Accuracy of the Number Needed to Treat Compared With Diagnostic Testing: Brief Critical Review.

CONTEXT.­: Number needed to treat (NNT) seems simple and is widely used. But its derivation from the absolute risk reduction is difficult to visualize. Like diagnostic sensitivity, absolute risk reduction is a measure of treatment accuracy. Thus, NNT is a measure of accuracy. NNT is inversely proportional to the relative risk reduction and the baseline risk that may be torturous when accuracy is poor. In order to better visualize the accuracy and weaknesses of NNT, NNT is compared with variables in diagnostic science that are often better understood, such as diagnostic sensitivity, specificity, and positive predictive value. OBJECTIVE.­: To better understand the accuracy of NNT. DATA SOURCES.­: Receiver operating characteristic curves are used to help visualize accuracy. It is shown that baseline risk and prevalence are highly correlated. Like positive predictive value, NNT is dependent on prevalence. Similar to diagnostic testing, symptoms and additional testing can increase prevalence and improve accuracy of the NNT. Examples are shown where changes in prevalence cause alterations in NNT. Moreover, data indicate that when accuracy of NNT is low, although the average NNT may be favorable, some subgroups may exhibit very poor response and even harm. It is shown that manipulations to increase prevalence can produce smaller, more selective groupings that can improve the accuracy and reduce the cost of expensive drugs. CONCLUSIONS.­: When the power of prediction is poor, the value of NNT must be carefully deliberated because it may be misleading. Indeed, the upper confidence interval may be a better reflection of NNT than the average.

The Predicament of Large Numbers of Observations and How We Got There: Critical Review.

BACKGROUND: Classical statistics were developed in a time when small sample sizes were the norm; thus, statistical significance typically ensured large clinical effects. Over the past 10-20 years, computational techniques have allowed studies with modest effects to reach statistical significance (usually P < 0.05) by analyzing very large numbers of patients. In this review, I discuss how this came about and provide an intuitive understanding of the strengths and weaknesses of various statistical parameters that provide insight into clinical effect sizes. CONTENT: In this review of the literature, a simple web-based program was used for calculations. Examples are shown. Odds and risk ratios are compared with ROC curves to allow better understanding of their predictive value. SUMMARY: In these complex times, an intuitive understanding of statistical procedures is increasingly important. This review will attempt to advance the reader's knowledge so that one can calculate the number needed to treat and its confidence interval, understand the meaning of a modest association, and determine when a study is likely to be accurate but with questionable clinical utility.

Non-High-Density Lipoprotein Cholesterol and Guidelines for Cholesterol Lowering in Recent History.

BACKGROUND: The National Cholesterol Education Program (NCEP) released guidelines for treating cholesterol in 1988, 1994, and 2002. After a hiatus, the guidelines were released again in 2013, 2016, 2017, and 2018. METHODS: In this article, I review these guidelines, factors that affected their release, how they evolved, and why recommended treatment targets are reasonable. Also, to aid reader understanding, I briefly discuss biochemical mechanisms and the pathophysiology of beta-lipoproteins, focusing on the importance on non-high-density cholesterol (non-HDLC) in assessing risk and as a target for treatment. The concepts discussed are important to laboratory clinicians because those workers inscribe target values on the reports and may consult with medical staff members. CONCLUSIONS: The newest recommendations, released in 2018, are an extension of the 2017 guidelines that defined non-HDLC as equivalent to low-density lipoprotein cholesterol (LDLC). For the reasons discussed herein, non-HDLC has advantages over LDLC. Laboratories reporting cholesterol results should include non-HDLC values and cutoffs in their reports.

Critical review of 2016 ACC guidelines on therapies for cholesterol lowering with reference to laboratory testing.

INTRODUCTION: This review discusses new guidelines that were released in 2016 and 2017 for assessing risk of coronary disease (ASCVD) and treatments and that appear to have replaced the 2013 guidelines which superseded the National Cholesterol Education Program Adult Treatment Panel Guidelines (ATP). To put the new guidelines in prospective, I briefly review the history of prior guidelines. The newest guidelines seem to support the idea that elevated LDL cholesterol (C) is a cause of ASCVD. The Review also discusses issues related to these guidelines, especially measurement of LDLC, the importance of nonHDLC and puzzling results that do not seem to support the cause and effect idea. METHODS: Literature review and critical discussion. CONCLUSIONS: The 2016 guidelines appear to eliminate most of the flaws in the prior 2013 guidelines that replaced the ATP. They do not seem to rely on randomized control studies alone but the totality of all of the evidence. The new guidelines still fail to identify some persons with characteristics of metabolic syndrome that may be of increased risk for ASCVD although they address some of the problems in treating this group. The guidelines identify LDL cholesterol and nonHDL cholesterol target concentrations that are important for laboratory professionals since they should be defined on reports and are important for consultation.

Analysis, detection and quantitation of mixed cryoglobulins in HCV infection: brief review and case examples.

Considering the high incidence of cryoglobulins in hepatitis C virus (HCV) infection together with the high worldwide prevalence of HCV infection, identification of clinically apparent mixed cryoglobulinemia syndrome is increasingly important as most patients who are identified can now be successfully treated. Different approaches for the detection, analysis and reporting of cryoglobulins have been described and there is a wide variation in results reported, ranging from a qualitative "negative" or "positive", to a quantitative report including cryoglobulin type and the total protein. Protein and immunofixation (IFE) electrophoresis are generally used to identify and characterize cryoglobulins, as these methods quantify and phenotype. Here, we present a brief review of the literature and demonstrate a case oriented approach for identifying mixed cryoglobulinemia from the preanalytical phase, leading up to and including the analytical phase with characterization by IFE. Most patients with mixed cryoglobulinemia can now be treated with success. Nevertheless, the high cost may limit treatment of those with symptoms unless there is laboratory evidence for mixed cryoglubulinemia. Low levels of cryoglobulins can be associated with severe symptoms; as a result, accurate identification of cryoglobulins may be of increasing importance since clear identification may be a good reason to initiate treatment.

Strengths and weaknesses of methods for identifying monoclonal free light chains of Ig: examples from two cases with renal disease.

BACKGROUND: Serum free light chain (FLC) analysis with ratio and urine immunofixation electrophoresis (IFE) are both available for routine use in helping to detect plasma cell dyscrasia and related diseases. CASES: Case reports showing one serum positive for serum FLC but that showed a hook effect and overestimated the amount of monoclonal FLC while urine IFE was negative for Bence Jones protein, and a second serum that showed elevated FLC κ and λ but a normal κ/λ ratio, while urine IFE was positive for Bence Jones protein. CONCLUSIONS: These two techniques complement one another. Neither of the techniques is truly quantitative, and both exhibit methodological defects.

Implications of reverse cholesterol transport: recent studies.

INTRODUCTION: There is a strong epidemiological relationship between high density lipoproteins and atherosclerotic coronary vascular disease (ASCVD). The process of reverse cholesterol transport (RCT) has been hypothesized to help explain this relationship. The corollary that raising HDL should reduce ASCVD is also drawn from this relationship. In recent years, the metabolism of HDL has become better understood. A hypothetical process for explaining RCT has been superimposed on the currently understood HDL metabolic pathways. METHODS: Outline of HDL metabolism and the superimposed RCT process. Literature review of studies of persons with genetic defects, HDL cholesterol raising clinical trials, Mendelian randomization studies and treatments with molecules that mimic HDL. CONCLUSIONS: Mutation studies of ABCA1, LCAT and SR-B1 genes in humans showed expected variations in HDLC but little association with ASCVD and there was no significant association between HDLC and ASCVD in Mendelian randomization studies. Elevations in HDLC due to treatment with niacin and cholesteryl ester transport protein inhibitors in randomized trials raised HDLC but did not significantly reduce risk of ASCVD. Treatment with molecules that mimic HDL did not seem to reduce ASCVD. Thus, recent evidence does not seem to support RCT as currently proposed. This hypothesis seems to need substantial revision.

Resolving a case of concurrent hepatitis B virus surface antigen (HBsAg) and surface antibody (HBsAb).

BACKGROUND: In most cases, patients appear to recover from acute hepatitis B virus (HBV) infection and do not exhibit the surface antigen (HBsAg). Chronic carriers are positive for HBsAg but HBsAb is usually not present. After acute infection only HBsAb remains. The presence of both HBsAg and HBsAb is unusual. METHODS: We report on a patient whose results were analytically and clinically discrepant--positive for HBsAg and HBsAb on one testing platform but only HBsAg on another platform. RESULTS: Reasons for this result: 1) Interference from endogenous antibodies; 2) HBsAg is from one strain and HBsAb is from another: and 3) The presence of HBsAb and HBsAg. Growing evidence indicates that both may be present in many patients. Low HBsAb may be neutralized and not recognized by the solid-phase HBsAg in the assay. Likewise, low HBsAg may be neutralized by HBsAb. It remains unclear whether HBsAg is always cleared after acute infection. CONCLUSIONS: Testing indicated that both HBsAb and HBsAb were present. The data shows that different testing platforms may produce different results depending on the kinetics, the exposure of the capture HBsAg and the extent of endogenous HBAg/HBsAb. We demonstrate a simple way to rule out test interference to presumptively identify true HBsAb.

Brief critical review: Statistical assessment of biomarker performance.

INTRODUCTION: Articles have suggested that C-statistics associated with receiver operator characteristic (ROC) curves are insufficiently sensitive compared to odds/risk ratios (OR/RR) defined by p-values. Some suggest reclassification techniques are more appropriate. METHODS: We review the concepts of p-values, OR/RR, and ROC curves. We construct a ROC curve, demonstrate parametric and nonparametric curves, and analyze a comparison of ROC curves. RESULTS: Using these illustrations, we show that the ROC curve is not simply a C-statistic but a continuum of sensitivity/specificity pairs over decision levels. We demonstrate that p-values provide little useful information about discrimination and that OR/RR is a limited expression of a sensitivity/specificity plot. We illustrate that low prevalence produces low positive predictive values, reclassification techniques are subject to the same rules of prevalence as other analysis, and that modifying the analysis can decrease the p-value comparing C-statistics without altering the sensitivity/specificity plot. CONCLUSIONS: ROC curves provide both visual and statistical information to support entry into large-scale trials, determining decision levels and the use of testing in the absence of such trials. If the sensitivity/specificity plot shows little improvement at appropriate decision point(s), statistically significant-improvement in other "novel" statistical indices should be suspect.

POEMS syndrome: importance of the clinical laboratory practitioner's role.

BACKGROUND: Polyneuropathy organomegaly, endocrinopathy, monoclonal (M) gammopathy, skin syndrome (POEMS) may be difficult to diagnose as a result of methodological and clinical idiosyncrasies. Four of eleven criteria (M-protein, VEGF, endocrinopathy and thrombocytosis/polycythemia) are closely associated with clinical laboratory testing. POEMS has been largely associated with λ-M-gammopathies. Vascular endothelial growth factor (VEGF) is a recent addition to the major diagnostic criteria. VEGF may alter vascular permeability causing some manifestations of POEMS. METHODS: Review of the literature that focuses on clinical laboratory issues--endocrinological findings, identification of monoclonal gammopathies by electrophoresis and case demonstration. RESULTS: Based on the criterion of VEGF, POEMS was diagnosed in a patient with a κ-M-gammopathy. CONCLUSIONS: Low-level IgA monoclonal proteins that are common in POEMS are often difficult to identify by serum electrophoresis (SPE). Immunofixation electrophoresis is required when polyneuropaties are investigated and an M-protein is not identified by SPE. VEGF may improve the sensitivity for diagnosis of POEMS and findings from capillary leak syndrome which are also associated with elevated VEGF and M-gammopathy suggests that κ-M-gammopathies may be implicated more often. It is demonstrated that due to computerized records, the laboratory practitioner is well suited to help the clinician make this complicated diagnosis.

Complementarily of urine analysis and serum free light chain assay for assessing response treatment response: illustrated by three case examples.

BACKGROUND: With the introduction of a new sensitive serum assay for monoclonal free light chains (FLC), there is a question as to whether or not traditional urine immunofixation electrophoresis (IFE) is still necessary. Therapy for B-cell disease has greatly improved in recent time so that noninvasive biomarkers that suggest complete remissions have become increasingly important. A Consensus Opinion stated that because of methodological imperfections, at concentrations of FLC <100 mg/l, urine IFE should supplement serum FLC ratio. METHODS: Examples are presented from a review of laboratory results and medical records from 3 patients on treatment after diagnosis of amyloidosis AL and multiple myeloma. RESULTS: Monoclonal FLC was identified in urine by IFE while the κ/λ ratio from the serum FLC assay was within the reference range. CONCLUSION: The results from these 3 cases suggest that as the FLC concentration drops, the serum FLC results may become more ambiguous. These results are consistent with guidelines cautioning about conclusions drawn from the serum FLC assay alone. Although more study is needed, the examples presented here illustrate that traditional urine IFE appear to adds important information regarding potential complete remissions and remains an important complementary assay to the serum FLC assay.

Urine immunofixation electrophoresis remains important and is complementary to serum free light chain.

Articles have debated whether or not urine analysis remains valuable for identifying monoclonal gammopathies. A general impression is that the newer serum free light chain (FLC) assay is more analytically sensitive, more quantitative and simpler to perform. Many laboratory directors may have seized on the idea of eliminating urine analysis because it is a tedious procedure and requires expert interpretation while most laboratories can perform automated serum FLC assay. Others have concluded that urine immunofixation electrophoresis (IFE) optimizes the diagnostic sensitivity and should be included when there is a clinical indication. Here, I show that papers faulting urine analysis often used inappropriate urine methodology and this helps explain why there was misinterpretation. Moreover, the literature, shows urine IFE is often more sensitive for identifying low-level monoclonal FLC than the serum assay because urine IFE is as sensitive when performed appropriately and generally more specific. Besides, the reference range for serum FLC assay is unclear which is a great problem in assessing response to treatment and in identifying diseases when there is low concentration monoclonal FLC. I conclude that urine IFE remains important and is complementary to serum FLC assay, although the best algorithms for use remains to be elucidated.

Polyclonal free light chain of Ig may interfere with interpretation of monoclonal free light chain κ/λ ratio.

There is controversy about whether a sensitive assay for the serum Ig free light chain (FLC) κ/λ ratio can replace urine immunofixation electrophoresis (UIFE). This report describes two untreated patients in whom monoclonal FLCs were identified in urine despite normal serum FLC κ/λ ratios. Unlike the classical serum electrophoretic patterns in multiple myeloma, both serum samples showed adequate amounts of polyclonal Ig. The most likely explanation is a masking effect by polyclonal FLC on the serum κ/λ ratio when sufficient concentrations of polyclonal FLC exist. These cases illustrate this likely effect and attest to the continued importance of UIFE for initial screening of patients for Bence-Jones protein.

Weak associations between prognostic biomarkers and disease in preliminary studies illustrates the breach between statistical significance and diagnostic discrimination.

INTRODUCTION: Prior to actual well-designed outcome studies, statistical significance does not imply diagnostic usefulness. Sensitivity/specificity graphs have been a mainstay in evaluating potential diagnostic usefulness. Unfortunately, many recent studies have been expressed as relative risk (RR) or odds ratios (OR) based on statistical significance that produces misleading constructs, especially when the relationship between the biomarker and disease is weak. The purpose of this review is to discuss the meaning of the diagnostic relationships when the association between a biomarker and a disease is weak such that it shows poor discrimination by sensitivity/specificity plots. METHODS: Review of the literature, illustrative example and calculation of positive predictive values. RESULTS: Information from the literature showing the equivalency between OR and sensitivity/specificity plots and defining weak diagnostic discrimination. Demonstration of how confounders can alter between study statistical significance when there is a weak diagnostic relationship. CONCLUSIONS: RR or OR should be expressed as diagnostic sensitivity/specificity plots. Usually predictive values will identify whether or not there is potential diagnostic value that is economically reasonable. Very weak diagnostic relationships produce very low positive predictive values. Great caution should be exercised when considering weak associations for diagnostic use, especially if the biomarker is not a modifiable risk factor.

Clinical validation of biomarkers for predicting risk.

BACKGROUND: A useful biomarker should improve clinical management in an economically reasonable way. This should be determined from well-designed outcome studies that show clinical management can be altered on the basis of the biomarker. It is important not to confuse results from testing prior to outcome with outcome studies. CONTENT: This chapter reviews statistical tests used to evaluate studies performed prior to final outcome studies and criteria that assess whether or not a biomarker should be considered for outcome studies at each step. I review how relative risk and odds ratios are related to receiver operator characteristic (ROC) plot analysis. Other statistical techniques such as reclassification and the Hosmer Lemeshow test that have been suggested for evaluation of diagnostic usefulness are considered. Weaknesses of each technique are discussed. SUMMARY: I consider ROC analysis to be a mainstay against which other statistical tests of diagnostic performance should be compared. The importance of expressing data in terms of predictive values is emphasized. Tests showing weak diagnostic associations with a disease are difficult to evaluate for outcome study application, because there is usually great difference in between-study variance so that the true relationship between the biomarker, its diagnostic ability, and predictive capability are unclear.