Utilization of assay performance characteristics to estimate hemoglobin A1c result reliability.

BACKGROUND: Allowable total error (TE(a)) goals for hemoglobin (Hb) A(1c) require minimal assay imprecision and bias and implementation of a robust QC monitoring program. Here, we compare the combined influence on the risk of reporting unreliable results of TE(a) goals, a routine QC practice, and assay performance characteristics of 6 Hb A(1c) instruments across 4 academic medical centers. METHODS: The CLSI protocols EP-5 and EP-9 were applied to investigate Hb A(1c) result imprecision and bias on the Variant II Turbo and Variant II (Bio-Rad), G8 (Tosoh), Capillarys 2 Flex Piercing (Sebia), COBAS Integra 800 (Roche), and DCA Vantage (Siemens). Patient-weighted σ values and the risk of reporting unreliable Hb A(1c) results were determined for each assay at TE(a) specifications of 5%, 6%, and 7%. RESULTS: A large range of patient-weighted σ values spanning 0.5 orders of magnitude at a 6% TE(a) was observed. Although imprecision for all instruments was <3%, bias impacted the majority of the σ changes observed. Estimates for reporting unreliable results varied almost 500-fold based on analytical performance alone. CONCLUSIONS: Considerable differences in the probability of reporting unreliable Hb A(1c) results between different NGSP (formerly the National Glycohemoglobin Standardization Program)-certified platforms were observed. At a 6% TE(a), our study indicates all but the Capillarys 2 Flex Piercing requires that the maximum affordable QC be run. Risk estimates for individual laboratories' Hb A(1c) methods can be used to assess QC practices and residual risk of an unreliable Hb A(1c) result.

Should I repeat my 1:2s QC rejection?

BACKGROUND: Repeating a QC that is outside 2SD from the mean (1:2s rule) appears to be a common practice. Although this form of repeat sampling is frowned on by many, the comparative power of the approach has not been formally evaluated. METHODS: We computed power functions mathematically and by computer simulation for 4 different 1:2s repeat-sampling strategies, as well as the 1:2s rule, the 1:3s rule, and 2 common QC multirules. RESULTS: The false-rejection rates for the repeat-sampling strategies were similarly low to those of the 1:3s QC rule. The error detection rates for the repeat-sampling strategies approached those of the 1:2s QC rule for moderate to large out-of-control error conditions. In most cases, the power of the repeat-sampling strategies was superior to the power of the QC multirules we evaluated. The increase in QC utilization rate ranged from 4% to 13% for the repeat-sampling strategies investigated. CONCLUSIONS: The repeat-sampling strategies provide an effective tactic to take advantage of the desirable properties of both the 1:2s and 1:3s QC rules. Additionally, the power of the repeat-sampling strategies compares favorably with the power of 2 common QC multirules. These improvements come with a modest increase in the average number of controls tested.

Synaptic ultrastructural alterations anticipate the development of neuroaxonal dystrophy in sympathetic ganglia of aged and diabetic mice.

Neuroaxonal dystrophy, a distinctive axonopathy characterized by marked enlargement of distal axons, is the hallmark pathologic alteration in aged and diabetic human prevertebral sympathetic ganglia and in corresponding rodent models. Neuroaxonal dystrophy is thought to represent the abnormal outcome of cycles of synaptic degeneration and regeneration; a systematic study of identified axon terminals in aged and diabetic prevertebral ganglia, however, has not previously been performed. We examined the initial changes that develop in presynaptic and postsynaptic elements in sympathetic ganglia of aged and diabetic mice and found numerous synaptic changes involving both presynaptic and postsynaptic elements. Early alterations in presynaptic axon terminal size, vesicle content, and morphology culminate in the development of anastomosing membranous tubulovesicular aggregates, accumulation of autophagosomes, and amorphous debris that form a continuum with progressively larger classically dystrophic swellings. Dendritic changes consist of the development of swellings composed of delicate tubulovesicular elements and mitochondriopathy characterized by increased numbers of small mitochondria and, exclusively in aged ganglia, megamitochondria. These results support the hypothesis that neuroaxonal dystrophy results from progressive changes in presynaptic axon terminals that likely involve membrane dynamics and which are accompanied by distinctive changes in postsynaptic dendritic elements.

Assessing the impact of the frequency of quality control testing on the quality of reported patient results.

BACKGROUND: The traditional measure used to evaluate QC performance is the probability of rejecting an analytical run that contains a critical out-of-control error condition. The probability of rejecting an analytical run, however, is not affected by changes in QC-testing frequency. A different performance measure is necessary to assess the impact of the frequency of QC testing. METHODS: I used a statistical model to define in-control and out-of-control processes, laboratory testing modes, and quality control strategies. RESULTS: The expected increase in the number of unacceptable patient results reported during the presence of an undetected out-of-control error condition is a performance measure that is affected by changes in QC-testing frequency. I derived this measure for different out-of-control error conditions and laboratory testing modes and showed that a worst-case expected increase in the number of unacceptable patient results reported can be estimated. The laboratory thus has the ability to design QC strategies that limit the expected number of unacceptable patient results reported. CONCLUSIONS: To assess the impact of the frequency of QC testing on QC performance, it is necessary to move beyond thinking in terms of the probability of accepting or rejecting analytical runs. A performance measure based on the expected increase in the number of unacceptable patient results reported has the dual advantage of objectively assessing the impact of changes in QC-testing frequency and putting focus on the quality of reported patient results rather than the quality of laboratory batches.

Clinical and economic impact of falsely decreased calcium values caused by gadoversetamide interference.

OBJECTIVE: Gadolinium is administered as a contrast agent in MRI procedures. Two gadolinium-based contrast agents, gadodiamide and gadoversetamide, interfere with colorimetric total serum calcium methods. The purpose of this prospective observational study was to examine the incidence of calcium interference after gadoversetamide procedures, associated clinical outcomes, and costs 20 months after implementation of quality assurance and physician education programs. MATERIALS AND METHODS: Records of patients who received gadoversetamide from June 24, 2006, to October 7, 2006, were reviewed to determine if a routine calcium test had been performed after the injection. Calcium values were repeated with an alternate method that is less susceptible to gadoversetamide interference. If the difference was > or = 2.0 mg/dL or if the initial test value was < or = 7.0 mg/dL, patient charts were reviewed for any related treatment. Costs associated with this algorithm were tracked. RESULTS: The initial calcium test was performed after gadoversetamide in 766 of 3,439 instances. The alternate test was performed in 633 of 766. One hundred twenty-five of 633 (20%) showed a difference in calcium values that was > or = 0.7 mg/dL, with 16 showing differences of > or = 1.6 mg/dL. Chart review for 56 instances revealed that calcium supplements were administered in 22 of 56 around the time of gadoversetamide injection. However, none appeared to be related to the spurious hypocalcemia. The total additional cost (reagent and technologist) for following this algorithm for just over 3 months was $6,807. CONCLUSION: Approximately 20% of patients receiving gadoversetamide exhibited spurious hypocalcemia. No patients were identified who received inappropriate calcium because of this interference. This may be attributable to the quality assurance and physician education programs.

The impact of QC frequency on patient results.

The expected number of unacceptable patient results due to an undetected, malfunction--E(Nu)--can be set as a patient-based quality goal. Using the number of patients tested between QC specimens as a design parameter allows one to design QC strategies that meet specified patient-based quality goals. The QC utilization rate can be minimized in a QC design for a given E(Nu). The QC-utilization rate achievable depends on how close analytical imprecision is to the total allowable error.

Assessing Quality Control Strategies for HbA1c Measurements From a Patient Risk Perspective.

BACKGROUND: Current laboratory risk management principles emphasize the importance of assessing laboratory quality control (QC) practices in terms of the risk of patient harm. Limited practical guidance or examples on how to do this are available. METHODS: The patient risk model described in a published laboratory risk management guideline was combined with a recently reported approach to computing the predicted probability of patient harm to produce a risk management index (RMI) that compares the predicted probability of patient harm for a QC strategy to the acceptable probability of patient harm based on the expected severity of harm caused by an erroneously reported patient result. RESULTS: Measurement procedure capability and quality control performance for two instruments measuring HbA1c in a laboratory were assessed by computing the RMI for each instrument individually and for the laboratory as a whole. CONCLUSIONS: This assessment provides a concrete example of how laboratory QC practices can be directly correlated to the risk of patient harm from erroneously reported patient results.

Statistical topics in the laboratory sciences.

This chapter concerns statistical concepts and procedures that are applicable to diagnostic testing performed in the clinical laboratory. Three important laboratory issues are addressed: the estimation of analytical imprecision, the design of an effective laboratory quality control strategy, and the establishment of population reference ranges. These three topics were selected because each demonstrates a valuable statistical principle. Estimation of analytical imprecision highlights the important role of study design. Evaluating laboratory quality control strategies emphasizes the importance of choosing appropriate statistical models. The estimation of population reference ranges demonstrates that there can be many different approaches to developing good statistical estimators.

A direct comparison between lamellar body counts and fluorescent polarization methods for predicting respiratory distress syndrome.

Our objective was to directly compare the diagnostic usefulness of lamellar body counting (LBC) and the TDx-FLM II assay (Abbott Laboratories, Abbott Park, IL) for predicting respiratory distress syndrome (RDS). This was a 5-year, retrospective, cohort study. A diagnosis of RDS was given to infants who received surfactant treatment and/or required ventilator support and/or continuous positive airway pressure for more than 24 hours. There were 172 infants without RDS and 12 with RDS included in the study. By using a TDx-FLM II cutoff of 55 mg/g or more for maturity, the sensitivity was 83%, specificity was 65%, predictive value of a mature result was 98%, and predictive value of an immature result was 14%. These results were similar to LBC using a cutoff of 50,000/microL or more with sensitivity of 92%, a specificity of 60%, a predictive value of a mature result of 99%, and a predictive value of an immature result of 14%. The LBC and TDx-FLM II methods have similar clinical usefulness.

Evaluation of polymerase chain reaction for group B streptococcus detection using an improved culture method.

OBJECTIVE: The administration of antibiotic prophylaxis to laboring women who harbor Group B streptococci (GBS) depends on identification of carriers. We sought to evaluate the diagnostic accuracy of real-time polymerase chain reaction (PCR) for detection of GBS using a more stringent culture method. METHODS: Two swabs were used simultaneously to obtain rectovaginal GBS samples from consenting women. One swab was analyzed using a stringent, validated culture technology, which included direct plating onto selective agar and inoculation of a selective broth. The other swab was used for a commercial real-time PCR assay, which uses amplification to detect the presence of the cfb gene sequence of GBS DNA. We calculated the assay accuracy using sensitivity and specificity. RESULTS: A total of 233 samples were available. Both the culture and PCR methods were positive for 59 and negative for 157 patients. The culture method was positive and PCR was negative in 9 patients. The culture was negative and the PCR positive for 8 patients. The sensitivity of the PCR assay was 86.8% and specificity was 95.2%. The positive predictive value was 88.1% and the negative predictive value was 94.6%. CONCLUSION: Although a rapid PCR assay may be useful to determine GBS status in the urgent intrapartum setting, the false-negative rate of 13.2% for the real-time PCR assay prohibits its use for standard GBS screening in the office.

Ischemia-modified albumin increases after skeletal muscle ischemia during arthroscopic knee surgery.

BACKGROUND: Ischemia can alter the ability of albumin to bind free metal atoms. Based on these biochemical changes, methods to quantify ischemia modified albumin (IMA) were developed to assist in the evaluation of patients with symptoms of cardiac ischemia. Since ischemia can occur in any vascular bed, the specificity of IMA for cardiac muscle ischemia is unclear and requires further investigation. METHODS: We evaluated the specificity of an IMA test in patients with skeletal muscle ischemia during arthroscopic knee surgery. A pressurized thigh cuff was continuously inflated to 300 mm Hg on the operative leg, in order to arrest blood flow during the procedure. Samples were collected before surgery, 15 min after surgery, and prior to discharge. RESULTS: Twenty-three patients were enrolled in the study. Median tourniquet time was 29 min (range 19-108). Median pre-operative IMA was 90.2 KU/l (range 77-101.6). Statistically significant (p<0.05) increases in IMA and myoglobin concentrations, and decreases in albumin concentrations were observed following tourniquet release and before discharge. CONCLUSIONS: Post-operative myoglobin elevations indicated that skeletal muscle ischemia was sufficient to produce detectable myocyte necrosis. Post-operative IMA increases are consistent with ischemic modification of albumin during exposure to ischemic conditions in skeletal muscle during and /or immediately after tourniquet application. However, the negative correlations between IMA and albumin results suggest that increases in IMA were in part due to lower post-operative albumin concentrations resulting in decreased cobalt binding.

Experimental rat models of types 1 and 2 diabetes differ in sympathetic neuroaxonal dystrophy.

Dysfunction of the autonomic nervous system is a recognized complication of diabetes, ranging in severity from relatively minor sweating and pupillomotor abnormality to debilitating interference with cardiovascular, genitourinary, and alimentary dysfunction. Neuroaxonal dystrophy (NAD), a distinctive distal axonopathy involving terminal axons and synapses, represents the neuropathologic hallmark of diabetic sympathetic autonomic neuropathy in man and several insulinopenic experimental rodent models. Although the pathogenesis of diabetic sympathetic NAD is unknown, recent studies have suggested that loss of the neurotrophic effects of insulin and/or insulin-like growth factor-I (IGF-I) on sympathetic neurons rather than hyperglycemia per se, may be critical to its development. Therefore, in our current investigation we have compared the sympathetic neuropathology developing after 8 months of diabetes in the streptozotocin (STZ)-induced diabetic rat and BB/ Wor rat, both models of hypoinsulinemic type 1 diabetes, with the BBZDR/Wor rat, a hyperglycemic and hyperinsulinemic type 2 diabetes model. Both STZ- and BB/Wor-diabetic rats reproducibly developed NAD in nerve terminals in the prevertebral superior mesenteric sympathetic ganglia (SMG) and ileal mesenteric nerves. The BBZDR/Wor-diabetic rat, in comparison, failed to develop superior mesenteric ganglionic NAD in excess of that of age-matched controls. Similarly, NAD which developed in axons of ileal mesenteric nerves of BBZDR/Wor rats was substantially less frequent than in BB/Wor- and STZ-rats. These data, considered in the light of the results of previous experiments, argue that hyperglycemia alone is not sufficient to produce sympathetic ganglionic NAD, but rather that it may be the diabetes-induced superimposed loss of trophic support, likely of IGF-I, insulin, or C-peptide, that ultimately causes NAD.

Establishment of reference intervals for markers of fetal thyroid status in amniotic fluid.

Fetal goiter can arise as a result of fetal hyper or hypothyroidism. Although this condition is rare, it can be life threatening. Detection of fetal goiter in utero is possible with the aid of ultrasound, but proper prenatal treatment depends on knowledge of hormonal status. Amniotic fluid (AF) sampling is less technically demanding and poses fewer risks to the fetus than cordocentesis for fetal serum sampling, but well-established reference ranges for AF thyroid studies are not available in the literature. We have established reference intervals for AF (TSH), total T(4) (tT(4)), and free T(4) using stored AF samples. The reference intervals were: TSH (n = 127), less than 0.1-0.5 mU/liter, with a median of 0.1 mU/liter; tT(4) (n = 129), 2.3-3.9 microg/dl (30-50 nmol/liter), with a median of 3.3 microg/dl (4 nmol/liter); and free T(4) (n = 119) less than 0.4-0.7 ng/dl (5-9 pmol/liter), with a median of 0.4 ng/dl (5 pmol/liter). These intervals represent the largest study done to date on third trimester AF using automated immunoassays. A literature search of fetal goiter revealed a number of cases of hypothyroidism. Seven cases reported AF TSH concentrations (range, 1.1-28.9 mU/liter) and four reported AF tT(4) concentrations [range, 0.98-1.25 micro g/ml (13-16 nmol/liter)], all of which fell outside our reference intervals. These data support the use of AF to diagnose fetal hypothyroidism, reducing the need to resort to a riskier procedure such as cordocentesis.

Rules-based detection of discrepancies between TSH and free T4 results.

BACKGROUND: Analytical errors in clinical laboratory testing are unavoidable. Recent reports have suggested the idea of "physiological profiling" which uses several results from a given patient to identify clinically unlikely results. The objective of this study was to establish rules-based criteria for identifying physiologically unlikely TSH and free T(4) (fT(4)) results. METHODS: For a 30-month period, all samples with fT(4) concentrations >2 ng/dl and TSH concentrations >0.1 micro IU/ml were investigated. RESULTS: Among 7918 plasma samples for which both TSH and fT(4) concentrations were measured, 18 (0.23%) had fT(4) and TSH exceeding the investigated limits. Of these, two were due to heterophile antibody interference (with the TSH assay), one was proven to be due to random error, four could be explained by the patients' conditions, three were from infants <1 week of age, and the remaining eight were unresolved, primarily due to insufficient sample, discarded samples, and inability to obtain patient histories. CONCLUSIONS: This study defined a rules-based alert system for clinically unlikely combinations of TSH and fT(4) results. This pilot study demonstrates that this system is capable of detecting at least two different types of laboratory errors that would have otherwise gone undetected.

Measurement of glycated hemoglobin percentages for use in the diagnosis and monitoring of diabetes mellitus in nonhuman primates.

OBJECTIVE: To identify a technique for measurement of glycated hemoglobin percentage in blood samples obtained from various species of nonhuman primates (NHPs), to determine whether these percentages varied with respect to glycemic control, and to assess whether this physiologic variable provided a suitable test for diagnosing diabetes mellitus in NHPs. SAMPLE POPULATION: 166 blood samples collected from 121 NHPs comprising 22 species from the Haplorhine and Strepsirhine suborders and including nondiabetic, treated-diabetic, and diabetic animals in 23 zoologic institutions throughout the United States. PROCEDURE: Hemoglobin A1c percentage was measured in 154 samples by use of high-performance liquid chromatography. Total glycated hemoglobin percentage was measured in 159 samples by use of a boronate-affinity chromatographic assay. Glucose concentration was measured in 157 samples with an autochemical analyzer by use of a hexose kinase method. RESULTS: The boronate-affinity chromatographic technique for measurement of total glycated hemoglobin percentage was the most suitable method. Nondiabetic Haplorhines had percentages higher than those in nondiabetic Strepsirhines. In Haplorhines, diabetic animals had percentages higher than those in treated-diabetic animals, which had percentages higher than those in nondiabetic animals. In Strepsirhines, this pattern was less pronounced. CONCLUSIONS AND CLINICAL RELEVANCE: Measurement of total glycated hemoglobin percentage provides useful information for diagnosing diabetes mellitus in Haplorhines and, possibly, in Strepsirhines. Until reference ranges are established for each species, it is recommended that results for samples from NHPs without clinical signs of diabetes mellitus be compared with results of samples collected concomitantly from NHPs with clinical signs of this condition.

Comparing analytic performance criteria: evaluation of HbA1c certification criteria as an example.

BACKGROUND: Direct comparison of analytical performance criteria that utilize different statistical approaches can be problematic. We describe a mathematical approach to compare performance criteria for hemoglobin A1c (HbA1c) analysis used by the NGSP standardization program and the College of American Pathologists (CAP) to enhance consistency between the schemes. METHODS: The imprecision (CV) and bias combinations required to pass each criterion at probabilities of 0.95, 0.99 and 0.999 were calculated and used to construct contour plots to compare them. The CV/bias requirements were calculated mathematically for the 2011-2012 CAP (3/3 results within ±7% of the target) and different proposed NGSP (33/40 to 40/40 results within ±7% of the target) criteria, and using computer simulations for the existing NGSP criterion (95% confidence interval of the differences between the method and NGSP within ±0.75% HbA1c). RESULTS: Requiring 37 of 40 results to be within ±7% of the NGSP target best matched the CAP criterion at zero bias (95% chance of passing). CONCLUSIONS: The NGSP Steering Committee recommended a certification criterion of 37 of 40 results within ±7% of the NGSP (reduced to ±6% in 2014). The described evaluation approach may be useful in other situations where comparison of different performance criteria is desired.

Validating the performance of QC procedures.

A methodology for computing the maximum expected number of unreliable patient results produced because of an out-of-control condition for a given quality control strategy is presented along with strategies for changing the expected number of unreliable results produced and reported. The expected number of unreliable patient results reported before and after the last accepted quality control evaluation before the detection of an out-of-control condition are discussed and used as design criteria for quality control strategies that meet a laboratory's risk criteria.