Etretinate: effect of milk intake on absorption.

Since etretinate, an aromatic retinoid useful in the treatment of psoriasis and other skin disorders is lipid-soluble, it may be poorly absorbed in the absence of a fat load. We therefore studied serum concentrations of etretinate and its major metabolite (Ro 10-1670) after the controlled administration of etretinate. After an overnight fast, 6 Darier's disease and 4 psoriatic patients received a 1 mg/kg morning dose of etretinate with water or 1 pint of whole milk (fat load). There was a 260% increase (p less than 0.0005) in the mean of each patient's increase in the baseline-corrected peak serum concentration of etretinate after administration with milk (115 +/- 15 micrograms/dl) compared to after administration with water (32 +/- 4 micrograms/dl). Over a 24-h period there was an overall 296 +/- 26% (p less than 0.0005) increase in serum etretinate after administration with milk compared to water in 5 patients with Darier's disease. In contrast to the serum etretinate, there was a 17% mean decrease (p less than 0.025) in the corrected peak serum concentration of Ro 10-1670 in all 10 patients after administration of etretinate with milk compared to water. The net result of these alterations is that the mean corrected serum concentration of etretinate is higher than Ro 10-1670 at all time points measured after milk administration. In contrast, after administration of etretinate with water the major retinoid in the serum is Ro 10-1670. Establishing the clinical significance of these alterations may require controlled clinical trials.

Diagnosis and therapeutic monitoring of invasive candidiasis by rapid enzymatic detection of serum D-arabinitol.

BACKGROUND: Using a rapid automated enzymatic assay, we prospectively investigated serum D-arabinitol (DA), a biochemical marker of invasive candidiasis, in a large population of high-risk patients to determine its potential diagnostic, therapeutic, and prognostic significance in invasive candidiasis. PATIENTS AND METHODS: A total of 3,223 serum samples were collected from 274 patients with cancer. Serum DA concentrations were determined in coded serum samples analyzed by rapid enzymatic assay. Creatinine also was analyzed in the same system to determine a serum DA and creatinine ratio (DA/Cr). The sensitivity, specificity, correlation with therapeutic response, and prognostic significance were analyzed for all patient study groups. RESULTS: A DA/Cr of > or = 4.0 mumol/L per mg/dL was detected in 31 (74%) of all 42 cases of fungemia and 25 (83%) of the 30 cases of the subset of persistent fungemia. Elevated DA/Cr was detected in 4 (40%) of 10 patients with tissue-proven, deeply invasive candidiasis and negative blood cultures (eg, hepatosplenic candidiasis or localized abscess) and 7 (44%) of 16 cases of deep mucosal candidiasis (eg, esophageal candidiasis). Elevated serial DA/Cr levels also were detected in persistently febrile and granulocytopenic patients requiring empirical amphotericin B. Among 26 assessable cases of fungemia, abnormally elevated DA/Cr values were detected in 14 (54%) before, 10 (38%) after, and 2 (8%) simultaneously with the first microbiologic report of fungemia. The trends of serial DA/Cr values correlated with therapeutic response in 29 (85%) of 34 patients with assessable cases of fungemia, decreasing in 8 (89%) of 9 patients with clearance of fungemia and increasing in 21 (84%) of 25 patients with persistence of fungemia. Among the 34 assessable patients with fungemia, mortality was directly related to the trend of serial DA/Cr determinations over time: 71% among fungemic patients who had persistently elevated or increasing DA/Cr, and 18% among the fungemic patients who had resolving DA/Cr or never had elevated DA/Cr (P < 0.01). CONCLUSIONS: Rapid enzymatic detection of DA in serially collected serum samples from high-risk cancer patients permitted detection of invasive candidiasis, early recognition of fungemia, and therapeutic monitoring in DA-positive cases. Serially collected serum DA determinations complement blood cultures for improving detection and monitoring therapeutic response in patients at risk for invasive candidiasis.

Comparison of automated and manual methods for urinalysis.

The authors compared results for accuracy and precision obtained by a semiautomated prototype International Remote Imaging Systems, Inc. (IRIS) urinalysis workstation (IUW) with those from quantitative manual urinalysis (QMU). Three technologists skilled in urinalysis each performed 172 urinalyses with both the IUW and QMU methods. The results show that the IUW method is likely to yield comparable counts for particulate analytes compared with the QMU, except for casts. The QMU reported significantly (P less than 0.001) more casts than the IUW method. This difference is related to at least a ninefold greater volume of untreated urine examined by the QMU method than the IUW method. The IUW method may provide a more accurate result than the QMU method at very low and high concentrations of particulate analytes. The result from 24 blind duplicate urines also analyzed by each of the three technologists with both methods showed comparable precision for particulate analytes between the two methods except for red blood cells; the QMU method had significantly (P less than 0.001) better precision for this analyte.

Kinetic analysis of the rapid intraoperative parathyroid hormone assay in patients during operation for hyperparathyroidism.

BACKGROUND: Rapid intraoperative parathyroid hormone (RI-PTH) assay is used to guide adequacy of resection during operation for hyperparathyroidism. We compared the RI-PTH assay (15 minutes) with a standard PTH assay, determined whether the PTH half-life varied between patients, and constructed a kinetic analysis of the RI-PTH data. METHODS: Forty-five patients with hyperparathyroidism had blood sampled at baseline and at times after parathyroid resection. Intact PTH was determined using RI-PTH and a standard assay. Values were fitted to an exponential decay curve using the baseline and the follow-up time points. PTH half-life and the new postexcision baseline value were calculated from the decay curve. RESULTS: The RI-PTH assay and the standard PTH assay correlated well. Average PTH half-life was 1.68 +/- 0.94 minutes (0.42 to 3.81 minutes). A kinetic analysis yielded a formula for the generation of a PTH decay curve. Using a 50% reduction in RI-PTH at 5 minutes as the criterion for adequate resection, 2 patients were incorrectly classified as not being cured. These patients were correctly classified using the kinetic analysis. CONCLUSIONS: PTH half-life can vary substantially. A kinetic analysis may be more accurate in assessing adequacy of resection. This method allows the surgeon to interpret RI-PTH data independent of the timing of samples.

Etretinate. Persistent serum levels after long-term therapy.

In 47 patients who received long-term etretinate therapy, we measured serum etretinate concentrations from one to 244 weeks after the discontinuation of therapy. The earliest posttreatment, nondetectable serum concentration of etretinate was observed at five weeks after treatment. Detectable serum concentrations (0.05 to 1.2 micrograms/dL) were observed more than two years (108, 111, 131, 136, and 150 weeks) following the discontinuation of therapy. Sequential serum concentrations obtained on eight individual patients were used to determine half-lives for this late-phase elimination. The median half-life for the 12 curves obtained was 12.5 weeks (range, 5.3 to 24.8 weeks). Since etretinate is stored in fat, we compared each patient's deviation from ideal body weight as a measure of excess body fat with various pharmacokinetic factors of etretinate elimination. Overweight patients tended to have slower elimination, maintain higher serum concentrations, and clear etretinate later.

Intravenous fluorescein interference with clinical laboratory tests.

The results of laboratory tests performed after fluorescein angiography may be erroneous because of interference by intravenous fluorescein. We investigated this potential interference in four adults at intervals of five minutes, three hours, six hours, and 12 hours after fluorescein injection. We used a panel of serum and urine chemistry tests on seven commonly used instruments. A significant change in the reported concentration of a serum or urine analyte was defined as a result beyond +/- 3 coefficients of variation of the preinjection baseline value for the test on a specific instrument. The determinations of creatinine, total protein, cortisol, digoxin, quinidine, and thyroxine in serum were affected by intravenous fluorescein. The urine tests were unaltered. The physician must be aware of the problem of interpreting clinical chemistry results after fluorescein angiography.

Sonic fragility of the head-tail bond of bacteriophage P22.

The binding of tail parts to the head of phage P22 is normally irreversible, but after adsorption to host cells sonication releases many of the tail parts intact. This release is dependent on the power of sonication, but is independent of the length of sonication from 2 to 32 s. This phenomenon has been used to show that the upper limit of the number of P22 particles that can bind to a cell is no lower than 700.

Lithium determinations evaluated in eight analyzers.

We compared five ion-selective electrodes (ISEs; AVL 985S, Baxter Lytening 2Z, Beckman EL-ISE, Ciba-Corning 654 Na/K/Li, and Nova CRT 11) and a colorimetric method (Ektachem) for determination of lithium with flame atomic absorption and atomic emission spectroscopy. We evaluated precision, recovery, interference by drugs (procainamide, N-acetylprocainamide, quinidine, lidocaine, carbamazepine, and valproic acid) and inorganic analytes (Na, K, Ca, Mg, and Br), and performance with sera from patients receiving lithium. Imprecision was < 5% (CV) for all analyzers except Ektachem and Beckman. Analytical recovery was 100% +/- 5% for all analyzers except Ektachem and Baxter. Some drug interference was seen with all analyzers except AVL and Corning. Calcium caused interference with AVL, Corning, and Ektachem analyzers, and sodium and potassium interfered with the Ektachem analyzer. The results with the Baxter, Beckman, and Corning analyzers were closest to those by flame atomic emission (mean differences, 0.01 to 0.02 mmol/L); the AVL and Nova analyzers gave lower results (mean differences, -0.11 to -0.13 mmol/L) and the Ektachem gave higher results (mean difference, 0.08 mmol/L).

Enzyme abnormalities of patients with acquired immunodeficiency syndrome.

We determined the enzyme activities of glucose-6-phosphate isomerase, alanine aminotransferase, aspartate aminotransferase, and lactate dehydrogenase in serum from 23 normal controls, 27 anti-HIV seropositive individuals confirmed by Western blot, and 53 patients with acquired immunodeficiency syndrome (AIDS). There is a significant difference for all four enzyme activities among controls, HIV seropositive individuals, and patients with AIDS, the enzyme activities showing a progressive increase as the disease progresses. Evidently these enzyme measurements may be adjunctive biochemical markers for progression of AIDS.

Nutritional status of patients with acquired immunodeficiency syndrome.

Nutritional status may be an important factor in the prognosis of morbidity and mortality. We assessed the nutritional status of individuals seropositive for human immunodeficiency virus (HIV) (as confirmed by Western blot) and of patients with AIDS, by determining the concentration in serum of total protein, albumin, prealbumin (transthyretin), and retinol-binding protein. HIV-seropositive individuals showed no significant difference from normal volunteers in values for prealbumin, albumin, and retinol-binding protein. Patients with AIDS showed significantly smaller prealbumin and albumin concentrations than did normal and HIV-positive individuals. There was no significant difference in the concentration of retinol-binding protein among the three groups. The concentration of total serum protein was significantly greater in HIV-positive individuals and in patients with AIDS than in normal individuals. Thus, the nutritional status of patients with AIDS may be a factor for morbidity and mortality.

Analytical bias for cholesterol and the percent of the population deemed at risk for coronary heart disease.

The effect of methodological bias on the population at risk is dependent on the location of the reference value in the distribution of the population. We fitted the cumulative distribution for cholesterol to a rational function and calculated the apparent reference values for four biased methods: Technicon SMAC (2.6%), DuPont aca (4.0 to 4.8%), Kodak DT-60 (-2.0 to -5.5%), and BMD Reflotron (-7.4 to -7.8%). With the true and apparent reference values for cholesterol and the rational function, we determined the percentage increase or decrease in the population deemed at risk for coronary heart disease. The population at risk increased by as much as 48% for methods with positive bias, and decreased by as much as 54% for methods with negative bias. If we restrict the percentage of the population incorrectly diagnosed to 3% and use reference values (cut points) recommended by the National Cholesterol Education Program, the maximum allowable methodological bias would be 1.6% for positive bias and -1.55% for negative bias. Therefore, an absolute methodological bias of 3% (as recommended by the Laboratory Standardization Panel) may be too liberal.

A model for the effect of bias for cholesterol on the population at risk.

The location of the Reference Value for an analyte within the population distribution affects the magnitude of error due to methodological bias. Using the gaussian distribution, we evaluated the effects of systematic and proportional biases of the method (positive and negative), mean value, and standard deviation on the magnitude of error. We chose four Reference Values for cholesterol as a model. For a population with a mean of 2.0 and SD of 0.36 g of cholesterol per liter, a 3% positive proportional bias causes sixfold more error at the 50th percentile than at the 97.5th. In general, the error for a given bias (proportional or systematic) is greater for a Reference Value within the body than at the tails of the distribution. Further, the magnitude of the error varies as a function of the mean and standard deviation of the population.

On the albumin-dependence of measurements of free thyroxin. II. Patients with non-thyroidal illness.

We studied the relation between thyroxin-binding proteins and free thyroxin (FT4) measurements by five radioimmunoassays (RIA) and an FT4 index (FT4I) in patients with non-thyroidal illness (NTI). The one-step FT4 RIAs and the FT4I frequently failed to identify the true FT4 status (as determined by equilibrium dialysis) of NTI patients. In these patients, falsely low FT4 results with one-step RIAs and FT4I were associated with decreasing total T3 and T4 concentrations, which, furthermore, paralleled decreasing serum albumin concentrations. All NTI patients with "low T3, low T4 syndrome" had subnormal albumin concentration. The two-step RIAs and equilibrium dialysis showed normal FT4 concentrations in most patients with NTI. However, sera from a subset of NTI patients with "low T3 syndrome" gave above-normal FT4 results with these methods. From their predictably poor performance in the presence of a subnormal albumin concentration, we conclude that the one-step FT4 RIAs and FT4I are inappropriate for testing the thyrometabolic status of NTI patients.

Positive interference in lithium determinations from clot activator in collection container.

We describe positive interference with the ion-selective electrode determination of lithium (Lytening 2Z analyzer; Dade) when blood is collected in a 10-mL plain red-top plastic Vacutainer Plus Tube (Becton Dickinson) containing a silica clot activator and silicone surfactant (prod. no. 36-7820). We evaluated both the original tube (blue-labeled) and a new tube formulated to contain less silicone surfactant (striped-labeled). We determined that the interference is from either the silica clot activator or the silicone surfactant used to fix the silica to the tube and is inversely related to the volume of blood in the tube. Long-term intermittent exposure of the Li ion-selective electrode to the silica clot activator or surfactant results in decreased Li values--in terms of both the positive interference by the silica clot activator or surfactant and the actual Li determinations. Moreover, this long-term interference with the Li ion-selective electrode for patient's specimens is undetected by the Dade control material (QCLytes).

Effects of specimen turbidity and glycerol concentration on nine enzymatic methods for triglyceride determination.

We compared the effects of specimen turbidity and glycerol concentration on nine enzymatic methods for triglyceride measurement. We assayed 51 specimens with triglyceride concentrations of 0.85-8.21 mmol/L (75-727 mg/dL) and turbidity at 420 nm equivalent to > or = 0.1 mmol/L (8.8 mg/dL) triglyceride (measured as part of our comparison method). The data were analyzed by multiple regression, which gave coefficients for the effects of glycerol concentration and the change in turbidity during the reaction. The effects of specimen turbidity and glycerol concentration were method-dependent and ranged from 6.20% to -15.67% of the measured result. The magnitude of the turbidity effect (in assays with a significant turbidity interference) was similar to that for glycerol (in assays with a significant glycerol interference). A triglyceride assay with a bichromatic measurement was less subject to interference from turbidity.

On the albumin dependence of measurements of free thyroxin. I. Technical performance of seven methods.

We evaluated three one-step (analog) and two two-step radioimmunoassay for free thyroxin (FT4), and a FT4 index calculated from the total T4 (TT4) and thyroxin-binding globulin (TBG) ratio for technical performance, for correlation with the reference FT4 method (equilibrium dialysis), and for dependence on TBG and albumin concentrations. The one-step methods (Amerlex, Coat-A-Count, and GammaCoat) showed greater precision than the two-step procedures (GammaCoat and Spiria). Results by the latter two techniques, however, correlated better with those by equilibrium dialysis than did those by the analog methods or by TT4/TBG. Only the GammaCoat two-step method had a slight but statistically significant (inverse) correlation with TBG concentration. All three analog methods and TT4/TBG showed a marked dependence on albumin concentration, whereas the Spiria technique showed only a slight dependence. Only equilibrium dialysis was independent of both the TBG and albumin concentration. Thus, despite their good precision, the analog (one-step) FT4 methods and the TT4/TBG approach cannot be expected to produce valid results when the concentration of albumin in serum is abnormally low or high.

A model for assessing interference.

Most studies of analytical interference indicate the magnitude but not the type of interference. We developed a model for interference that assesses the magnitude of the following types of interference: independent of analyte concentration, dependent on analyte concentration, and a combination of the two. The experimental design for the model is an orthogonally arranged matrix with progressively increasing concentrations of analyte and interfering agent. Multiple regression with these independent variables (concentration of analyte, concentration of interfering agent, and the product of the two) is used to determine the magnitude, direction, and significance of each type of interference. Applying the model to study interference by hemoglobin with determinations of creatinine, direct bilirubin, and total bilirubin showed that hemoglobin interferes with determination of creatinine independent of the analyte concentration, interferes with total bilirubin dependent on the analyte concentration, and interferes with direct bilirubin by a combination of these two.

An enzymatic method for 5-fluorocytosine.

We developed an enzymatic method for determination of 5-fluorocytosine in serum, using creatine iminohydrolase (EC 3.5.4.21), the Cobas-Bio analyzer, and an extant ammonia method. Analytical recovery (y) of drug added to serum (x) was good, with y = 0.97x-0.7, Sy.x = 3.6, r = 0.997 (n = 65) over the range 6.25 to 150 mg/L. Comparison with an HPLC method (x) showed good agreement: y = 0.98x + 1.34, Sy.x = 3.7 (n = 37), as analyzed with the Deming debiased regression. Precision was good, CVs being less than 3% for within-run and less than 6% for between-run controls. Ammonia, amphotericin B, glucose, urea, and hemolysis do not interfere, but bilirubin shows analyte-dependent interference and lipemia interferes when triglycerides exceed 5 g/L. This assay is accurate, inexpensive, and easy to perform. It can be easily adapted for routine or emergency use.

Direct and indirect techniques for free thyroxin compared in patients with nonthyroidal illness. II. Effect of prealbumin, albumin, and thyroxin-binding globulin.

We studied the correlation of thyroxin (T4)-binding proteins with the apparent free T4 (FT4) in 101 patients with nonthyroidal illness (NTI). Most patients (95%) were seriously ill at the time of blood collection. Concentrations of T4-binding prealbumin (transthyretin), albumin, and T4-binding globulin (TBG) often were low in the sera of these patients. Albumin was the most frequently subnormal, TBG the least. FT4 in serum was determined by five methods represented in 16 different assays. With few exceptions, analog (one-step) FT4 RIAs--both the binding-rate-based RIA and the related FT4 indices (calculated from triiodothyronine-macroaggregated albumin uptake and total T4)--and T4/TBG ratios correlated positively and usually highly significantly (P less than 0.01) with concentrations of prealbumin, albumin, and TBG. Equilibrium dialysis values for FT4 did not correlate with prealbumin concentrations but showed a weakly (P less than 0.03) positive association with albumin and a highly significant (P less than 0.002) positive correlation with TBG. Of the three two-step FT4 RIAs tested, the only statistically significant but weakly (P less than 0.02) positive correlation with T4-binding proteins was between Spiria FT4 and TBG. Thus, in these NTI patients, FT4 estimates vary with methodology and, to a lesser extent, with the particular assay used. The results from two-step FT4 RIAs are least associated with binding protein concentrations.