Martha E. Lyon

Celiac Disease and IgA Deficiency: Complications of Serological Testing Approaches Encountered in the Clinic

BACKGROUND IgA deficiency causes false-negative IgA-based celiac serology results in patients with celiac disease. Using a case-finding strategy, we examined the prevalence of IgA deficiency, physician evaluation, and management of IgA deficiency during serological testing for celiac disease. METHODS We reviewed consecutive IgA-endomysial antibody (EMA) and serum IgA results from the laboratory database over 17 months. We cross-referenced seronegative patients with IgA deficiency (IgA <0.06 g/L) to the pathology database to evaluate intestinal biopsy results. Ordering physicians received a questionnaire regarding the management of seronegative patients with IgA deficiency who had no biopsy record. RESULTS Among the 9533 patients tested for IgA-EMA, 4698 (49%) were tested for IgA deficiency. IgA deficiency occurred in 35 of 4698 (0.75%) patients screened for IgA deficiency. Only 19 of 35 (54%) IgA-deficient patients were diagnosed appropriately with either intestinal biopsy (17 patients) or measurement of IgG-tissue transglutaminase (2 patients). Thirteen (76%) of the 17 IgA-deficient patients who underwent upper endoscopy with or without colonoscopy displayed gastrointestinal pathology on biopsies, including 3 (18%) with celiac disease. No further evaluation to exclude celiac disease was performed for the remaining 16 of 35 (46%) IgA-deficient, EMA-negative patients because of inappropriate management (6 patients), administrative error (7 patients), or patient/physician refusal (3 patients). CONCLUSIONS IgA deficiency occurred in 1:131 patients tested for celiac disease, and celiac disease occurred in 1:6 of those properly evaluated. Inadequate evaluation of IgA deficiency while testing for celiac disease occurred frequently and resulted in the underdiagnosis of both. Changes in testing algorithms and reporting of results were made to improve testing for celiac disease and IgA deficiency.

Simulation of repetitive diagnostic blood loss and onset of iatrogenic anemia in critical care patients with a mathematical model

Anemia is prevalent among critical care patients and is attributed to pathologic and iatrogenic processes and bleeding. The extent that diagnostic blood loss contributes to anemia among adult critical care patients is controversial and multi-factorial. The aim of this study is to describe an erythrokinetic model that integrates rates of phlebotomy, erythropoiesis and red cell senescence with patient characteristics to predict the onset of iatrogenic anemia in an objective manner. Using sex-specific parameters, the model predicts that adults with (average body weight and average blood volume), initial hemoglobin concentration at mid-reference interval, active erythropoiesis and losing 53 mL of blood per day by phlebotomy would require 40-70 days to reach 70 g/L of hemoglobin. To mimic critical care patients with low initial hemoglobin and suppressed erythropoiesis, the influence of daily blood loss and total blood volume was predicted. Simultaneous lack of erythropoiesis, initial hemoglobin concentrations at the lower limit of the reference interval (110 g/L), low body weight and increased phlebotomy accelerated onset of ~70 g/L hemoglobin transfusion threshold to 9-14 days. This computer simulation depicts the extent that adult critical care patients with anemia risk factors could benefit from conservative test ordering practices and subsequent reduced diagnostic phlebotomy.

Impact of a Transcutaneous Bilirubinometry Program on Resource Utilization and Severe Hyperbilirubinemia

Objectives: Our goal was to assess the impact of programmatic and coordinated use of transcutaneous bilirubinometry (TcB) on the incidence of severe neonatal hyperbilirubinemia and measures of laboratory, hospital, and nursing resource utilization. Methods: We compared the neonatal hyperbilirubinemia-related outcomes of 14 796 prospectively enrolled healthy infants ≥35 weeks gestation offered routine TcB measurements in both hospital and community settings by using locally validated nomograms relative to a historical cohort of 14 112 infants assessed by visual inspection alone. Results: There was a 54.9% reduction (odds ratio [OR]: 2.219 [95% confidence interval (CI): 1.543–3.193]; P < .0001) in the incidence of severe total serum bilirubin values (≥342 µmol/L; ≥20 mg/dL) after implementation of routine TcB measurements. TcB implementation was associated with reductions in the overall incidence of total serum bilirubin draws (134.4 vs 103.6 draws per 1000 live births, OR: 1.332 [95% CI: 1.226–1.446]; P < .0001) and overall phototherapy rate (5.27% vs 4.30%, OR: 1.241 [95% CI: 1.122–1.374]; P < .0001), a reduced age at readmission for phototherapy (104.3 ± 52.1 vs 88.9 ± 70.5 hours, P < .005), and duration of phototherapy readmission (24.8 ± 13.6 vs 23.2 ± 9.8 hours, P < .05). There were earlier (P < .01) and more frequent contacts with public health nurses (1.33 vs 1.66, P < .01) after introduction of the TcB program. Conclusions: Integration of routine hospital and community TcB screening within a comprehensive public health nurse newborn follow-up program is associated with significant improvements in resource utilization and patient safety.

Interference studies with two hospital-grade and two home-grade glucose meters.

BACKGROUND Interference studies of four glucose meters (Nova Biomedical [Waltham, MA] StatStrip [hospital grade], Roche Diagnostics [Indianapolis, IN] Accu-Chek Aviva [home grade], Abbott Diabetes Care [Alameda, CA] Precision FreeStyle Freedom [home grade], and LifeScan [Milpitas, CA] SureStep Flexx [hospital grade]) were evaluated and compared to the clinical laboratory plasma hexokinase reference method (Roche Hitachi 912 chemistry analyzer). These meters were chosen to reflect the continuum of care from hospital to home grade meters commonly seen in North America. METHODS Within-run precision was determined using a freshly prepared whole blood sample spiked with concentrated glucose to give three glucose concentrations. Day-to-day precision was evaluated using aqueous control materials supplied by each vendor. Common interferences, including hematocrit, maltose, and ascorbate, were tested alone and in combination with one another on each of the four glucose testing devices at three blood glucose concentrations. RESULTS Within-run precision for all glucose meters was <5% except for the FreeStyle (up to 7.6%). Between-day precision was <6% for all glucose meters. Ascorbate caused differences (percentage change from a sample without added interfering substances) of >5% with pyrroloquinolinequinone (PQQ)-glucose dehydrogenase-based technologies (Aviva and Freestyle) and the glucose oxidase-based Flexx meter. Maltose strongly affected the PQQ-glucose dehydrogenase-based meter systems. When combinations of interferences (ascorbate, maltose, and hematocrit mixtures) were tested, the extent of the interference was up to 193% (Aviva), 179% (FreeStyle), 25.1% (Flexx), and 5.9% (StatStrip). The interference was most pronounced at low glucose (3.9-4.4 mmol/L). CONCLUSIONS All evaluated glucose meter systems demonstrated varying degrees of interference by hematocrit, ascorbate, and maltose mixtures. PQQ-glucose dehydrogenase-based technologies showed greater susceptibility than glucose oxidase-based systems. However, the modified glucose oxidase-based amperometric method (Nova StatStrip) was less affected in comparison with the glucose oxidase-based photometric method (LifeScan SureStep Flexx).

Impact of skin tone on the performance of a transcutaneous jaundice meter

Aim:  To evaluate the performance of the Konica Minolta/Air‐Shields® JM‐103 jaundice meter on the basis of infant skin tone during the early neonatal period.

Simulation Models of Misclassification Error for Single Thresholds of High-Sensitivity Cardiac Troponin I Due to Assay Bias and Imprecision

BACKGROUND Clinical outcome studies for cardiac troponins (cTn) are expensive and difficult to design owing to variation in patients, in the assays, and in the incidence of different types of myocardial infarction (MI). To overcome these difficulties, simulation models were used to estimate the rate of misclassification error for MI and risk prediction resulting from assay bias and imprecision. METHODS Finite mixture analysis of Abbott high-sensitivity cTnI (hs-cTnI) results at time 0 h in patients presenting early with acute coronary syndrome (ACS) symptoms to the emergency department (ED) [n = 145, Reducing the Time Interval for Identifying New Guideline (RING) study] allowed derivation of a simulation data set (n = 10000). hs-cTnI concentrations were modified by addition of bias or imprecision error. The percentage of all 10000 modified hs-cTnI results that were misclassified for MI at thresholds of 2, 5, 26.2, and 52 ng/L was determined by Monte Carlo simulation. Analyses were replicated with an all-comer emergency department (ED) population (n = 1137) ROMI (Optimum Troponin Cutoffs for ACS in the ED) study. RESULTS In the RING study, simulation at 26.2-ng/L (99th percentile) and 52-ng/L thresholds were affected by both bias ±2 ng/L and imprecision (10%-20%) and had misclassification rates of 0.4% to 0.6%. Simulations at the 2-ng/L and 5-ng/L thresholds were only affected by bias. Misclassification rates at bias of ±1 ng/L were 10% for the 2-ng/L threshold, and 5% for the 5-ng/L threshold. CONCLUSIONS Simulation models predicted that hs-cTnI results are seldom misclassified (<1% of patients) when interpretative thresholds are near or exceed the overall 99th percentile. However, simulation models also predicted that low hs-cTnI results, as recommended in guidelines, are prone to misclassification of 5%-10% of patients.

Evaluation of the ESPGHAN Celiac Guidelines in a North American Pediatric Population

OBJECTIVES:We retrospectively examined the performance of the tissue transglutaminase (TTG), endomysial antibody (EMA) tests, and the ESPGHAN (European Society of Paediatric Gastroenterology, Hepatology and Nutrition) nonbiopsy criteria in a pediatric population.METHODS:Consecutive celiac serologies and corresponding intestinal biopsy results were obtained on children <18 years old over 3.5 years. Patients were classified into three categories: positive TTG, negative TTG, and IgA deficiency.RESULTS:Of the 17,505 patients with celiac serology performed, 775 had a positive TTG, 574 with a negative TTG were biopsied, and 25 were IgA deficient. Of the patients with a TTG ≥10 × upper limit of normal (ULN), positive EMA, and symptoms, 98.2% had biopsies consistent with celiac disease (CD). Four human leukocyte antigen (HLA) DQ2/DQ8-positive patients who met the ESPGHAN nonbiopsy criteria did not have CD. In the group with a TTG 3–10 × ULN, 75.7% EMA-positive patients and only 40% EMA-negative patients had CD (P<0.001). Of those with a TTG 1–3 × ULN, 52.2% EMA-positive patients vs. only 13.3% EMA-negative patients had CD (P<0.01). Of the patients with bulbar and duodenal biopsies, 9.8% had CD confined only in the bulb, especially those with a low titer TTG (P<0.01). CD prevalence in our cohort was 34.6%. Sensitivity, specificity, and positive predictive value of the TTG were 98.7%, 86.4%, and 79.4%, respectively.CONCLUSIONS:The TTG is a very sensitive screen for CD, but positive predictive value improves with a positive EMA titer. To apply the new ESPGHAN guidelines, clinicians must understand the performance of their celiac serology tests.

Patient acuity exacerbates discrepancy between whole blood and plasma methods through error in molality to molarity conversion: "Mind the gap!".

OBJECTIVE A mathematical constant factor is proposed to convert measured whole blood glucose molality to plasma-equivalent molarity. The objective of this study was to determine the distributions of conversion factors for groups of patients with different acuity and to assess the gap or error in plasma-equivalent glucose reporting that would occur when a mathematical constant conversion factor is used in patients. METHODS Distributions of hematocrit, red blood cell water and plasma water were determined in patients from the community, hospital and adult intensive care unit. Volume displacement conversion factor distributions and glucose error were determined for each group. RESULTS With increasing patient acuity the median hematocrit decreased, median plasma water increased and variation of these parameters increased. In hospital patients, the molality to molarity conversion factor distribution interval was 1.04-1.16, rather than a constant 1.11. Assuming direct electrode glucose devices only have error attributed to analytical imprecision (coefficient of variation of 5%), it is predicted that only 2% of community patients will have glucose results that exceed 10% of the target values. In the same device, due to variance in hematocrit and plasma water affecting the factor for conversion of molality to molarity, it is predicted that 8.2% of adult intensive care unit patients would have glucose results that exceed 10% of the target value. CONCLUSIONS Changes in hematocrit and plasma water concentration are predicted to affect a gap or error between whole blood direct reading biosensors and central laboratory plasma methods. This error increases and becomes more variable as patient acuity increases.

Investigating Interferences of a Whole-Blood Point-of-Care Creatinine Analyzer: Comparison to Plasma Enzymatic and Definitive Creatinine Methods in an Acute-Care Setting

BACKGROUND Although measurement of whole-blood creatinine at the point of care offers rapid assessment of renal function, agreement of point-of-care (POC) results with central laboratory methods continues to be a concern. We assessed the influence of several potential interferents on POC whole-blood creatinine measurements. METHODS We compared POC creatinine (Nova StatSensor) measurements with plasma enzymatic (Roche Modular) and isotope dilution mass spectrometry (IDMS) assays in 119 hospital inpatients. We assessed assay interference by hematocrit, pH, pO(2), total and direct bilirubin, creatine, prescribed drugs, diagnosis, red blood cell water fraction, and plasma water fraction. RESULTS CVs for POC creatinine were 1.5- to 6-fold greater than those for plasma methods, in part due to meter-to-meter variation. Regressioncomparison of POC creatinine to IDMS results gave a standard error (S(y|x)) of 0.61 mg/dL (54 μmol/L), whereas regression of plasma enzymatic creatinine to IDMS was S(y|x) 0.16 mg/dL (14 μmol/L). By univariate analysis, bilirubin, creatine, drugs, pO(2), pH,plasma water fraction, and hematocrit were not found to contribute to method differences. However, multivariate analysis revealed that IDMS creatinine, red blood cell and plasma water fractions, and hematocrit explained 91.8% of variance in POC creatinine results. CONCLUSIONS These data suggest that whole-blood POC creatinine measurements should be used with caution. Negative interferences observed with these measurements could erroneously suggest adequate renal function near the decision threshold, particularly if estimated glomerular filtration rate is determined. Disparity between whole-blood and plasma matrices partially explains the discordance between whole-blood and plasma creatinine methods.

Estimates of Total Analytical Error in Consumer and Hospital Glucose Meters Contributed by Hematocrit, Maltose, and Ascorbate

Background: Patients and physicians expect accurate whole blood glucose monitoring even when patients are anemic, are undergoing peritoneal dialysis, or have slightly elevated ascorbate levels. The objective of this study was to estimate analytical error in two consumer and two hospital glucose meters contributed by variations in hematocrit, maltose, ascorbate, and imprecision. Method: The Influence of hematocrit (20–60%), maltose, and ascorbate were tested alone and in combination with each glucose meter and with a reference plasma glucose method at three concentrations of glucose. Precision was determined by consecutive analysis (n = 20) at three levels of glucose. Multivariate regression analysis was used to estimate the bias associated with the interferences, alone and in combination. Total analytical error was estimated as |% bias| + 1.96 (% imprecision). Results: Three meters demonstrated hematocrit bias that was dependent upon glucose concentration. Maltose had profound concentration-dependent positive bias on the consumer meters, and the extent of maltose bias was dependent on hematocrit. Ascorbate produced small but statistically significant biases on three meters. Coincident low hematocrit, presence of maltose, and presence of ascorbate increased the observed bias and was summarized by estimation of total analytical error. Among the four glucose meter devices assessed, estimates of total analytical error in glucose measurement ranged from 6 to 68% under the conditions tested. Conclusions: The susceptibility of glucose meters to clinically significant analytical biases is highly device-dependent, and low hematocrit exacerbated the observed analytical error.