Potential of mid-infrared spectroscopy to aid the triage of patients with acute chest pain.

A total of 1,429 serum samples from 389 consecutive patients with acute chest pain were analyzed with the goal to aid the rapid diagnosis of acute myocardial infarction. To the best of our knowledge this is the largest and most comprehensive study on mid-infrared spectroscopy in cardiology. We were able to identify those signatures in the mid-infrared spectra of the samples, which were specific to either acute myocardial infarction or chest pain of other origin (angina pectoris, oesophagitis, etc). These characteristic spectral differences were used to distinguish between the cause of the donor's acute chest pain using robust linear discriminant analysis. A sensitivity of 88.5% and a specificity of 85.1% were achieved in a blind validation. The area under the receiver operating characteristics curve amounts to 0.921, which is comparable to the performance of routine cardiac laboratory markers within the same study population. The biochemical interpretation of the spectral signatures points towards an important role of carbohydrates and potentially glycation. Our studies indicate that the "Diagnostic Pattern Recognition (DPR)" method presented here has the potential to aid the diagnostic procedure as early as within the first 6 hours after the onset of chest pain.

Comparison of mid-infrared and Raman spectroscopy in the quantitative analysis of serum.

Mid-infrared or Raman spectroscopy together with multivariate data analysis provides a novel approach to clinical laboratory analysis, offering benefits due to its reagent-free nature, the speed of the analysis and the possibility of obtaining a variety of information from one single measurement. We compared mid-infrared and Raman spectra of the sera obtained from 247 blood donors. Partial least squares analysis of the vibrational spectra allowed for the quantification of total protein, cholesterol, high and low density lipoproteins, triglycerides, glucose, urea and uric acid. Glucose (mean concentration: 154 mg/dl) is frequently used as a benchmark for spectroscopic analysis and we achieved a root mean square error of prediction of 14.7 and 17.1 mg/dl for mid-infrared and Raman spectroscopy, respectively. Using the same sample set, comparable sample throughput, and identical mathematical quantification procedures Raman and mid-infrared spectroscopy of serum deliver similar accuracies for the quantification of the analytes under investigation. In our experiments vibrational spectroscopy-based quantification appears to be limited to accuracies in the 0.1 mmol/l range.

The safety of treatment with recombinant human erythropoietin in clinical use: a review of controlled studies.

Recombinant human erythropoietin (rhEPO) has now been approved for the treatment of renal anemia, anemia of prematurity, cancer-associated anemia, AIDS-associated anemia and as concomitant treatment for patients with or without autologous blood donation awaiting elective surgery. The purpose of this review is to provide an overview, based on the results of controlled studies, of the anticipated safety profile of rhEPO in various indications and to assess whether treatment with rhEPO influences the incidences of certain adverse events in these indications. The anticipated adverse events differ from indication to indication and generally reflect the corresponding underlying illness. With most indications, no relevant differences in the incidences of adverse events are observed between rhEPO and placebo-control/patients. Only in the rhEPO therapy of renal anemia is an increased incidence of hypertensive events observed in the rhEPO groups, a finding that is not reproduced with the other indications. The controlled studies forming the basis of this review provide no evidence of a relevant increase in the risk of thromboembolic events during rhEPO therapy. Overall, it may be stated that rhEPO treatment, where strictly indicated, is a safe form of therapy. As with any other treatment, the risk of side effects in certain predisposed patients must also be weighed against the desired clinical benefits.

Decomposing event-related potentials: a new topographic components model.

"Component" notions inherently used with measurement approaches, raw peak determination, PCA and generator approaches are discussed. By combining aspects of them all, a new model of ERP decomposition is established; quite profitable and surprising mathematical properties are illustrated and discussed.

Doubting these doubts--a reply to Collet.

In reply to Collet (1989) it is argued that principal component analysis (PCA) of event-related potentials is not invalidated or disproved by the arguments of his comment. The main points of this reply are as follows. First, since Collet's analysis was based on the correlation matrix only, it cannot disprove assumptions of PCA which do not constrain the correlation (or covariance) matrix. Second, his model utilizes results of PCA which invalidate the comparison and parameter count. Third, his model does not allow specific predictions and does not lead to real data reduction, which reverses Collet's argument based on the principle of parsimony. Fourth, his model is less general than PCA, as it could apply (if at all) to slow brain potentials only.

Short term changes of event related potentials during concept learning.

In a concept learning task, the time course of P3 amplitudes following the stimuli to be categorized was investigated. That time course did not coincide with the learning curve: it was not until the 4th to 6th trial after learning that P3 amplitudes grew larger. Some implications of this effect are discussed.

Testing and estimation for variable single evoked brain potentials.

The task of validly quantifying variations in electrical activity recorded when the brain is processing external or internal events is addressed. Three new statistical tests sensitive to different types of response changes, which test the null hypothesis that there is a homogeneous signal, are presented. In the case of latency jitter, the testing procedure was developed together with a procedure for estimating the unknown signal shifts. The tests provide a statistically valid and powerful tool in investigating signal variation, even with strong colored noise. Moreover, the differential sensitivity to different types of variations allows a study of the underlying signal variability even though the single signal cannot be estimated.

Effect of insulin analogs on the decline of hemoglobin in diabetic patients with nephropathy.

AIM: Decrease of hemoglobin occurs in diabetic patients with nephropathy earlier than in nondiabetic patients, probably due to impaired synthesis of erythropoietin (EPO). Apart from EPO, insulin also stimulates erythropoiesis. We investigate whether there are differences between human insulin and insulin analogs in respect of their erythropoiesis stimulating effect. PATIENTS AND METHODS: Hemoglobin concentration and other factors which may influence hemoglobin levels were analyzed retrospectively in 203 type 1 diabetic patients with various degrees of kidney function. Eighty-six patients were treated with human insulin and 117 patients received an insulin analog. RESULTS: Hemoglobin concentration did not differ in patients with normal renal function (creatinine clearance (CCL) >90 ml/min) treated with human insulin or insulin analogs. In patients with impaired renal function (CCL<90 ml/min) there was a significant decrease of hemoglobin with declining kidney function in patients treated with human insulin (r=0.463; p<0.003) but not in patients treated with insulin analog (r=-0.12; p=0.4). This result remained significant after adjustment of multiple potential confounders such as age, gender, diabetes duration, BMI, metabolic control, kidney function, chronic inflammation or use of ACE-inhibitors or AT1-blockers. CONCLUSION: Insulin analogs mitigate the decline of hemoglobin in diabetic patients with impaired renal function. This might be due to a stimulating effect of insulin analogs on erythropoiesis via IGF receptor or a sustained activation of the insulin receptor.

Cardiovascular mortality in haemodialysis patients treated with epoetin beta - a retrospective study.

AIMS: Anaemia is a major risk factor in end-stage renal disease (ESRD) and leads to enhanced cardiac output and left-ventricular hypertrophy (LVH). Recombinant human erythropoietin (rhEPO) partially corrects anaemia and reduces LVH in ESRD. The current study retrospectively analysed mortality data from haemodialysis patients included in the clinical development database for epoetin-beta. METHODS: The unselected database, set up to monitor safety from clinical studies of epoetin-beta, comprised 3,111 adult haemodialysis patients included in 17 clinical trials in the clinical development programme (1987-1994). 1,726 and 466 patients were treated for >1 and >2 years, respectively. Untreated control patients (n = 246) were available in two studies. Mortality was measured from fatal adverse event documentation. RESULTS: The controlled studies showed an approximately 20% reduction in the mortality risk for epoetin-beta versus controls after 1 year. For the overall patient population, all-cause mortality fell from a peak (after about 150 days) of about 10 to about 6 deaths per 100 patient-years in the 3rd year of treatment (p < 0.01). The proportion of deaths due to cardiovascular causes fell from 50 to 20-30% over 3 years. The decline in cardiovascular mortality could not be explained by changes in covariate distribution or by drop-outs. CONCLUSIONS: The cardiovascular mortality risk decreased over time in this population of ESRD patients. The beneficial effects of long-term anaemia correction by epoetin-beta therapy was a likely cause of this favourable development.

Principal component analysis of event-related potentials: a note on misallocation of variance.

Wood and McCarthy (1984) found a 'misallocation of variance' when applying PCA, including Varimax rotation, to simulated data. Here it is demonstrated that this effect can be produced by Varimax rotation, without PCA as an intervening step. PCA does not distort or lose information when extracting components, since it is shown that the prototypes may be perfectly reconstructed from the unrotated solution. However, it is stressed that infinitely many sets of prototypes may render the same final solution, a fact which cannot be overcome by any method. The role of the rotation step within this framework is discussed.

Graphical representation of multidimensional EEG data and classificatory aspects.

EEG data are intrinsically multidimensional and this applies also to a set of broad band spectral parameters (6 parameters per derivation). Non-metric multidimensional scaling (MDS) and principal component analysis (PCA) are compared with respect to their capacity to achieve an adequate few-dimensional representation for the graphical comparison of groups and the identification of subgroups. In an application to neurophysiological aspects of mental retardation and learning disability, MDS proved to be somewhat superior to PCA. Both methods profited from the introduction of (1-p)-convex hulls to define a normative region.

How to select epochs of the EEG at rest for quantitative analysis.

Before performing a quantitative analysis of the EEG at rest, it has to be specified how many seconds are to be used later on, and a criterion has to be set down according to which data are selected for further analysis. In this paper it is investigated which duration is appropriate for an analysis of background, non-transient activity. Furthermore, different criteria for automatically selecting such epochs for quantitative analysis are compared with respect to broad-band parameters of a sample of 42 children. A length of 20 sec turned out to be sufficient, since using 40 sec or 60 sec did not improve the statistical variability. Moreover, a criterion based on minimizing EOG power from 1.5 to 7.5 Hz was mildly superior to other criteria considered. The particular choice of epoch was, however, less crucial than anticipated.

Variability of single visual evoked potentials evaluated by two new statistical tests.

In this paper the trial-to-trial variability of brain responses (as opposed to the non-event-related background variability) in flash-evoked potential experiments is investigated by means of two new statistical tests, proposed by M ocks et al. (1984). In two groups of normal children, a considerable number of inhomogeneous responders could be detected, and evidence was found that two different modes of response variability are independently present in the data. The paper also gives a precise description of the computing algorithm and an empirical validation of the applicability to EP data of the two tests.

SELAVCO: a method to deal with trial-to-trial variability of evoked potentials.

The possibility and the potential dangers of the inhomogeneity of potentials to repeated stimuli have been recognized for some time. The predominant method of averaging sweeps of brain activity time locked to the stimuli relies on the assumption of an invariant response. In this paper, a method is proposed for dealing more effectively with the variability of single potentials than does averaging. This method operates by weighted averaging and the weights are determined from the data after suitable filtering operations. SELAVCO is validated and compared to the average for real data and in a controlled design close to reality.

Transformations towards the normal distribution of broad band spectral parameters of the EEG.

In this report transformations are evaluated for obtaining normally distributed broad band spectral parameters (both for absolute and relative power) in a sample of healthy individuals. This problem leads automatically to the question of defining a normative sample, both in terms of sampling and recording conditions; as to the latter, the electro-oculogram proved to be crucial. For relative power the transformation log(x/(1-x)) was excellent in all regards, whereas the best transformation for absolute power--log(x)--was not fully satisfactory for all bands.

The transfer of EOG activity into the EEG for eyes open and closed.

The transfer of EOG activity into the EEG is investigated for eyes open (EO) eyes closed (EC), relying on spontaneously produced EOG activity. Blinks were prominent with EO and eye movements with EC. A frequency domain approach is more appropriate compared to treating the EOG influence as constant over frequency (i.e., assuming that a fraction of the EOG amplitude is present at EEG derivations). A major problem is to take into account the coherent EEG activity present at the EOG derivations, and this holds true in particular for the higher frequency bands were EOG power is relatively low. Blinks and eye movements have different spectral patterns and are also transferred differently to the locations on the skull. Eye movements are transferred best at low frequencies. The gains of blinks peak in the theta band. Due to high random variability in the individual gain functions, sample average gain functions ('grand means') rather than individual ones were studied.

Trial-to-trial variability of single potentials: methodological concepts and results.

Variability of single visual evoked potentials was investigated by means of three statistical tests sensitive to amplitude variations, gradual changes, and latency jitter, respectively. In a sample of (n = 78) normal children, a considerable number of inhomogeneous responders was found, and most prominent were gradual potential changes and latency jitter. Removal of latency jitter demonstrated that the gradual changes are not of latency type and only partly of the amplitude type. As found from empirical densities, there is strong indication that there were subpopulations differing in their response style. On the whole, however, it was concluded that there was no clear, interindividually stable type of response variation in these data.