a conventional ragweed preparation that had been enriched for antigen E (Amb a I), designated as Pool 2. The second group was given fragments of Pool 2 (fSRW) prepared by peptic digestion and the third group was injected with histamine as a placebo. Groups treated with the fSRW and Pool 2 had significantly reduced symptom-medication scores compared with the placebo-treatment group. However, fSRW-treated patients fared significantly better than Pool 2 patients (P less than 0.02). fSRW injections caused a significant rise in preseasonal specific IgG, antibodies as well as suppression of the seasonal anamnestic specific IgE increase. Similar, but not quite as marked changes occurred with Pool 2 treatment. fSRW was well tolerated and non-toxic. Thus, allergen modification by enzymatic degradation, as demonstrated here, appears to be a promising new approach for allergen immunotherapy.

Concentrations of phosphorylated zidovudine (ZDV) in patient leukocytes do not correlate with ZDV dose or plasma concentrations.

Zidovudine (ZDV) elicits its antiviral effect through intracellular metabolism to the 5'-triphosphate, which interferes with viral replication. Monitoring of the active metabolites of ZDV in cells could lead to an intracellular therapeutic range. This study was performed to determine whether a radioimmunoassay, previously used for in vitro quantitation of total phosphorylated ZDV inside peripheral blood leukocytes, could be used for similar determinations in patient samples. The relationship between ZDV dose, plasma concentrations, and intracellular metabolite concentrations was also examined. Ten-milliliter blood samples were drawn from each of 13 human immunodeficiency virus-infected patients and were assayed. Intracellular concentrations of phosphorylated ZDV ranged from 0.33 to 3.54 pmol/10(6) cells, similar to those observed in vitro. Phosphorylated ZDV was independent of dose, and did not correlate with plasma concentrations. Intracellular concentration in the patient population as a whole did not change during the 4-h dosing interval, while plasma concentration decayed normally. Later determinations in the same patients gave intracellular values within 31% of earlier values. Intraassay variability was less than 10%. Thus, the method is valid for measurement of phosphorylated ZDV in patient cells. Although individual concentrations showed no clear change during the 3-month study period, intracellular concentrations decreased with increasing length of therapy (up to 3 years) in the population as a whole. This suggests a decreased cellular ability to phosphorylate ZDV after prolonged exposure to drug. The lack of intracellular decay implies a half-life longer than the 1-h half-life of plasma ZDV. These data suggest that smaller doses or longer dosing intervals might maintain intracellular concentrations once steady state is achieved.(ABSTRACT TRUNCATED AT 250 WORDS)

Cyclosporine dose reduction by ketoconazole administration in renal transplant recipients.

Cyclosporine metabolism occurs in the liver via hepatic cytochrome P-450 microsomal enzymes. Ketoconazole, an imidazole derivative, has been shown to inhibit the cytochrome P-450 enzyme system. Thirty-six renal transplant recipients receiving cyclosporine as part of a triple immunosuppressive drug regimen were started on 200 mg/day of oral ketoconazole. The dose of cyclosporine was reduced by 70% at the start of ketoconazole; this dose reduction was based on our previous experience with concomitant cyclosporine-ketoconazole therapy. Ketoconazole was started in patients who had been on cyclosporine for between 10 days and 74 months. The mean cyclosporine dose was 420 mg/day (5.9 mg/kg/day) before starting ketoconazole and 66 mg/day (0.9 mg/kg/day) one year after the addition of ketoconazole; this represents a cyclosporine dose reduction of 84.7% (P less than 0.0001). The mean trough whole-blood cyclosporine concentrations measured by HPLC, were 130 ng/mL preketoconazole and 149 ng/mL after 1 year of combination therapy. Mean serum creatinine and BUN levels were unchanged before and during ketoconazole administration, and no changes in liver function tests were noted. Cyclosporine pharmacokinetics were performed before and after at least three weeks of ketoconazole. Hourly whole-blood samples were measured by HPLC (parent cyclosporine only) and TDX (parent + metabolites). Combination therapy resulted in decreases in the maximum blood concentration and the steady-state volume of distribution divided by the fractional absorption, and increases in mean residence time and the parent-to-parent plus metabolite ratio (calculated by dividing the HPLC by the TDX value). The addition of ketoconazole to cyclosporine-treated patients resulted in a significant inhibition of cyclosporine metabolism and decrease in the dosage. There was minimal nephrotoxicity, and only four rejection episodes occurred on combined therapy. The concomitant administration of the two drugs was well tolerated, and there was no deleterious effect on the immunosuppressive activity of cyclosporine. This drug interaction provides a significant reduction in the costs associated with organ transplantation.