Posts Tagged ‘multivariate analysis’

A Future for Plasma Metabolomics in Cardiovascular Disease Assessment

Curator: Larry H Bernstein, MD, FCAP



Plasma metabolomics reveals a potential panel of biomarkers for early diagnosis
in acute coronary syndrome  

CM. Laborde, L Mourino-Alvarez, M Posada-Ayala,
G Alvarez-Llamas, MG Serranillos-Reus, et al.
Metabolomics – manuscript draft

In this study, analyses of peripheral plasma from Non-ST Segment Elevation
Acute Coronary Syndrome patients and healthy controls by gas chromatography-
mass spectrometry permitted the identification of 15 metabolites with statistical
differences (p<0.05) between experimental groups.
In our study, 6 amino acids were found decreased in NSTEACS patients when
compared with healthy control group suggesting either a decrease in anabolic
activity of these metabolites or an increase in the catabolic pathways. Of both
possibilities, the increased catabolism of the amino acids can be explained
considering simultaneously the capacity of glycogenic and ketogenic amino
acids along with the gradual hypoxic condition to which cardiac muscle cells
have been exposed.

Additionally, validation by gas chromatography-mass spectrometry and liquid
chromatography-mass spectrometry permitted us to identify a potential panel
of biomarkers formed by 5-OH tryptophan, 2-OH-butyric acid and 3-OH-butyric
acid. Oxidative stress conditions dramatically increase the rate of hepatic
synthesis of glutathione. It is synthesized from the amino acids cysteine, glutamic
acid and glycine. Under these conditions of metabolic stress, the supply of cysteine
for glutathione synthesis become limiting and homocysteine is used to form
cystathionine, which is cleaved to cysteine and 2-OH-butyric acid. Thus elevated
plasma levels of 2-OH-butyric acid can be a good biomarker of cellular oxidative
stress for the early diagnosis of ACS.  Another altered metabolite of similar
structure was 3-OH-butyric acid, a ketone body together with the acetoacetate,
and acetone. Elevated levels of ketone bodies in blood and urine mainly occur
in diabetic ketoacidosis. Type 1 diabetes mellitus (DMI) patients have decreased
levels of insulin in the blood that prevent glucose enter cells so these cells use
the catabolism of fats as energy source that produce ketones as final products.
This panel of biomarkers reflects the oxidative stress and the hypoxic state that
disrupts the myocardial cells and consequently constitutes a metabolomic
signature that could be used for early diagnosis of acute coronary syndrome.
We hypothesize that the hypoxia situation comes to “mimic” the physiological
situation that occurs in DMI. In this case, the low energy yield of glucose
metabolism “forces” these cells to use fat as energy source (through catabolism
independent of aerobic/anaerobic conditions) occurring ketones as final
products. In our experiment, the 3-OH-butyric acid was strongly elevated in
NSTEACS patients.


Current Methods Used in the Protein Carbonyl Assay
Nicoleta Carmen Purdel, Denisa Margina and Mihaela Ilie.
Ann Res & Rev in Biol 2014; 4(12): 2015-2026.

The attack of reactive oxygen species on proteins and theformation of
protein carbonyls were investigated only in the recent years. Taking into
account that protein carbonyls may play an important role in the early
diagnosis of pathologies associated with reactive oxygen species
overproduction, a robust and reliable method to quantify the protein
carbonyls in complex biological samples is also required. Oxidative
stress represents the aggression produced at the molecular level by
the imbalance between pro-oxidant and antioxidant agents, in favor of
pro-oxidants, with severe functional consequences in all organs and
tissues. An overproduction of ROS results in oxidative damages
especially to proteins (the main target of ROS), as well as in lipids,or
DNA. Glycation and oxidative stress are closely linked, and both
phenomena are referred to as ‘‘glycoxidation’’. All steps of glycoxidation
generate oxygen-free radical production, some of them being common
with lipidic peroxidation pathways.
The initial glycation reaction is followed by a cascade of chemical
reactions resulting in the formation of intermediate products (Schiff base,
Amadori and Maillard products) and finally to a variety of derivatives
named advanced glycation end products (AGEs). In hyperglycemic
environments and in natural aging, AGEs are generated in increased
concentrations; their levels can be evaluated in plasma due to the fact
that they are fluorescent compounds. Specific biomarkers of oxidative
stress are currently investigated in order to evaluate the oxidative status
of a biological system and/or its regenerative power. Generaly, malondi-
aldehyde, 4-hydroxy-nonenal (known together as thiobarbituric acid
reactive substances – TBARS), 2-propenal and F2-isoprostanes are
investigated as markers of lipid peroxidation, while the measurement
of protein thiols, as well as S-glutathionylated protein are assessed
as markers of oxidative damage of proteins. In most cases, the
oxidative damage of the DNA has 8-hydroxy-2l-deoxyguanosine
(8-OHdG) as a marker.  The oxidative degradation of proteins plays an
important role in the early diagnosis of pathologies associated with
ROS overproduction. Oxidative modification of the protein structure
may take a variety of forms, including the nitration of tyrosine residues,
carbonylation, oxidation of methionine, or thiol groups, etc.

The carbonylation of protein represents the introduction of carbonyl
groups (aldehyde or ketone) in the protein structure, through several
mechanisms: by direct oxidation of the residues of lysine, arginine,
proline and threonine residues from the protein chain, by interaction
with lipid peroxidation products with aldehyde groups (such as 4-
hydroxy-2-nonenal, malondialdehyde, 2-propenal), or by the
interaction with the compounds with the carbonyl groups resulting
from the degradation of the lipid or glycoxidation. All of these
molecular changes occur under oxidative stress conditions.
There is a pattern of carbonylation, meaning that only certain
proteins can undergo this process and protein structure determines
the preferential sites of carbonylation. The most investigated
carbonyl derivates are represented by gamma-glutamic
semialdehyde (GGS) generated from the degradation of arginine
residue and α-aminoadipic semialdehyde (AAS) derived from lysine.

A number of studies have shown that the generation of protein
carbonyl groups is associated with normal cellular phenomena like
apoptosis, and cell differentiation and is dependent on age, species
and habits (eg. smoking) or severe conditions’ exposure (as
starvation or stress). The formation and accumulation of protein
carbonyls is increased in various human diseases, including –
diabetes and cardiovascular disease.

Recently, Nystrom [7] suggested that the carbonylation process
is associated with the physiological and not to the chronological
age of the organism and the carbonylation may be one of the causes
of aging and cell senescence; therefore it can be used as the marker
of these processes. Jha and Rizvi, [15] proposed the quantification of
protein carbonyls in the erythrocyte membrane as a biomarker of aging

PanelomiX: A threshold-based algorithm to create panels of

X Robin, N Turck, A Hainard, N Tiberti, F Lisacek. 
T r a n s l a t i o n a l  P r o t e o m i c s   2 0 1 3; 1: 57–64.

The computational toolbox we present here – PanelomiX – uses
the iterative combination of biomarkers and thresholds (ICBT) method.
This method combines biomarkers andclinical scores by selecting
thresholds that provide optimal classification performance. Tospeed
up the calculation for a large number of biomarkers, PanelomiX selects
a subset ofthresholds and parameters based on the random forest method.
The panels’ robustness and performance are analysed by cross-validation
(CV) and receiver operating characteristic(ROC) analysis.

Using 8 biomarkers, we compared this method against classic
combination procedures inthe determination of outcome for 113 patients
with an aneurysmal subarachnoid hemorrhage. The panel classified the
patients better than the best single biomarker (< 0.005) and compared
favourably with other off-the-shelf classification methods.

In conclusion, the PanelomiX toolbox combines biomarkers and evaluates
the performance of panels to classify patients better than single markers
or other classifiers. The ICBT algorithm proved to be an efficient classifier,
the results of which can easily be interpreted. 

Multiparametric diagnostics of cardiomyopathies by microRNA
CS. Siegismund, M Rohde, U Kühl,  D  Lassner.
Microchim Acta 2014 Mar.

MicroRNAs (miRNAs) represent a new group of stable biomarkers
that are detectable both in tissue and body fluids. Such miRNAs
may serve as cardiological biomarkers to characterize inflammatory
processes and to differentiate various forms of infection. The predictive
power of single miRNAs for diagnosis of complex diseases may be further
increased if several distinctly deregulated candidates are combined to
form a specific miRNA signature. Diagnostic systems that generate
disease related miRNA profiles are based on microarrays, bead-based
oligo sorbent assays, or on assays based on real-time polymerase
chain reactions and placed on microfluidic cards or nanowell plates.
Multiparametric diagnostic systems that can measure differentially
expressed miRNAs may become the diagnostic tool of the future due
to their predictive value with respect to clinical course, therapeutic
decisions, and therapy monitoring.

Nutritional lipidomics: Molecular metabolism, analytics, and
JT. Smilowitz, AM. Zivkovic, Yu-Jui Y Wan, SM. Watkins, et al.
Mol. Nutr. Food Res2013, 00, 1–17.

The term lipidomics is quite new, first appearing in 2001. Its definition
is still being debated, from “the comprehensive analysis of all lipid
components in a biological sample” to “the full characterization of
lipid molecular species and their biological roles with respect to the
genes that encode proteins that regulate lipid metabolism”. In principle,
lipidomics is a field taking advantage of the innovations in the separation
sciences and MS together with bioinformatics to characterize the lipid
compositions of biological samples (biofluids, cells, tissues, organisms)
compositionally and quantitatively.

Biochemical pathways of lipid metabolism remain incomplete and the
tools to map lipid compositional data to pathways are still being assembled.
Biology itself is dauntingly complex and simply separating biological
structures remains a key challenge to lipidomics. Nonetheless, the
strategy of combining tandem analytical methods to perform the sensitive,
high-throughput, quantitative, and comprehensive analysis of lipid
metabolites of very large numbers of molecules is poised to drive
the field forward rapidly. Among the next steps for nutrition to understand
the changes in structures, compositions, and function of lipid biomolecules
in response to diet is to describe their distribution within discrete functional
compartments lipoproteins. Additionally, lipidomics must tackle the task
of assigning the functions of lipids as signaling molecules, nutrient sensors,
and intermediates of metabolic pathways.


Read Full Post »

Landscape of Cardiac Biomarkers for Improved Clinical Utilization

Curator and Author: Larry H Bernstein, MD, FCAP


This reviewer has been engaged in the development, the application, and the validation of cardiac biomarkers for over 30 years. There has been a nonlinear introduction of new biomarkers in that period, with an explosion of methods discovery and large studies to validate them in concert with clinical trials. The improvement of interventional methods, imaging methods, and the unraveling of patient characteristics associated with emerging cardiovascular disease is both cause for alarm (technology costs) and for raised expectations for both prevention, risk reduction, and treatment. What is strikingly missing is the kind of data analyses on the population database that could alleviate the burden of physician overload. It is an urgent requirement for the EHR, and it needs to be put in place to facilitate patient care.


This is a journey through the current status of biochemical markers in cardiac evaluation. 

In the traditional use of cardiac biomarkers, the is a timed blood sampling from the decubital fossa. This was the case with alanine aminotransferase (AST, then SGOT), creatine kinase (CK) or its isoenzyme MB, and lactic dehydrogenase (or the isoenzyme-1). The time of sampling was based on time to appearance from time of damage, and the release of the biomarker is a stochastic process. The earliest studies of CK-MB appearance, peak height, and disappearance was by Burton Sobel and associates related to measuring the extent of damage, and determined that reperfusion had an effect. A significant reason for using a combination of CK-MB and LD-1 was that a patient who is a late arrival might have a CK-MB on the decline (peak at 18 h) while the LD-1 is rising (peak at 48 h).

The introduction of the troponins was accompanied by a serial 4 h measurement, usually for 4 draws (0, 4, 8, 12 h). The computational power of laboratory information systems was limited until recently, so it is somewhat surprising, given what we have seen – in addition to published work in the 1980’s – that this capability is not in use today, when regression and nonparametric classification algorithms are now so advanced that would enable much improved and effective communication to the physician needing the information.

J Adan, LH Bernstein, J Babb. Can peak CK-MB segregate patients with acute myocardial infarction into different outcome classes? Clin Chem 1985; 31(2):996-997. ICID: 844986.

RA Rudolph, LH Bernstein, J Babb. Information induction for predicting acute myocardial infarction. Clin Chem 1988; 34(10):2031-2038. ICID: 825568.

LH Bernstein, IJ Good, GI Holtzman, ML Deaton, J Babb. Diagnosis of acute myocardial infarction from two measurements of creatine kinase isoenzyme MB with use of nonparametric probability estimation. Clin Chem 1989; 35(3):444-447. ICID: 825570.

L H Bernstein, A Qamar, C McPherson, S Zarich, R Rudolph. Diagnosis of myocardial infarction: integration of serum markers and clinical descriptors using information theory. Clin Chem 1999; 72(1):5-13. ICID: 825618

Vermunt, J.K. & Magidson, J. (2000a). “Latent Class Cluster Analysis”, chapter 3 in J.A. Hagenaars and A.L. McCutcheon (eds.), Advances in Latent Class AnalysisCambridge University Press.

Vermunt, J.K. & Magidson, J. (2000b). Latent GOLD 2.0 User’s Guide. Belmont, MA: Statistical Innovations Inc.

LH Bernstein, A Qamar, C McPherson, S Zarich. Evaluating a new graphical ordinal logit method (GOLDminer) in the diagnosis of myocardial infarction utilizing clinical features and laboratory data. Yale J Biol Med. 1999; 72(4):259-268. ICID: 825617.

L Bernstein, K Bradley, S Zarich. GOLDmineR: improving models for classifying patients with chest pain.
Yale J Biol Med. 2002; 75(4):183-198. ICID: 825624

SA Haq, M Tavakol, LH Bernstein, J Kneifati-Hayek, M Schlefer, et al. The ACC/ESC Recommendation for 99th Percentile of the Reference NormalTroponin I Overestimates the Risk of an Acute Myocardial Infarction: a novel enhancement in the diagnostic performance of troponins. “6th Scientific Forum on Quality of Care and Outcomes Research in Cardiovascular Disease and Stroke.” Circulation 2005; 111(20):e313-313. ICID: 939931.

LH Bernstein, MY Zions, SA Haq, S Zarich, J Rucinski, B Seamonds, …., John F Heitner. Effect of renal function loss on NT-proBNP level variations. Clin Biochem 2009; 42(10-11):1091-1098. ICID: 937529

SA Haq, M Tavakol, S Silber, L Bernstein, J Kneifati-Hayek, et al. Enhancing the diagnostic performance of troponins in the acute care setting. J Emerg Med 2008; ICID: 937619

Gil David, LarryH Bernstein, Ronald Coifman. Generating Evidence Based Interpretation of Hematology Screens via Anomaly Characterization. OCCJ 2011; 4(1):10-16. ICID: 939928

The use and limitations of high-sensitivity cardiac troponin and natriuretic peptide concentrations in at risk populations

Background: High-sensitivity cardiac troponin (hs-cTn) assays are now available that can detect measurable troponin in significantly more individuals in the general population than conventional assays. The clinical use of these hs-cTn assays depends on the development of proper reference values. However, even with a univariate biomarker for risk and/or severity of ischemic heart disease, a single reference value for the cardiac biomarker does not discriminate the probabilities between 2 or 3 different cardiac disorders, or identify any combination of these, such as, heart failure or renal disease > stage 2 and acute coronary syndrome. True, the physician has a knowledge of the history and presentation as a guide. Do we know how adequate the information is in a patient who has an atypical presentation? Again, the same problem arises with the use of the natriuretic peptides, but the value of these tests is improved over the previous generation tests. Let us parse through the components of this diagnostic problem, which is critical for reaching the best decisions under the circumstances.

Issue 1. The use of the clinical information, such as, patient age, gender, past medical history, known medical illness, CHEST PAIN, ECG, medications, are the basis of longstanding clinical practice. These may be sufficient in a patient who presents with acute coronary syndrome and a Q-wave not previously seen, or with ST-elevation, ST-depression, T-wave inversion, or rhythm abnormality. Many patients don’t present that way.

Issue 2. The use of a single ‘decision-value’ for critical situations decribed, leaves us with a yes-no answer. If you use a receiver-operator characteristic curve, all of the patients used to construct the sensitivity/specificity analysis have to be decisively known for identification. Otherwise, one might just take the median of a very large population, and the median represents the best value for a data set that is not normal distribution. However, the ROC method may inform about an acute event, if that is the purpose, but with a single value for a single variable, it can’t identify a likelihood of an event in the next six months.

Issue 3. There are several quantitative biomarkers that are considerably better than were available 15 years prior to this discussion. These can be used alone, but preferably in combination for diagnostic evaluation, for predictiong prognosis, and for therapeutic decision-making. What is now available was unimagined 20 years ago, both in test selection and in treatment selection.

Cardiac troponin assays were recently reviewed in Clin Chem by Fred Apple and Amy Seenger. (The State of Cardiac Troponin Assays: Looking Bright and Moving in the Right Direction).

Cardiac troponin assays have evolved substantially over 20 years, owing to the efforts of manufacturers to make them more precise and sensitive. These enhancements have led to high-sensitivity cardiac troponin assays, which ideally would give measureable values above the limit of detection (LoD) for 100% of healthy individuals and demonstrate an imprecision (CV) of ≤10% at the 99th percentile.

As laboratorians, we wish to comment on the recently published “ACCF 2012 Expert Consensus Document on Practical Clinical Considerations in the Implementation of Troponin Elevations”. Our purpose is to address 8 analytical issues that we believe have the potential to cause confusion and that therefore deserve clarification.

Since the initial publications by the National Academy of Clinical Biochemistry (NACB) in 1999 and by the European Society of Cardiology/American College of Cardiology in 2000, when both organizations endorsed cardiac troponin I (cTnI) or cTnT as the preferred biomarker for the detection of myocardial infaction, numerous other organizations have followed suit and promoted the sole use of cardiac troponin in this clinical application. The American College of Cardiology Foundation (ACCF) 2012 Expert Consensus Document summarizes the recently published 2012 Third Universal Definition of Myocardial Infarction by the Global Task Force, thus providing some practical recommendations on the use and interpretation of cardiac troponin in clinical practice.

This commentator has already expressed the view that there is no ‘silver bullet’, and the potential for confusion is not yet going to be resolved. The potential for greater accuracy in diagnosis is bolstered by currently available imaging.

Current strength of cardiac biomarker opportunities:

A recent study measured hs-tnI in 1716 (93%) of the community-based study cohort and 499 (88%) of the healthy reference cohort. Parameters that significantly contributed to higher hs-cTnI concentrations in the healthy reference cohort included age, male sex, systolic blood pressure, and left ventricular mass. Glomerular filtration rate and body mass index were not independently associated with hs-cTnI in the healthy reference cohort. Individuals with diastolic and systolic dysfunction, hypertension, and coronary artery disease (but not impaired renal function) had significantly higher hs-cTnI values than the healthy reference cohort.

The authors concluded that hs-cTnI assay with the aid of echocardiographic imaging in a large, well-characterized community-based cohort demonstrated hs-cTnI to be remarkably sensitive in the general population, and there are important sex and age differences among healthy reference individuals. Even though the results have important implications for defining hs-cTnI reference values and identifying disease, the reference value is not presented, and the question remains about how many subjects in the 88% (499) healthy reference consort had elevated systolic blood pressure or left ventricular hypertrophy (LVH) measured by imaging. Furthermore, while impaired renal function dropped out as an independent predictor of associated hs-cTnI, one would expect it to have a strong association with LVH.

Defining High-Sensitivity Cardiac Troponin Concentrations in the Community.
PM McKie, DM Heublein, CG. Scott, ML Gantzer, …and AS Jaffe.
Depart Med & Lab Med and Pathology, Mayo Clinic and Foundation, Rochester, MN; Siemens Diagnostics, Newark, DE. Clin Chem 2013.

hsTnI with NSTEMI

Another study looks at the prognostic performance of hs-TnI assay with non-STEMI. High-sensitivity assays for cardiac troponin enable more precise measurement of very low concentrations and improved diagnostic accuracy. However, the prognostic value of these measurements, particularly at low concentrations, is less well defined. (This is the sensitivity vs specificity dilemma raised with regard to the impoved hs-cTn assays.) But the value of low measured values is a matter for prognostic evaluation, based on the hypothesis that any cTnI that is measured in serum is leaked from cardiomyocytes. This assay evaluation used the Abbott ARCHITECT. The data were 4695 patients with non–ST-segment elevation acute coronary syndromes (NSTE-ACS) from the EARLY-ACS (Early Glycoprotein IIb/IIIa Inhibition in NSTE-ACS) and SEPIA-ACS1-TIMI 42 (Otamixaban for the Treatment of Patients with NSTE-ACS–Thrombolysis in Myocardial Infarction 42) trials. The primary endpoint was cardiovascular death or new myocardial infarction (MI) at 30 days. Baseline cardiac troponin was categorized at the 99th percentile reference limit (26 ng/L for hs-cTnI; 10 ng/L for cTnT) and at sex-specific 99th percentiles for hs-cTnI.

All patients at baseline had detectable hs-cTnI compared with 94.5% with detectable cTnT. With adjustment for all other elements of the TIMI risk score, patients with hs-cTnI ≥99th percentile had a 3.7-fold higher adjusted risk of cardiovascular death or MI at 30 days relative to patients with hs-cTnI <99th percentile (9.7% vs 3.0%; odds ratio, 3.7; 95% CI, 2.3–5.7; P < 0.001). Similarly, when stratified by categories of hs-cTnI, very low concentrations demonstrated a graded association with cardiovascular death or MI (P-trend < 0.001). Thus, Application of this hs-cTnI assay identified a clinically relevant higher risk of recurrent events among patients with NSTE-ACS, even at very low troponin concentrations.

Prognostic Performance of a High-Sensitivity Cardiac Troponin I Assay in Patients with Non–ST-Elevation Acute Coronary Syndrome. EA Bohula May, MP Bonaca, P Jarolim, EM Antman, …and DA Morrow. Clin Chem 2013.

Combination test with cTnI and a troponin

The next study looks at the value of a combination of cTnT and N-Terminal pro-B-type-natriuretic-peptide (NT proBNP) to predict heart failure risk. Recall that NT proBNP has been a stabd-alone biomarker for CHF. The study was done with the consideration that heart failure (HF) is projected to have the largest increases in incidence over the coming decades. Therefore, would cardiac troponin T (cTnT) measured with a high-sensitivity assay and N-terminal pro-B–type natriuretic peptide (NT-proBNP), biomarkers strongly associated with incident HF, improve HF risk prediction in the Atherosclerosis Risk in Communities (ARIC) study?

Using sex-specific models, we added cTnT and NT-proBNP to age and race (“laboratory report” model) and to the ARIC HF model (includes age, race, systolic blood pressure, antihypertensive medication use, current/former smoking, diabetes, body mass index, prevalent coronary heart disease, and heart rate) in 9868 participants without prevalent HF; area under the receiver operating characteristic curve (AUC), integrated discrimination improvement, net reclassification improvement (NRI), and model fit were described.

Over a mean follow-up of 10.4 years, 970 participants developed incident HF. Adding cTnT and NT-proBNP to the ARIC HF model significantly improved all statistical parameters (AUCs increased by 0.040 and 0.057; the continuous NRIs were 50.7% and 54.7% in women and men, respectively). Interestingly, the simpler laboratory report model was statistically no different than the ARIC HF model.

Troponin T and N-Terminal Pro-B–Type Natriuretic Peptide: A Biomarker Approach to Predict Heart Failure Risk: The Atherosclerosis Risk in Communities Study. V Nambi, X Liu, LE Chambless, JA de Lemos, SS Virani, et al.
Clin Chem 2013.

BCM Researchers Discover Simpler, Improved Biomarkers to Predict Heart Failure As Accurate As Complex Models     Posted by: Anna Ishibashi Sep 17, 2013

Biomarkers for heart failure Researchers at the Baylor College of Medicine and the Michael E. DeBakey Veterans Affairs hospital discovered two improved biomarkers in the bloodstream that predict who is at higher risk of having heart failure in 10 years. The study was published in the journal Clinical Chemistry.

In the Atherosclerosis Risk in Communities (ARIC) clinical study, researchers measured the blood concentration of troponin T and N-terminal-pro-B-type natriuretic peptide (NT-proBNP) in the models, while also collecting age and race data. The important point taken from the study was that researchers did not find any difference in the accuracy of heart failure risk prediction statistically between this simpler test and the traditional, more complex one, which includes information of age, race, systolic blood pressure, antihypertensive medication use, smoking status, diabetes, body-mass index, prevalent coronary heart disease and heart rate.

Troponin T is an indicator of damaged heart muscle and can be detected in low levels even in individuals with no symptoms through this simpler, improved testing method. Similarly, NT-proBNP is a by-product of brain natriuretic peptide (BNP), which is a small neuropeptide hormone that has been shown to be effective in diagnosing congestive heart failure.

The critical issues that we must now address is what lifestyle and drug therapies can prevent the development of heart failures for individuals who are at high risk – according to Dr. Christie Ballantyne, professor of medicine and section chief of cardiology and cardiovascular research at BCM and the Houston Methodist Center for Cardiovascular Disease Prevention.

Although chest pain is widely considered a key symptom in the diagnosis of myocardial infarction (MI), not all patients with MI present with chest pain. This study was done the frequency with which patients with MI present without chest pain and to examine their subsequent management and outcome. A total of 434,877 patients with confirmed MI enrolled June 1994 to March 1998 in the National Registry of Myocardial Infarction, which includes 1674 hospitals in the United States. Outcome measures were prevalence of presentation without chest pain; clinical characteristics, treatment, and mortality among MI patients without chest pain vs those with chest pain.

Of all patients diagnosed as having MI, 142,445 (33%) did not have chest pain on presentation to the hospital. This group of MI patients was, on average, 7 years older than those with chest pain (74.2 vs 66.9 years), with a higher proportion of women (49.0% vs 38.0%) and patients with diabetes mellitus (32.6% vs 25.4%) or prior heart failure (26.4% vs 12.3%). Also, MI patients without chest pain had a longer delay before hospital presentation (mean, 7.9 vs 5.3 hours), were less likely to be diagnosed as having confirmed MI at the time of admission (22.2% vs 50.3%), and were less likely to receive thrombolysis or primary angioplasty (25.3% vs 74.0%), aspirin (60.4% vs 84.5%), β-blockers (28.0% vs 48.0%), or heparin (53.4% vs 83.2%). Myocardial infarction patients without chest pain had a 23.3% in-hospital mortality rate compared with 9.3% among patients with chest pain (adjusted odds ratio for mortality, 2.21 [95% confidence interval, 2.17-2.26]).

We tested the hypotheses that MI patients without chest pain compared with those with chest pain would present later for medical attention, would be less likely to be diagnosed as having acute MI on initial evaluation, and would receive fewer appropriate medical treatments within the first 24 hours. We also evaluated the association between the presence of atypical presenting symptoms and hospital mortality related to MI.

Our results suggest that patients without chest pain on presentation represent a large segment of the MI population and are at increased risk for delays in seeking medical attention, less aggressive treatments, and in-hospital mortality.

Prevalence, Clinical Characteristics, and Mortality Among Patients With Myocardial Infarction Presenting Without Chest Pain. JG Canto, MG Shlipak, WJ Rogers, JA Malmgren, PD Frederick, et al. JAMA 2013; 283(24):3223-3229.

cTnT degraded forms in circulation

This recent study questions whether degraded cTnT forms circulate in the patient’s blood. Separation of cTnT forms by gel filtration chromatography (GFC) was performed in sera from 13 AMI patients to examine cTnT degradation. The GFC eluates were subjected to Western blot analysis with the original antibodies from the Roche immunoassay used to mimic the clinical cTnT assay. GFC analysis of AMI patients’ sera revealed 2 cTnT peaks with retention volumes of 5 and 21 mL. Western blot analysis identified these peaks as cTnT fragments of 29 and 14–18 kDa, respectively. Furthermore, the performance of direct Western blots on standardized serum samples demonstrated a time-dependent degradation pattern of cTnT, with fragments ranging between 14 and 40 kDa. Intact cTnT (40 kDa) was present in only 3 patients within the first 8 h after hospital admission.

Time-Dependent Degradation Pattern of Cardiac Troponin T Following Myocardial Infarction. EPM Cardinaels, AMA Mingels T van Rooij, PO Collinson, FW Prinzen and MP van Dieijen-Visser. Clin Chem 2013.

Older patients with higher cTNI

One of the problems of interpretation of cTnI is the age relationship to the 99th percentile of the elderly. cTnI was measured using a high-sensitivity assay (Abbott Diagnostics) in 814 community-dwelling individuals at both 70 and 75 years of age. The cTnI 99th percentiles were determined separately using nonparametric methods in the total sample, in men and women, and in individuals with and without CVD.

The cTnI 99th percentile at baseline was 55.2 ng/L for the total cohort. Higher 99th percentiles were noted in men (69.3 ng/L) and individuals with CVD (74.5 ng/L). The cTnI 99th percentile in individuals free from CVD at baseline (n = 498) increased by 51% from 38.4 to 58.0 ng/L during the 5-year observation period. Relative increases ranging from 44% to 83% were noted across all subgroups. Male sex [odds ratio, 5.3 (95% CI, 1.5–18.3)], log-transformed N-terminal pro-B-type natriuretic peptide [odds ratio, 1.9 (95% CI, 1.2–3.0)], and left-ventricular mass index [odds ratio, 1.3 (95% CI, 1.1–1.5)] predicted increases in cTnI concentrations from below the 99th percentile (i.e., 38.4 ng/L) at baseline to concentrations above the 99th percentile at the age of 75 years.

cTnI concentration and its 99th percentile threshold depend strongly on the characteristics of the population being assessed. Among elderly community dwellers, higher concentrations were seen in men and individuals with prevalent CVD. Aging contributes to increasing concentrations, given the pronounced changes seen with increasing age across all subgroups. These findings should be taken into consideration when applying cTnI decision thresholds in clinical settings.

KM Eggers, Lars Lind, Per Venge and Bertil Lindahl. Factors Influencing the 99th Percentile of Cardiac Troponin I Evaluated in Community-Dwelling Individuals at 70 and 75 Years of Age/. Clin Chem 2013.

Background: Atrial natriuretic peptide (ANP) has antihypertrophic and antifibrotic properties that are relevant to AF substrates. The −G664C and rs5065 ANP single nucleotide polymorphisms (SNP) have been described in association with clinical phenotypes, including hypertension and left ventricular hypertrophy. A recent study assessed the association of early AF and rs5065 SNPs in low-risk subjects. In a Caucasian population with moderate-to-high cardiovascular risk profile and structural AF, we conducted a case-control study to assess whether the ANP −G664C and rs5065 SNP associate with nonfamilial structural AF.
Methods: 168 patients with nonfamilial structural AF and 168 age- and sex-matched controls were recruited. The rs5065 and −G664C ANP SNPs were genotyped.
Results: The study population had a moderate-to-high cardiovascular risk profile with 86% having hypertension, 23% diabetes, 26% previous myocardial infarction, and 23% left ventricular systolic dysfunction. Patients with AF had greater left atrial diameter (44 ± 7 vs. 39 ± 5 mm; P , 0.001) and higher plasma NTproANP levels (6240 ± 5317 vs. 3649 ± 2946 pmol/mL; P , 0.01). Odds ratios (ORs) for rs5065 and −G664C gene variants were 1.1 (95% confidence interval [CI], 0.7–1.8; P = 0.71) and 1.2 (95% CI, 0.3–3.2; P = 0.79), respectively, indicating no association with AF. There were no differences in baseline clinical characteristics among carriers and noncarriers of the −664C and rs5065 minor allele variants.
Conclusions: We report lack of association between the rs5065 and −G664C ANP gene SNPs and AF in a Caucasian population of patients with structural AF. Further studies will clarify whether these or other ANP gene variants affect the risk of different subphenotypes of AF driven by distinct pathophysiological mechanisms.

P Francia, A Ricotta, A Frattari, R Stanzione, A Modestino, et al.
Atrial Natriuretic Peptide Single Nucleotide Polymorphisms in Patients with Nonfamilial Structural Atrial Fibrillation.
Clinical Medicine Insights: Cardiology 2013:7 153–159

Cystatin C and eGFR predict AMI or CVD mortality

BACKGROUND: The estimated glomerular filtration rate (eGFR) independently predicts cardiovascular death or myocardial infarction (MI) and can be estimated by creatinine and cystatin C concentrations. We evaluated 2 different cystatin C assays, alone or combined with creatinine, in patients with acute coronary syndrome.
METHODS: We analyzed plasma cystatin C, measured with assays from Gentian and Roche, and serum creatinine in 16 279 patients from the PLATelet Inhibition and Patient Outcomes (PLATO) trial. We evaluated Pearson correlation and agreement (Bland–Altman) between methods, as well as prognostic value in relation to cardiovascular death or MI during 1 year of follow up by multivariable logistic regression analysis including clinical variables, biomarkers, c-statistics, and relative integrated discrimination improvement (IDI).
RESULTS: Median cystatin C concentrations (interquartile intervals) were 0.83 (0.68–1.01) mg/L (Gentian) and 0.94 (0.80–1.14) mg/L (Roche). Overall correlation was 0.86 (95% CI 0.85–0.86). The level of agreement was within 0.39 mg/L (2 SD) (n = 16 279).
The areas under the curve (AUCs) in the multivariable risk prediction model with cystatin C (Gentian, Roche) or Chronic Kidney Disease Epidemiology Collaboration eGFR (CKD-EPI) added were 0.6914, 0.6913, and 0.6932. Corresponding relative IDI values were 2.96%, 3.86%, and 4.68% (n = 13 050). Addition of eGFR by the combined creatinine–cystatin C equation yielded AUCs of 0.6923 (Gentian) and 0.6924 (Roche) with relative IDI values of 3.54% and 3.24%.
CONCLUSIONS: Despite differences in cystatin C concentrations, overall correlation between the Gentian and Roche assays was good, while agreement was moderate. The combined creatinine–cystatin C equation did not outperform risk prediction by CKD-EPI.
A Åkerblom, L Wallentin, A Larsson, A Siegbahn, et al.
Cystatin C– and Creatinine-Based Estimates of Renal Function and Their Value for Risk Prediction in Patients with Acute Coronary Syndrome: Results from the PLATelet Inhibition and Patient Outcomes (PLATO) Study.

T2Dm has many subphenotypes in the prediabetic phase

For decades, glucose, hemoglobin A1c, insulin, and C peptide have been the laboratory tests of choice to detect and monitor diabetes. However, these tests do not identify individuals at risk for developing type 2 diabetes (T2Dm) (so-called prediabetic individuals and the subphenotypes therein), which would be a prerequisite for individualized prevention. Nor are these parameters suitable to identify T2Dm subphenotypes, a prerequisite for individualized therapeutic interventions. The oral glucose tolerance test (oGTT) is still the only means for the early and reliable identification of people in the prediabetic phase with impaired glucose tolerance (IGT). This procedure, however, is very time-consuming and expensive and is unsuitable as a screening method in a doctor′s office. Hence, there is an urgent need for innovative laboratory tests to simplify the early detection of alterations in glucose metabolism.
The search for diabetic risk genes was the first and most intensively pursued approach for individualized diabetes prevention and treatment. Over the last 20 years cohorts of tens of thousands of people have been analyzed, and more than 70 susceptibility loci associated with T2Dm and related metabolic traits have been identified. But despite extensive replication, no susceptibility loci or combinations of loci have proven suitable for diagnostic purposes.
Why did the genomic studies fail? One reason might be that T2Dm is a polygenetic disease, but there is another more important reason. The large diabetes cohorts investigated in these studies were very heterogeneous, consisting of poorly characterized individuals who were usually selected because they had an increase in blood glucose. Subsequently it has become clear that many different subphenotypes already exist in the prediabetic phase.
Metabolomics represents a new potential approach to move the diagnosis of diabetes beyond the application of the classical diabetic laboratory tests.
Rainer Lehmann. Diabetes Subphenotypes and Metabolomics: The Key to Discovering Laboratory Markers for Personalized Medicine?

Ca2+/calmodulin-dependent protein kinase II (CaMKII) has recently emerged as a ROS activated proarrhythmic signal

Background—Atrial fibrillation is a growing public health problem without adequate therapies. Angiotensin II (Ang II) and reactive oxygen species (ROS) are validated risk factors for atrial fibrillation (AF) in patients, but the molecular pathway(s) connecting ROS and AF is unknown. The Ca2+/calmodulin-dependent protein kinase II (CaMKII) has recently emerged as a ROS activated proarrhythmic signal, so we hypothesized that oxidized CaMKII􀄯(ox-CaMKII) could contribute to AF.
Methods and Results—We found ox-CaMKII was increased in atria from AF patients compared to patients in sinus rhythm and from mice infused with Ang II compared with saline. Ang II treated mice had increased susceptibility to AF compared to saline treated WT mice, establishing Ang II as a risk factor for AF in mice. Knock in mice lacking critical oxidation sites in CaMKII􀄯 (MM-VV) and mice with myocardial-restricted transgenic over-expression of methionine sulfoxide reductase A (MsrA TG), an enzyme that reduces ox-CaMKII, were resistant to AF induction after Ang II infusion.
RyR and Ca+ release from SR
ANS-   autonomic innervation of heart
mongillo_fig1  regulation of cardiac Ca++ cycling by ANS
jce561317.fig3    cardiac contraction
serum levels of MAA differentiated stable CAD from MI. For IgM antibodies to MAA, results were consistent with IgGantibodies to MAA
Conclusions—Our studies suggest that CaMKII is a molecular signal that couples increased ROS with AF and that therapeutic strategies to decrease ox-CaMKII may prevent or reduce AF.
Key words: atrial fibrillation, calcium/calmodulin-dependent protein kinase II, angiotensin II, reactive oxygen species, arrhythmia (mechanisms)
A Purohit, AG Rokita, X Guan, B Chen, et al.  Oxidized CaMKII Triggers Atrial Fibrillation.  Circulation. Sep 12, 2013;

Microparticles (MP)s give clues about vascular endothelial injury

BACKGROUND: Endothelial dysfunction is an early event in the development and progression of a wide range of cardiovascular diseases. Various human studies have identified that measures of endothelial dysfunction may offer prognostic information with respect to vascular events. Microparticles (MPs) are a heterogeneous population of small membrane fragments shed from various cell types. The endothelium is one of the primary targets of circulating MPs, and MPs isolated from blood have been considered biomarkers of vascular injury and inflammation.
CONTENT: This review summarizes current knowledge of the potential functional role of circulating MPs in promoting endothelial dysfunction. Cells exposed to different stimuli such as shear stress, physiological agonists, proapoptotic stimulation, or damage release MPs, which contribute to endothelial dysfunction and the development of cardiovascular diseases. Numerous studies indicate that MPs may trigger endothelial dysfunction by disrupting production of nitric oxide release from vascular endothelial cells and subsequently modifying vascular tone. Circulating MPs affect both proinflammatory and proatherosclerotic processes in endothelial cells. In addition, MPs can promote coagulation and inflammation or alter angiogenesis and apoptosis in endothelial cells.
SUMMARY: MPs play an important role in promoting endothelial dysfunction and may prove to be true biomarkers of disease state and progression.
Fina Lovren and Subodh Verma.  Evolving Role of Microparticles in the Pathophysiology of Endothelial Dysfunction.
Outcomes of STEMI and NSTEMI different predicted by NPs after MI
Patients with increased blood concentrations of natriuretic peptides (NPs) have poor cardiovascular outcomes after myocardial infarction (MI). Data from 41 683 patients with non–ST-segment elevation MI (NSTEMI) and 27 860 patients with ST-segment elevation MI (STEMI) at 309 US hospitals were collected as part of the ACTION Registry®–GWTG™ (Acute Coronary Treatment and Intervention Outcomes Network Registry–Get with the Guidelines) (AR-G) between July 2008 and September 2009.

B-type natriuretic peptide (BNP) or N-terminal pro-BNP (NT-proBNP) was measured in 19 528 (47%) of NSTEMI and 9220 (33%) of STEMI patients. Patients in whom NPs were measured were older and had more comorbidities, including prior heart failure or MI. There was a stepwise increase in the risk of in-hospital mortality with increasing BNP quartiles for both NSTEMI (1.3% vs 3.2% vs 5.8% vs 11.1%) and STEMI (1.9% vs 3.9% vs 8.2% vs 17.9%). The addition of BNP to the AR-G clinical model improved the C statistic from 0.796 to 0.807 (P < 0.001) for NSTEMI and from 0.848 to 0.855 (P = 0.003) for STEMI. The relationship between NPs and mortality was similar in patients without a history of heart failure or cardiogenic shock on presentation and in patients with preserved left ventricular function.

NPs are measured in almost 50% of patients in the US admitted with MI and appear to be used in patients with more comorbidities. Higher NP concentrations were strongly and independently associated with in-hospital mortality in the almost 30 000 patients in whom NPs were assessed, including patients without heart failure.

BM Scirica, MB Kadakia, JA de Lemos, MT Roe, DA Morrow, et al. Association between Natriuretic Peptides and Mortality among Patients Admitted with Myocardial Infarction: A Report from the ACTION Registry®–GWTG™.

Predictive value of processed forms of BNP in circulation

B-type natriuretic peptide (BNP) is secreted in response to pathologic stress from the heart. Its use as a biomarker of heart failure is well known; however, its diagnostic potential in ischemic heart disease is less explored. Recently, it has been reported that processed forms of BNP exist in the circulation. We characterized processed forms of BNP by a newly developed mass spectrometry–based detection method combined with immunocapture using commercial anti-BNP antibodies.

Measurements of processed forms of BNP by this assay were found to be strongly associated with presence of restenosis. Reduced concentrations of the amino-terminal processed peptide BNP(5–32) relative to BNP(3–32) [as the index parameter BNP(5–32)/BNP(3–32) ratio] were seen in patients with restenosis [median (interquartile range) 1.19 (1.11–1.34), n = 22] vs without restenosis [1.43 (1.22–1.61), n = 83; P < 0.001] in a cross-sectional study of 105 patients undergoing follow-up coronary angiography. A sensitivity of 100% to rule out the presence of restenosis was attained at a ratio of 1.52. Processed forms of BNP may serve as viable potential biomarkers to rule out restenosis.

H Fujimoto, T Suzuki, K Aizawa, D Sawaki, J Ishida, et al. Processed B-Type Natriuretic Peptide Is a Biomarker of Postinterventional Restenosis in Ischemic Heart Disease. Clin Chem 2013.

Circulating proteins from patients requiring revascularization

More than a million diagnostic cardiac catheterizations are performed annually in the US for evaluation of coronary artery anatomy and the presence of atherosclerosis. Nearly half of these patients have no significant coronary lesions or do not require mechanical or surgical revascularization. Consequently, the ability to rule out clinically significant coronary artery disease (CAD) using low cost, low risk tests of serum biomarkers in even a small percentage of patients with normal coronary arteries could be highly beneficial. METHODS: Serum from 359 symptomatic subjects referred for catheterization was interrogated for proteins involved in atherogenesis, atherosclerosis, and plaque vulnerability. Coronary angiography classified 150 patients without flow-limiting CAD who did not require percutaneous intervention (PCI) while 209 required coronary revascularization (stents, angioplasty, or coronary artery bypass graft surgery). Continuous variables were compared across the two patient groups for each analyte including calculation of false discovery rate (FDR [less than or equal to]1%) and Q value (P value for statistical significance adjusted to [less than or equal to]0.01).

Significant differences were detected in circulating proteins from patients requiring revascularization including increased apolipoprotein B100 (APO-B100), C-reactive protein (CRP), fibrinogen, vascular cell adhesion molecule 1 (VCAM-1), myeloperoxidase (MPO), resistin, osteopontin, interleukin (IL)-1beta, IL-6, IL-10 and N-terminal fragment protein precursor brain natriuretic peptide (NT-pBNP) and decreased apolipoprotein A1 (APO-A1). Biomarker classification signatures comprising up to 5 analytes were identified using a tunable scoring function trained against 239 samples and validated with 120 additional samples. A total of 14 overlapping signatures classified patients without significant coronary disease (38% to 59% specificity) while maintaining 95% sensitivity for patients requiring revascularization. Osteopontin (14 times) and resistin (10 times) were most frequently represented among these diagnostic signatures. The most efficacious protein signature in validation studies comprised osteopontin (OPN), resistin, matrix metalloproteinase 7 (MMP7) and interferon gamma (IFNgamma) as a four-marker panel while the addition of either CRP or adiponectin (ACRP-30) yielded comparable results in five protein signatures.

Proteins in the serum of CAD patients predominantly reflected

  1. a positive acute phase, inflammatory response and

  2. alterations in lipid metabolism, transport, peroxidation and accumulation.

    There were surprisingly few indicators of growth factor activation or extracellular matrix remodeling in the serum of CAD patients except for elevated OPN. These data suggest that many symptomatic patients without significant CAD could be identified by a targeted multiplex serum protein test without cardiac catheterization thereby eliminating exposure to ionizing radiation and decreasing the economic burden of angiographic testing for these patients.

WA Laframboise, R Dhir, LA Kelly, P Petrosko, JM Krill-Burger, et al. Serum protein profiles predict coronary artery disease in symptomatic patients referred for coronary angiography.
BMC Medicine (impact factor: 6.03). 12/2012; 10(1):157.

miRNAs in CAD

MicroRNAs are small RNAs that control gene expression. Besides their cell intrinsic function, recent studies reported that microRNAs are released by cultured cells and can be detected in the blood. To address the regulation of circulating microRNAs in patients with stable coronary artery disease. To determine the regulation of microRNAs, we performed a microRNA profile using RNA isolated from n=8 healthy volunteers and n=8 patients with stable coronary artery disease that received state-of-the-art pharmacological treatment. Interestingly, most of the highly expressed microRNAs that were lower in the blood of patients with coronary artery disease are known to be expressed in endothelial cells (eg, miR-126 and members of the miR-17 approximately 92 cluster). To prospectively confirm these data, we detected selected microRNAs in plasma of 36 patients with coronary artery disease and 17 healthy volunteers by quantitative PCR. Consistent with the data obtained by the profile, circulating levels of miR-126, miR-17, miR-92a, and the inflammation-associated miR-155 were significantly reduced in patients with coronary artery disease compared with healthy controls. Likewise, the smooth muscle-enriched miR-145 was significantly reduced. In contrast, cardiac muscle-enriched microRNAs (miR-133a, miR-208a) tend to be higher in patients with coronary artery disease. These results were validated in a second cohort of 31 patients with documented coronary artery disease and 14 controls. Circulating levels of vascular and inflammation-associated microRNAs are significantly downregulated in patients with coronary artery disease.

S Fichtlscherer, S De Rosa, H Fox, T Schwietz, A Fischer, et al. Circulating microRNAs in patients with coronary artery disease. Circulation Research 09/2010; 107(5):677-84.

Imaging modalities compared

This review compares the noninvasive anatomical imaging modalities of coronary artery calcium scoring and coronary CT angiography to the functional assessment modality of MPI in the diagnosis and prognostication of significant CAD in symptomatic patients. A large number of studies investigating this subject are analyzed with a critical look on the evidence, underlying the strengths and limitations. Although the overall findings of the presented studies are favoring the use of CT-based anatomical imaging modalities over MPI in the diagnosis and prognosticating of CAD, the lack of a high number of large- scale, multicenter randomized controlled studies limits the generalizability of this early evidence. Further studies comparing the short- and long-term clinical outcomes and cost-effectiveness of these tests are required to determine their optimal role in the management of symptomatic patients with suspected CAD.

Y Hacioglu, M Gupta, Matthew J Budoff. Noninvasive anatomical coronary artery imaging versus myocardial perfusion imaging: which confers superior diagnostic and prognostic information?
Journal of computer assisted tomography 34(5):637-44.

Three Dimensional In-Room Imaging (3DCA) in PCI

Introduction: Coronary angiography is a two-dimensional (2D) imaging modality and thus is limited in its ability to represent complex three-dimensional (3D) vascular anatomy. Lesion length, bifurcation angles/lesions, and tortuosity are often inadequately assessed using 2D angiography due to vessel overlap and foreshortening. 3D Rotational Angiography (3DRA) with subsequent reconstruction generates models of the coronary vasculature from which lesion length measurements and Optimal View Maps (OVM) defining the amount of vessel foreshortening for each gantry angle can be derived. This study sought to determine if 3DRA-assisted percutaneous coronary interventions resulted in improved procedural results by minimizing foreshortening and optimizing stent selection.
 Rotational angiographic acquisitions were performed and a 3D model was generated from two images greater than 30° apart. An optimal view map identifying the least amount of vessel foreshortening and overlap was derived from the 3D model.
The clinical validation of in-room image-processing tools such as 3DCA and optimal view maps is important since FDA approval of these tools does not require the presentation of any data on clinical experience and impact on clinical outcomes. While the technology of 3DRA and optimal view calculations has been well validated by the work of Chen and colleagues, this study is important in demonstrating how clinical care may be impacted [4,5,7]. This study was biased toward minimizing the impact of these tools on clinical decision-making since the study site, cardiologists, and staff have extensive experience in rotational angiography, 3-D modeling and reconstruction, and the impact of foreshortening on the assessment of lesion length and choice of stent size.
3DRA assistance significantly reduced target vessel foreshortening when compared to operator’s choice of working view for PCI (2.99% ± 2.96 vs. 9.48% ± 7.56, p=0.0001). The operators concluded that 3DRA recommended better optimal view selection for PCI in 14 of 26 (54%) total cases. In 9 (35%) of 26 cases 3DRA assistance facilitated stent positioning. 3DRA based imaging prompted stent length changes in 4/26 patients (15%).
MH. Eng, PA Hudson, AJ Klein, SYJ Chen, … , JA Garcia. Impact of Three Dimensional In-Room Imaging (3DCA) in the Facilitation of Percutaneous Coronary Interventions. J Cardio Vasc Med 2013; 1: 1-5.


Related References from

Genomics & Genetics of Cardiovascular Disease Diagnoses: A Literature Survey of AHA’s Circulation Cardiovascular Genetics, 3/2010 – 3/2013
Curators: Aviva Lev-Ari, PhD, RN and Larry H. Bernstein, MD, FCAP…cs-32010-32013/

Prognostic Marker Importance of Troponin I in Acute Decompensated Heart Failure (ADHF)
Larry H Bernstein and  Aviva Lev-Ari…-heart-failure

A Changing expectation from cardiac biomarkers.
Larry H Bernstein…ith-biomarkers/

Dealing with the Use of the High Sensitivity Troponin (hs cTn) Assays
Larry H Bernstein and Aviva Lev-Ari…-hs-ctn-assays/

For Disruption of Calcium Homeostasis in Cardiomyocyte Cells, see

Part VI: Calcium Cycling (ATPase Pump) in Cardiac Gene Therapy: Inhalable Gene Therapy for Pulmonary Arterial Hypertension and Percutaneous Intra-coronary Artery Infusion for Heart Failure: Contributions by Roger J. Hajjar, MD

Aviva Lev-Ari, PhD, RN

Part VII: Cardiac Contractility & Myocardium Performance: Ventricular Arrhythmias and Non-ischemic Heart Failure – Therapeutic Implications for Cardiomyocyte Ryanopathy (Calcium Release-related Contractile Dysfunction) and Catecholamine Responses

Justin Pearlman, MD, PhD, FACC, Larry H Bernstein, MD, FCAP and Aviva Lev-Ari, PhD, RN

Part VIII: Disruption of Calcium Homeostasis: Cardiomyocytes and Vascular Smooth Muscle Cells: The Cardiac and Cardiovascular Calcium Signaling Mechanism

Justin Pearlman, MD, PhD, FACC, Larry H Bernstein, MD, FCAP and Aviva Lev-Ari, PhD, RN

Read Full Post »

Regression: A richly textured method for comparison and

classification of predictor variables

 The multivariable Case

Larry H. Bernstein, MD


Keywords:  bias correction, chi square, linear regression, logistic regression, loglinear analysis, multivariable regression, normal distribution, odds ratio, ordinal regression, regression methods


Multivariate statistical analysis is used to extend this analysis to two or more predictors.   In this case a multiple linear regression or a linear discriminant function would be used to predict a dependent variable from two or more independent variables.   If there is linear association dependency of the variables is assumed and the test of hypotheses requires that the variances of the predictors are normally distributed.  A method using a log-linear model circumvents the problem of the distributional dependency in a method called ordinal regression.    There is also a relationship of analysis of variance, a method of examining differences between the means of  two or more groups.  Then there is linear discriminant analysis, a method by which we examine the linear separation between groups rather than the linear association between groups.  Finally, the neural network is a nonlinear, nonparametric model for classifying data with several variables into distinct classes. In this case we might imagine a curved line drawn around the groups to divide the classes. The focus of this discussion will be  the use of linear regression  and explore other methods for classification purposes.


Multivariate statistical analysis extends regression analysis and introduces combinatorial analysis for two or more predictors.   Multiple linear regression or a linear discriminant function would be used to predict a dependent variable from two or more independent variables.   If there is linear association dependency of the variables is assumed and the test of hypotheses requires that the variances of the predictors are normally distributed.  Linear discriminant analysis examines the linear separation between groups rather than the linear association between groups, and it also requires adherence to distributional assumption. There is also a relationship of analysis of variance as a special case of linear regression, a method of examining differences between the means of two or more groups. A method using a log-linear model circumvents the problem of the distributional dependency in a method called ordinal regression.  Finally, the neural network is a nonlinear, nonparametric model for classifying data with several variables into distinct classes. In this case we might imagine a curved line drawn around the groups to divide the classes.

Regression analysis.

The use of linear regression, linear discriminant analysis and analysis of variance has to meet the following assumptions:

The variables compared are assumed to be independent measurements.

The correlation coefficient is a useful measure of the strength of the relationship between two variables only when the variables are linearly related.

The correlation coefficient is bounded in absolute value by 1.

All points on a straight line imply correlation of 1.

Correlation of 1 implies all points are on a straight line.

A high correlation coefficient between two variables does not imply a direct effect of one variable on another (both may be influenced by a hidden explanatory variable).

The correlation coefficient is invariant to linear transformations of either variable.

The correlation coefficient is also expressed as the covariance or product of the deviations of X and Y from their means standardized by dividing by the respective standard deviations.

These assumptions may be valid if the amount of data compared is very large, and if the data is parametric.  This is not necessarily the case.  There are also special applications in laboratory evaluations and crossover studies between methods and instruments that require correction for bias or for differences in the error variance term.

How do we measure the difference if there is any?  We use the t-test (19, 21).   If t is small than the null hypothesis is satisfied and no difference is detected in the means.   The conclusion is that the null hypothesis is accepted and the means are essentially the same .  However, the ability to accept or reject the null hypothesis is dependent on sample size, or power.  If the null hypothesis is rejected, bias has to be suspected.  This is useful when analyzing certain data, where the results of OLS are unsatisfactory. This test is here applied to linear increasing values of Y on X measured by A and B methods.   Of course the measurements are plotted and a line is fitted to the scatterplot.   OLS gives the fit of the line based on the least squares error, where the slope of the line is given by (20,22).

B = å (xi – mean x)yi .

å(xi – mean x)2

It is assumed that there are n pairs of values of x and y, and xi and yi denote the ith pair of values.   The slope defines the regression line of y on x.  An intercept that differs from zero is the bias.  It is worthwhile to mention that there is a difference here between the correlation measurement and the least squares fit of y on x.   We are measuring X by methods A and B.  We can then determine that the is a linear association with r valued between 0 and 1 (-values excluded).  In the case of the regression model, we are predicting B from A by plotting B on y from A on x.   Of course, experimentally, we are expecting the prediction to hold over a range of measurements, and the agreement drops off at some value of the coordinates (xi, yi).

Multiple regression is an extension of linear regression where the dependent variable is predicted by several independent variables.

In this case, the extended equation is (23)

Y = b0 + b1x1 + b2x2 + b3x3 …bnxn.

The model assumes a linear relationship between many predictor variables and the dependent variable.  The model usually assumes that the independent variables are not correlated with each other, which may not be the case.  The model can be tested by stepwise removal of predictor variables to assess their contribution to the model.     The model is  considered to be parametric, and so it requires that the inputs are normally distributed.  The bs (or betas) are also partial correlation coefficients.  The partial F test is the measure of the contribution of each variable after all the variables are in the equation.

Figures 1-3 are scatterplots of eGFR (glomerular filtration rate calculated by MDRD equation) and of hemoglobin vs Nt-proBNP, and a boxplot of Nt-proBNP by WHO criteria for anemia. Figure 2 is a 3D plot of NT-proBNP spliced by eGFR and hemoglobin.  The linear regression model is presented in Table I.  The correlation coefficient (R ) for the model is weak, but not insignificant. What do you think is the effect of the large variance in the dependent variable?   Figure 3 is a 3D plot of the eGFR and hemoglobin vs a transformed variable – age normalized 1000*Log(Nt-proBNP)/eGFR.  The variance is reduced on the transformed variable.  Table II is the regression model on the data.  The correlation coefficient R is improved.

Analysis of variance (ANOVA) and Analysis of covariance (ANCOVA)

ANOVA is used if the dependent variable is continuous and all of the independent variables are categorical.   One-way ANOVA is used for a single independent variable, and multi-way ANOVA is used for multiple independent variables.   The ANOVA is based on the general linear model.   The F-test is used to compare the difference between the means of groups.   The independent variable has discrete values is not used as a measure.   The t-test can be used between each pair in the groups.   The goal of ANOVA is to explain the total variation found in the study.   An example of this application is shown in Figure 4.

Figure 4.  BNP determined within ejection fraction above or within 40

Figure 5 is the means and 95% confidence intervals for a comparison of D-dimer and positive or negatine venous duplex scans.  There are only two variables so the corresponding ANOVA is one-way.  The F-value is high and corresponds to a high t in the t-test.  F is the same as t2 and p = 0.0001 (Table III).  Our interest here is in multiple variables so we’ll hold the discussion of difference testing between two variables.

Figure 5.

Table III.

If some of the independent variables are categorical (nominal, ordinal or dichotomous) and some are continuous ANCOVA is used.   The ANCOVA procedure first adjusts the dependent variable on the basis of the continuous independent variable and then does ANOVA on the adjusted dependent variable.

Generalized linear and generalized additive models

Generalized linear models transform the response by assuming that a transformation of the expected response is a linear function of the predictor variables.   The variance of the response is a function of the mean response.   When the relationship between the parameters is not linear, a generalized linear model can’t be used.   A generalized additive model can be used to fit nonlinear data-dependent functions of the predictor.   Tree-based models are used for exploratory analysis and are related to clustering, which is a method for studying the structure of the data, creating clusters of data with similar characteristics.

Discriminant analysis

The discriminant analysis is a modification of the general linear regression model.   The method is used to assign data to any of distinct classes as the dependent variable.   The linear regression model predicts based on a linear relationship between the dependent and the independent variables.   They are codependent.   In the discriminant function they are independent.   The function determines a separation between the classes to which the data assigns patients.   The goal is to assign a new incoming patient based on the independent variables to one of the different groups.   The mathematical function can be linear, quadratic, or another function.   The stepwise linear regression with removal or addition of variables is viewed in the same way.   However, the discriminant function produces a separation between the classes rather than through them.  The same qualifications for the method fit pertaining to distributional assumptions that applies to multiple linear regression applies to the linear discriminant function, but the analysis of data on congestive heart failure, renal insufficiency and anemia partitioned with NT-proBNP, creatinine, age and hemoglobin concentration shown in Figure 6 and Table IV uses a quadratic equation.  I re-classify the data using the transformed variable age-normalized 1000*Log(NT proBNP)/eGFR presented in Figure 7 and Table V.  The use of the logarithmic transform and removal of age and hemoglobin as predictors give impressive results.

Figure  6.

Table IV.

Figure 7.

Table V.

Mahalanobis D2

The euclidean distance between two coordinates having the position (x1y1), (x2y2) is given by the distance D = ([x1 – x2]2 + [y1 – y2]2)1/2.   This is generalized for N-dimensional space, and the square of the distance is D2.   The two points are the centroids in a cloud of points in space separated by D, the euclidean distance between the points in an N-dimensional space.   The multiplication of a vector and a variance-covariance matrix T-1 yields the linear discriminant functions.   The Mahalanobis distance can be used to evaluate the distances of centroids and also the distances of objects towards the centroid of their class.

Logistic regression

The linear probability model (logistic regression) is the standard regression model applied to data for which the dependent variable is dichotomous (0,1). It fits a logistic function to the dependent variables valued at 0 or 1 and estimates the probabilities associated with each observation (24).  The predicted values from the model are interpreted as a probability that the response is a 1.  The test of significance of the model is the Maximum Likelihood Estimator (MLE).  The significance is determined by adjusting the parameters to maximize the likelihood of the observed data arising from the linear sum of the variables.

There are problems in using the linear probability model (49).

The residuals don’t have a constant variance so that estimates from regression are not best linear unbiased, therefore, not minimum variance.

Standard errors of regression coefficients can be erroneous giving invalid confidence intervals.

The predicted values from regression can range outside the interval [0,1], whereas probabilities are bounded by that interval

The linearity assumption inherently imposes constraints on the marginal effects of predictor variables that are not taken into account by the OLS estimation.

The linearity assumption implies that the marginal effect of a predictor is constant across its range.

The usual r squared measure is problematic.

Ordinal regression

I now turn to the application of a special nonparametric regression program developed by Jay Magidson (GOLDmineR; Statistical Innovations Inc., Belmont, MA), referred to as Ordinal regression, or universal regression (25-28).   Let’s look at the application of this tool, which makes outcomes analysis easy.   This method brings a powerful tool to the analysis of laboratory data for clinical validation of diagnostic tests.  It overcomes serious limitations of logistic analysis when there is more than two possible outcomes to consider.   This has become more important as we introduce tests that have results that are affected by morbid conditions so that a range of probabilities might be associated with scaled “dummy values” of the test (possibly because of hidden or unspecified variables).

Ordinal dependent variables are multivalued and have an ordered relationship to the predictor variable(s).   Magidson (25-28), inspired by the work of Leo Goodman (29,30), suggests the existence of a single regression model that can accomodate dependent variables of any metric – dichotomous, ordinal, or continuous.   This supermodel holds true under the assumption of bivariate normality and under other distributional assumptions and subsumes linear distribution and logistic regression as special cases (25).    It uses a log odds model fit and the odds ratio is obtained from the log(odds ratio).   In the linear probability model, the coefficients (bi) are partial correlation coefficients.   In the logit model the coefficients are partial log(odds-ratio).

The monotonic regression of X on Y is described by:


E(Y|X = x) = å   Pj.x yj


Where Pj.x, the conditional probability of the occurrence of Y=yj (an ordinal dependent variable) given X=x (qualitative or quantitative predictor variables), is estimated from a sample of N observations using 2 steps.

1)      Conditional logits Yj.x are predicted using the generalized logit model, where Yj.x*  is: Yj.x = aj + (b1x1* +  b2x2*  + bMxM*)yj*    j= 1,2,…, J.
The Y-scores, which determine the ordering and relative spacing of the J outcomes, may be specified or if unspecified, they are treated as model parameters and estimated with other parameters.   Yj* , the relative Y-score, is the difference between yj and some Y-reference score y0 defined as a weighted average of the original Y-scores.

2)      The predicted logits are transformed to predicted probabilities using the identity:


Pj.x º exp(Yj.x)/å exp(Yj.x)


For a given X=x, the generalized logit is defined as

Yj.x º ln(Pj.x/P0.x)

where Pj.x is the conditional probability of the jth outcome occurring when X=x


and P0.x =  P (Pj.x)ej


I performed a nonparametric regression using the universal regression program GOLDminer, developed by Jay Magidson (25-28) at Statistical Innovations, Belmont, MA.  The universal regression program is a logistic regression if the dependent variable is a binary outcome, and it is a polytomous regression if there are more than two dependent variables, but it can accommodate a paired comparison of covariates.  The measure of association is phi and R2.  The measure of fit is L2 (chi square).  The logarithmic form transforms into a probability model, which we aren’t concerned with here.

Graphical Ordinal Logit Display (GOLDminer)

I have mentioned the nonparametric universal regression of Magidson (25-28), based on work with log-linear modeling with Prof. Leo Goodman (29,30).  The logistic regression and linear regression models can be viewed as special cases of this more general model.  This regression model has greatest use for examining structure in data where there are more than two dependent variables, and the independent variables are scaled to intervals (25-28).  The model is more general than the logistic regression and is not constrained by the conditions encountered with logistic regression identified above.

I cite a number of publications of its use in clinical laboratory outcomes analysis.

Example  1.  The association between predictors of nutrition risk and malnutrition risk

I use here data obtained by Linda Brugler and coworkers at St.FrancisHospital in Wilmington, DE (31) that examines association between the malnutrition assessed before intervention with three predictors of malnutrition risk.  Poor oral intake and malnutrition related diagnosis are categorical, and the laboratory-derived serum albumin is scaled to form an ordinal predictor.   The strength of the predictors is given by Table VI:

Table VI.  Ordinal regression model for combined 3 predictors of malnutrition risk.

The model is defined by the following:  L2 = 267.68, R2 = 0.405, phi = 1.1134,

Df (3, 42), p = 9.7e-58.

Example 2:  Ordinal regression for thalassemia risk

Table VII shows the odds-ratios for the combinatorial scaled results of Mentzer score (ratio of MCV: red cell count), MCV, and Hgb A2(e)(by electrophoresis is higher than by HPLC).  The presence of only a single positive test gives an unlikely result for thalassemia, while two or more positive tests give a high likelihood of thalassemia.   This is summarized as follows: 0,0,0-0,0,1-0,1,0-1,0,0 = 0; 1,1,1-1,0,1-0,1,1-1,1,0 = 1.

Table VIII.   Expected Odds Ratios – Diagnosis Thalassemia

Example 3. Ordinal regression for risk of newborn respiratory distress syndrome

A study by Kaplan, Chapman and coworkers (32) extending work by Bernstein and Rundell (33) looked at the relationship between gestational age and RDS of the newborn and used the ordinal regression model to predict expected outcomes (33).  Table IX gives probabilities for the prediction of risk.

Table IX.   Probabilities of RDS given by gestational age and S/A ratio.

Example 4.  Prediction of myocardial infarction risk by EKG and troponin T at 0.1 ng/ml

Bernstein, Zarich and Qamar (34) carried out a study in which the physicians were blinded to the troponin T results.  A randomized prospective study of over 800 patients followed (35-37).  The chest pain characteristics, EKG findings and troponin T results were reviewed for consecutive patients entered into the study (34).   EKG results were scaled as: negative, nonspecific, 0; ST depression or T wave inversion, 1, ST elevation or new Q-wave, 2.  Troponin T was scaled as follows: 0-0.075 ng/ml, 0; 0.076-0.099, 1; > 0.1.The diagnoses were as follows: noncardiac, cardiac and nonischemic, 1; Unstable angina with MI ruled out, 2; non ST or ST elevation MI, 3.  Table X is the table of odds ratios and probabilities.

Table X. Ordinal regression of EKG and troponin T on diagnoses

Ovarian Cancer Survival

Rosman and Schwartz have reported a relationship between CA125 post-chemotherapy of ovarian carcinomatosis and serum half-life of CA125.  We examined a published data set provided by Dr. Martin Rosman.  Data were analyzed from 55 women who were treated at YaleUniversity, had an evaluable CA125 half-life (t1/2), and were followed for disease recurrence for at least 3 years.  We modeled survival or remission for ovarian cancer using operative findings, stage, and CA125 halflife (46).  Figure 9 is a plot of the CA125 elimination half-life vs the Kullback-Liebler distance using the data provided by Dr. Martin Rosman. The K-L distance is the difference between the total entropy of the data in which association is removed and the observed entropy for each value of CA125.  The t1/2 is 10 days.  What Rudolph and Bernstein (43) have referred to as effective information is KL distance. This was done to determine the value of CA125 that best predicts survival.

Figure 9 CA125 halflife

The next step was to carry out a Kaplan Meier survival plot with Cox regression on the data vs the time to death or remission.  A survival of 30 months is considered a cure.  A survival less is considered a remission.  Some patients died only shortly into chemotherapy.   The study result is shown in Figure 11.

Figure 10.  Kaplan Meier plot

We also examined the associations between OPERATIVE FINDINGS and CA125 to REMISSION and NONREMISSION or RELAPSE using a universal regression model under bivariate normality with estimation of generalized odds-ratios developed by Jay Magidson (Statistical Innovations, Inc., Belmont, MA).  It uses a parallel log-odds model based on adjacent odds to describe the data.  The universal regression is carried out after scaling the continuous variables with intervals we determined as follows: halflife- 0-5, 6-10, 11-15, 16-20, >20.   A crosstabulation is constructed using the scaled variables as treatment vs. the effect (full, short remission or none), to obtain the frequency tabulation of treatment level vs remission, relapse or nonremission.

Table XI is a cross-tabulation of the observed and expected outcome frequencies in remission (rem), short remission (short,< 30 months) and non-remission (none) versus the scaled half-lives.   Relapse and failure to achieve remission were combined into one outcome class.  The means and standard error of the means (SEM) of half-life versus remission or non-remission/relapse are effectively separated (F=7.42, p < 0.01) as follows: Remission, 7.9, 2.8, [19];  Relapse/Non-remission, 17.4, 2.05, [36].

Table XII.  Observed and expected odds and odds-ratios of remission, relapse and no response by half-life

Perspective for the Future

Linear regression has been used extensively for methods comparison and for quality control, exclusively based on distributional assumptions and distance from the center of the population sample.   This is essential to analytical chemistry principles, but it has reached a limit.  The last 30 years has seen the development of very powerful regression tools that are not dependent on distributional assumptions and that move the method into classification and prediction.  The development of the Akaike Information Criterion (38-40) brought together two major disciplines that had separate developments, information theory and statistics.   The work by Bernstein et al. (41-42) in predicting myocardial infarction using bivariate density estimation, and with Kullback-Liebler Distance (43, 44), an extension of work by Rypka (45) is closely related. The use of tables and the scaling of data has been the dominant approach to statistics that uses ordinal and categorical data in outcomes research.  This has become a powerful method used in studies of placebo and drug effects.   The approach is readily amenable to studies of laboratory tests and outcomes.   Outcomes studies will be designed and carried out for laboratory tests that will ask questions appropriate for the clinical laboratory sciences, and that will not be subordinated to pharmaceutical evaluations, which currently have exclusion criteria that are inappropriate for laboratory investigations.


Regression has a long history in the development of modern science since the 18th century.  Regression has had a role in the emergence of physics, anthropology, psychology, and chemistry.  But its development was initially tied to linear association and assumption of normal distribution.   There are many associations that are tied to frequency of discrete events.  The use of chi-square as a measure of goodness of fit has such a tie to genetic analysis and to classification tables.   The importance of outcomes management and the recognition of a multivariable data structure that needs to be explored leads us to a new domain of regression models and includes an assumption that the dependent variable may not be know with certainty.  This is the case with the emerging models known as mixture models, structural equation models and latent class models.  This type of model is not traditionally a regression model and looks at defined variables and also unmeasured, hidden or latent variables (factors) in the model.  However, there are factor analysis and regression forms of the LCM that are included in the LCM software releases of Statistical Innovations, Inc. (Latent Gold). This important subject is beyond the scope of this review, but Demidenko (47) has written an excellent text on the subject.


19. Hoel PG. Elementary Statistics, Testing Hypotheses: The difference between two means. Chapter 3.3. pp133-117. 1960. Wiley, New York.

20. Hoel PG. Ibid. Regression. Chapter 9. pp141-153.

21. Norman GR, Streiner DL. Biostatistics: The Bare Essentials. Two repeated observations: The paired t-test and alternatives. Chapter 10. pp89-93. 2000, BC Deckker, Hamilton, Ont., Canada.

22. Norman GR, Streiner DL. Ibid. Simple regression and correlation. Chapter 13. pp118-126.

23. Norman GR, Streiner DL. Ibid. Multiple regression. Chapter 14. pp127-137.

24. Norman GR, Streiner DL.Ibid. Logistic regression. Chapter 15. pp139-144.

25.  Magidson J.  “Multivariate Statistical Models for Categorical Data,” Chapters 3 & 4   in Bagozzi R, Advanced Methods of Marketing Research, Blackwell, 1994.

26.  Magidson J. Introducing a new graphical method for the analysis of an ordered categorical response – Part I. Journal of Targeting, Measurement and Analysis for Marketing (UK). 1995; IV(2):133-148.

27.  Magidson J.  Introducing a new graphical model for the analysis of  an ordered categorical response – Part II. Ibid. 1996;IV(3):214-227.

28.  Magidson J.  Maximum likelihood assessment of clinical trials based on an ordered categorical response. Drug information Journal. 1996;30:143-170.

29.   Goodman LA.  Simple models for the analysis of associations in cross-  classifications having ordered categories.  Journal of the American Statistical Association. 1979;74: 537-552.  Reprinted in The Analysis of Cross-Classified Data Having Ordered Categories. 1984, HarvardUniversity Press.

30. Goodman LA.  Association models and the bivariate normal for contingency tables with ordered categories. Biometrika 1981;68:347-355.

31.Brugler L, Stankovic AK, Schlefer M, Bernstein L. A simplified nutrition screen for hospitalized patients using readily available laboratory and patient information. Nutrition 2005;21:650-658.

32. Kaplan LA, Chapman JF, Bock JL, Santa Maria E, Clejan S, et al. Prediction of respiratory distress syndrome using the Abbott FLM-II amniotic fluid assay. Clin Chim Acta 2002;326[1-2]:61-68.

33.  Bernstein LH, Stiller R, Menzies C, McKenzie M, Rundell C. Amniotic fluid    polarization of fluorescence and lecithin/sphingomyelin ratio decision criteria assessed. Yale J Biol Med 1995; 68(2):101-117.

34.  Bernstein LH, Qamar A, McPherson C, Zarich S.   Evaluating a new graphical   ordinal logit method (GOLDminer) in the diagnosis of myocardial infarction utilizing clinical features and laboratory data.   Yale J Biol Med 1999; 72:259-268.

35. Bernstein L, Bradley K, Zarich S. GOLDmineR: Improving Models for Classifying Patients with Chest Pain. Yale J Biol Med 2002;75: 183-198.

36. Zarich S, Bradley K, Seymour J, Ghali W, Traboulsi A, et al. Impact of troponin T determinations on hospital resources and costs in the evaluation of patients with suspected myocardial ischemia. Amer J Cardiol 2001;88:732-6.

37. Zarich SW, Qamar AU, Werdmann MJ, Lizak LS, McPhersonCA, Bernstein LH. Value of a single troponin T at the time of presentation as compared to serial CK-MB determinations in patients with suspected myocardial ischemia. Clin Chim Acta 2002;326:185-192.

38. Akaike H. Information theory and an extension of maximum likelihood principle.    In B.N. Petrov and F. Csake (eds.), Second International Symposium on Information Theory. 1973, Akademiai Kiado, pp 267-281, Budapest.

39. Akaike H. A new look at the statistical model identification.  IEEE Transactions on Automation Control, AC-19, 1974; 716-723.

40. Dayton CM. Information Criteria for the Paired-Comparisons Problem.  American Statistician. 1998;52: 144-151.

41. Bernstein LH, Good IJ, Holtzman GI, Deaton ML, Babb J:  Diagnosis of myocardial infarction from two enzyme measurements of creatine kinase isoenzyme MB with use of nonparametric probability estimation.  Clin Chem 1989;35:444-7.

42. Bernstein LH, Good IJ, Holtzman GI, Deaton ML, and Babb J. Diagnosis of heart attack from two enzyme measurements by means of bivariate probability density estimation: statistical details. J Statistical Computation and Simulation. 1989.

43. Rudolph RA, Bernstein LH, Babb J. Information-induction for the diagnosis of myocardial infarction. Clin Chem 1988;34:2031-8.

44. Kullback S, Liebler RA. On information and sufficiency. Ann Mathematical Statistics 1951;22:79-86.

45. Rypka EW. Methods to evaluate and develop the decision process in the selection of tests. Clinics in Laboratory Med 1992;12[2]: 351-385.

46. Bernstein LH. Outcomes-based Decision Support: How to Link Laboratory Utilization to Clinical Endpoints. Chapter 8. Pp91-128. In Bissell MG, ed. Laboratory-Related Measures of Patient Outcomes: An Introduction. 2000. AACC Press. Washington, DC.

47. Demidenko E.  Mixture models: Theory and applications.  2004.  Wiley-Interscience. Hoboken, NJ.

48. Martin RF. General Deming regression for estimating systematic bias and confidence interval in method-comparison studies. Clin Chem 2000;46:100-104.

49. Magidson J.  Opportunities grow on trees. A general alternative to linear regression. Monotonic regression of dichotomous, ordinal and grouped continuous dependent variables.  1998. Statistical Innovations, Inc. Belmont, MA.

 Figures and Tables Version 8 Multivariable

Table I.  Regression of eGFR and hemoglobin to predict Nt-proBNP

Step number : 0
R : 0.376
R-square : 0.141


In Effect Coefficient Standard Error Std.
Tolerance df F-ratio p-value
1 Constant
2 eGFR -83.499 14.063 -0.297 0.951 1 35.256 0.000
3 Hgb -910.224 260.436 -0.175 0.951 1 12.215 0.001

Information Criteria

AIC 7785.028
AIC (Corrected) 7785.139
Schwarz’s BIC 7800.628


Dependent Variable NTproBNP
N 365
Multiple R 0.376
Squared Multiple R 0.141
Adjusted Squared Multiple R 0.137
Standard Error of Estimate 10287.156

Analysis of Variance

Source SS df Mean Squares F-ratio p-value
Regression 6.309E+009 2 3.155E+009 29.809 0.000
Residual 3.831E+010 362 1.058E+008

Table II. Linear regression of NKLog(Nt-proBNP0/eGFR by eGFR and hemoglobin

Log transform flattens the high Nt-proBNP scale and eGFR and age are normalized







In Effect Coefficient Standard Error Std.
Tolerance df F-ratio p-value
1 Constant
2 eGFR -1.873 0.144 -0.573 0.933 1 170.011 0.000
3 Hgb -4.259 2.436 -0.077 0.933 1 3.056 0.081

Information Criteria

AIC 4299.786
AIC (Corrected) 4299.899
Schwarz’s BIC 4315.331


Dependent Variable NKLogNTGFR
N 360
Multiple R 0.597
Squared Multiple R 0.357
Adjusted Squared Multiple R 0.353
Standard Error of Estimate 94.260

Regression Coefficients B = (X’X)-1X’Y

Effect Coefficient Standard Error Std.
Tolerance t p-value
CONSTANT 256.151 27.745 0.000 . 9.232 0.000
MDRD_GFR -1.873 0.144 -0.573 0.933 -13.039 0.000
Hgb -4.259 2.436 -0.077 0.933 -1.748 0.081

Table III. One-way ANOVA of D-dimer for positive and negative scans

Dependent Variable D_DIMER
N 817

Analysis of Variance

Source Type III SS df Mean Squares F-ratio p-value
VENDUP 43456570.851 1 43456570.851 68.278 0.000
Error 5.187E+008 815 636461.763

Table 4.   Discriminant function for CHF, renal insufficiency and anemia by age, NT-proBNP, creatinine and hemoglobin

Group Frequencies
0 1 2
135 335 235
Group Means
  0 1 2
NTproBNP (pg/ml) 1516.369 5964.054 12902.662
Creatinine 0.716 1.654 2.103
Hgb 11.972 11.533 11.305
Age 60.570 71.373 74.966
Between Groups F-matrix
df : 4 699
  0 1 2
0 0.000
1 23.445 0.000
2 45.108 11.788 0.000

Wilks’s Lambda







Approx. F-ratio










Classification Functions
  0 1 2
CONSTANT -32.018 -35.196 -37.394


F-to-remove Tolerance
5 NTproBNP
13.489 0.801
6 Creatinine 21.368 0.799
7 Hgb 0.190 0.928
3 Age 38.632 0.948
Test Statistic
Statistic Value Approx. F-ratio


Wilks’s Lambda 0.778 23.337 8 1398 0.000
Pillai’s Trace 0.226 22.295 8 1400 0.000
Lawley-Hotelling Trace 0.279 24.382 8 1396 0.000

Table V.  The DFA calculations for Figure 9.

Group Frequencies
  0 1 2
221.000 631.000 571.000
NKLgNTproGFRe 15.589 55.971 81.159
MDRD 123.130 61.940 48.748
Group 0 Discriminant Function Coefficients
MDRDest Constant
NKLgNTproGFRe -0.015
MDRD -0.001 0.000
Constant 0.588 0.052 -15.590
Group 1 Discriminant Function Coefficients
MDRDest Constant
NKLgNTproGFRe 0.000
MDRD 0.000 -0.001
Constant 0.024 0.089 -12.106
Group 2 Discriminant Function Coefficients
MDRDest Constant
NKLgNTproGFRe 0.000
MDRD 0.000 -0.001
Constant 0.015 0.147 -13.077
Between Groups F-matrix
df : 2 1419
  0 1 2
0 0.000
1 236.650 0.000
2 335.228 21.342 0.000

Wilks’s Lambda for the Hypothesis







Approx. F-ratio










Classification Matrix (Cases in row categories classified into columns)
  0 1 2 %correct
0 206 15 0 93
1 237 363 31 58
2 69 459 43 8
Total 512 837 74 43
Jackknifed Classification Matrix
  0 1 2 %correct
0 205 16 0 93
1 237 363 31 58
2 69 462 40 7
Total 511 841 71 43
Test Statistic
Statistic Value Approx. F-ratio


Wilks’s Lambda 0.671 156.542 4 2838 0.000
Pillai’s Trace 0.330 140.347 4 2840 0.000
Lawley-Hotelling Trace 0.488 173.026 4 2836 0.000
Canonical Discriminant Functions
  1 2
Constant -1.912 -1.075
NKLgNTproGFRe 0.001 0.009
MDRD 0.028 0.008
Canonical Discriminant Functions : Standardized by Within Variances
  1 2
NKLgNTproGFRe 0.085 1.061
MDRD 1.026 0.284
Canonical Scores of Group Means
  1 2
0 1.576 0.034
1 -0.122 -0.069
2 -0.476 0.063

Table VI  Ordinal regression model for combined 3 predictors of malnutrition risk.

Predictor                                              L2                     p                      exp(beta)

Poor oral intake                                    60.29               8.2e-15              5.3

Malnutrition related condition    46.29               1.0e-11              3.06

Albumin                                                152.01             6.3e-35              3.16

Table VII.   Expected Odds Ratios – Diagnosis Thalassemia


Me,M,A2(e)                 Thalassemia

1,1,1                                 9713

1,1,0                                 1696

1,0,1                                   263

0,1,1                                   212          

1,0,0                                     46

0,1,0                                     37

0,0,1                                       6

0,0,0                                       1

Table VIII.   Probabilities of RDS given by gestational age and S/A ratio.

Dependent variable: Respiratory outcome (Resp_Sca)

Predictors: Surfactant to albumin (S/A) Ratio_45: 0, > 45; 1, 21-44; 2, < 21;

Gestational age at delivery: 0, > 36; 1, 34-36; 2, < 34.

S/A Ratio_45               p = 8.7*10-22

Gestational Age at Delivery Scaled        p = 4.2*10-9

Combined variables: ChiSq = 130.14,   p = 5.1*10-28,   R2 = 0.433,   phi = 0.8231,   exp(beta) = 2.16 (S/A),   1.88 (GA)

Definition (S/A, GA) Exp. Probabilities Exp. Odds-Ratios
0-20, < 34 0.84 4427
0-20, 34-36 0.64 668
21-44, < 34 0.57 441
0-20, > 36 0.31 101
21-44, 34-36 0.25 67
> 45, < 34 0.19 44
21-44, > 36 0.06 10
> 45, 34-36 0.04 7
> 45, > 36 0.01 1

Table IX. Ordinal regression of EKG and troponin T on diagnoses

Association Summary               L²                     df         p-value             R²        phi

Explained by Model                  206.52             2          1.4e-45            0.686   1.3856

Residual                                         48.64               14        1.0e-5

Total                                               255.16             16        4.5e-45

Odds Ratios and probabilities for diagnoses

average                        1                2                              0          1          2

score                            0.00        0.00

2,3       2.87                             466.82   10086.03         0.01     0.11     0.88

2,2       2.67                             105.78    1087.95           0.04     0.20     0.75

1,3       2.64                             95.35          931.05            0.05     0.21     0.74

2,1       1.95                             23.97          117.35            0.26     0.27     0.47

1,2       1.87                             21.61           100.43           0.29     0.26     0.45

0,3       1.79                             19.48             85.95           0.32     0.26     0.42

1,1       0.67                             4.90                10.83          0.73     0.15     0.12

0,2       0.61                             4.41                   9.27          0.75     0.14     0.11

0,1       0.12                             1.00                 1.00            0.95     0.04     0.01

Table X.  Observed and expected odds and odds-ratios of remission, relapse and no response by half-life

Half-life          exp. odds     exp. odds-ratios

(range, days)      Rem    short    none     Rem   short  none

> 20                           1      4.16    17.11      1    12.49  56.07

16-20                         1      2.21     4.84      1     6.64  44.16

11-15                          1      1.18     1.37      1     3.53  12.49

6-10                            1     0.63     0.39      1     1.88   3.53

< 6                               1     0.33     0.11       1       1         1

HL-ref                         1    0.33      0.11       1       1         1

Figure 1.  log_NT-proBNP vs eGRF

Figure 2.   Boxplots of NT-proBNP and WHO criteria

Figure 3.  NT-proBNP vs Hb

Figure 4.   3D plot of NT-proBNP, MDRD eGFR, Hb

Figure 5.   3D plot of Normalized K*Log_NTproBNP/eGFR, eGFR, Hb

Figure  6. D-dimer Confidence Intervals vs Imaging

Figures 7 & 8.   Canonical Scores Plots

Figure 9.  Entropy Plot of CA125 halflife (x) vs Effective Information
(Kullback Entropy) showing sharp drop in Entropy at 10 days (equivalent to information added to resolve uncertainty).  AS developed by Rosser R Rudolph

Figure 10.  Kaplan Meier Plot of CA125 half-life vs Survival in Ovarian Cancer

Read Full Post »