Echocardiographic markers for mitral valve surgery to preserve left ventricular performance in mitral regurgitation.

BACKGROUND: The timing of mitral valve (MV) surgery to preserve left ventricular (LV) contractility in patients with mitral regurgitation (MR) has been defined by complex cardiac catheterization techniques. Whether noninvasive methods can identify patients with MR, a normal LV ejection fraction, and early LV contractile impairment is unknown. We hypothesized that echocardiographic measures would separate patients with MR and a normal LV ejection fraction into those with and without contractile dysfunction and, thus, prospectively predict the response of LV size and performance to MV surgery. METHODS AND RESULTS: We studied 27 patients with micromanometer LV pressures and radionuclide angiography to obtain a determination of LV volumes and ejection fraction and calculate chamber elastance, a measure of LV contractility, before MV surgery. Echocardiographic studies were performed before MV surgery and repeated at 3 and 12 months after surgery. Age, New York Heart Association class, LV plus maximum pressure per unit change in time, LV systolic and end-diastolic pressures, and echocardiographic posterior wall thickness and radius to wall thickness ratio did not identify preoperative LV contractile dysfunction. However, other echocardiographic measures were related to LV contractility, including LV end-diastolic dimension (r = -0.50, P <.005), LV end-systolic dimension (r = -0.60, P <.0001), and LV fractional shortening (r = 0.50, P =.005). From analysis of receiver operator characteristic curves, an LV end-systolic dimension of >/=40 mm was identified as most predictive for separating patients with MR before surgery into those with and without LV contractile dysfunction (sensitivity of 82% and specificity of 100%). The patients with MR and impaired preoperative LV contractility showed a dramatic deterioration in LV fractional shortening at 3 months after MV surgery (P =.01), which recovered to within the normal range for fractional shortening at 12 months (P =.02) from a progressive reduction in LV end-systolic dimension. This response in LV size and performance temporally differed from that in the patients with MR and normal contractility (2-way analysis of variance P <.0001). However, at 12 months after MV surgery, LV end-diastolic dimension, end-systolic dimension, and fractional shortening were normal in both groups of patients with MR. CONCLUSION: We conclude that echocardiographic measures, particularly an end-systolic dimension of >/=40 mm, may be useful for identifying patients with MR before surgery with early, occult LV contractile dysfunction in whom MV surgery may be recommended to preserve LV systolic performance.

Multimodality staging optimizes resectability in patients with pancreatic and ampullary cancer.

Few patients with pancreatic cancer have resectable disease at the time of diagnosis, and a variety of nonsurgical techniques are available to provide effective palliation of jaundice and pain. Accurate preoperative staging is essential to identify patients with unresectable disease, thereby minimizing unnecessary surgery. Currently used diagnostic tests include contrast-enhanced computerized tomography (CT), visceral angiography, endoscopic ultrasound, and laparoscopy, but their utility remains controversial. To evaluate the accuracy of these various diagnostic tests, 30 consecutive patients with histologically proven pancreatic or ampullary adenocarcinoma treated between 1992 and 1996 were evaluated. All 30 patients had contrast-enhanced CT and laparoscopy, 22 patients (73%) had visceral angiography, and 16 patients (53%) had endoscopic ultrasound. Individual and combined predictive values of resectability and unresectability as well as the sensitivities and specificities were determined for all diagnostic tests and compared with intraoperative findings. When CT, visceral angiography, and laparoscopy were combined, the predictive values of resectability and unresectability were 75 and 90 per cent, respectively, with a sensitivity of 75 per cent and a specificity of 90 per cent. Therefore, the combined use of selected diagnostic tests proved more effective than any single diagnostic test for accurately staging patients with pancreatic head and ampullary cancers and should be considered to minimize unnecessary surgery.

Anterior shoulder stability: Contributions of rotator cuff forces and the capsular ligaments in a cadaver model.

The purpose of this study was to quantify in a biomechanical model the contributions to shoulder joint stability that are made by tensions in the four tendons of the rotator cuff and by static resistance of defined portions of the capsular ligaments. A materials testing machine was used to directly determine anterior joint laxity by measurement of the force required to produce a standard anterior subluxation. Shoulders were tested in external or neutral humeral rotation. Data were analyzed by multiway analysis of variance with regression analysis. This model simulated tensions in the rotator cuff musculature by applying static loads at the tendon insertion sites acting along the anatomic lines of action. A load in any of the cuff tendons resulted in a measurable and statistically significant contribution to anterior joint stability. The contributions between different tendons were not significantly different and did not depend on the humeral rotation (neutral or external). In neutral humeral rotation the superior and middle glenohumeral ligaments together function equally with the inferior glenohumeral ligament as primary stabilizers against anterior humeral translation. The posterior capsule is a secondary stabilizer. The external rotation of the abducted humerus increases anterior stability by more than doubling the stability contribution from the inferior glenohumeral ligament. The stability contribution from the posterior capsule is larger in external rotation than in neutral rotation but is still of secondary magnitude. In external rotation the stability contribution of the anterior capsule, including the superior glenohumeral ligament and the middle glenohumeral ligament, becomes insignificant. The model presented here simulates the combined effect of two major sources of shoulder stability. This versatile model permits the direct measurement of the contributions to anterior shoulder stability that are made by tensions in the rotator cuff tendons and by static resistance of defined capsular zones. The use of multiple regression analysis-a standard statistical technique but one relatively new to the orthopaedic literature-permits quantitative determination of the contribution of each independent variable to the dependent variable, shoulder stability.

A revised indirect estimate of mutation rates in Amerindians.

We have previously raised the possibility that the mutation rate resulting in rare electrophoretic variants is higher in tribal/tropical-dwelling/nonindustrialized societies than in civilized/temperate-dwelling/industrialized societies. Here, we report the results of examining 11 additional proteins for the occurrence of rare electrophoretic variants in 10 Amerindian tribes, for a total of 8,968 determinations and a total of 17,648 locus tests. When these data are combined with the results of all our previous similar studies of Amerindians, a total of 272,298 polypeptides, the products of 43 different loci, have been examined for the occurrence of rare electrophoretic variants. On the assumption that these variants are maintained by mutation pressure and are essentially neutral in their phenotypic effects, we have calculated by three different approaches that it requires an average mutation rate of 1.3 X 10(-5)/locus per generation to maintain the observed variant frequency. Concurrently, we are reporting elsewhere that a direct estimate of the mutation rate resulting in electromorphs in various studies of civilized industrialized populations is 0.3 X 10(-5)/locus per generation. Although this difference appears to have statistical significance, the nonquantifiable uncertainties to both approaches are such that our enthusiasm for a true difference in mutation rates between the two types of populations has diminished. However, even the lower of these estimates, when corrected for all the types of genetic variation that electrophoresis does not detect, implies total locus and gametic mutation rates well above those which in the past have dominated genetic thinking.

A model of weak selection in the infinite alleles framework.

Ewens (1972) proposed a model in the infinite allele framework for populations with neutrality of all alleles at a particular locus. This paper proposes a generalisation of Ewens' result for situations where there is a form of weak selection. The models considered here are continuous time, discrete state space Markov processes.

Adequacy of tissue oxygenation in intact dog intestine.

Changes in O2 consumption, O2 extraction, and intramural pH, resulting from a decreasing O2 delivery, were studied in the intact dog intestine. The O2 delivery was decreased by ischemia, hypoxia, and combined hypoxia-ischemia. A noninvasive approach for determining intramural pH based on the principle of tonometry was used. There was a strong correlation between the changes in intramural pH and intestinal O2 consumption as O2 delivery was decreased. Intramural pH and O2 consumption were initially maintained in the face of decreasing O2 delivery, but after a critical point they decreased. This critical point was 60.3 +/- 1.6% of base-line O2 delivery in the ischemic group and 51.3 +/- 2.7% of base line in the hypoxic-ischemic group. Despite a decrease to 36.0 +/- 5.6% of base-line O2 delivery, the intramural pH and O2 consumption did not decrease in the hypoxic group. O2 extraction increased with decreasing O2 delivery but did not plateau, indicating no diffusion limitation. The data suggest that blood flow is the major factor limiting intestinal O2 consumption. It is concluded that the noninvasive measure of intramural pH is a good marker of the adequacy of tissue oxygenation in canine intestine.

Estimation of phylogenetic relationships from DNA restriction patterns and selection of endonuclease cleavage sites.

The distribution of cleavage sites and their related sequences have been analyzed for 54 restriction endonucleases in the genome of human mitochondrial DNA; in three papova viruses, BK, simian virus 40, and polyoma; and in three bacteriophages, phi X174, fd, and G4. The results show that the cleavage sites and related sequences for most of the restriction enzymes tested are distributed nonrandomly. These results (i) constitute prima facie evidence for the action of natural selection, either direct or indirect on the restriction sites, and (ii) suggest that estimates of phylogenetic relationship, based on a phenetic approach using restriction enzyme data, will be biased.

Assigning a probability for paternity in apparent cases of mutation.

Any direct estimator of mutation in a human population is subject to error due to nonpaternity. This paper deals with the quantification of this error by producing, under certain assumptions, the probability for paternity. In addition, a new direct estimator of the mutation rate is introduced.

The age structure of populations of Saccharomyces cerevisiae.

A discrete deterministic model is described for the growth of an age-structured population of yeast, Saccharomyces cerevisiae, incorporating recent information on the asymmetry of cell division and control of the cell cycle in this species. Solutions are obtained for the age structure of the population at equilibrium, and for the equilibrium distribution of relative frequency of cells through the cell cycle. The model is applied to experimental data on the changing age structure of nonequilibrium populations of yeast. The model predicts well both the transient behavior and the equilibrium structure of such populations. It is shown that the asymmetry of cell division explains (1) the excess of newly formed daughter cells in the population as compared to the frequency of older cells and (2) the damped oscillations in the frequencies of cells of different ages as demographic equilibrium is approached.

Indirect estimates of mutation rates in tribal Amerindians.

Three different formulations have been used to estimate the average rate/locus/generation with which mutation results in electrophoretically detectable variants of 28 proteins in 12 tribal Amerindian populations. All methods are indirect--i.e., they assume a reasonable approximation to equilibrium between mutation and loss of mutants from the population--and are based on the further assumption that the biochemical traits under consideration are essentially neutral in their phenotypic effects. Despite the fact that the methods draw on somewhat different aspects of the available data, there is satisfactory agreement between them, the average of the three estimates being 1.6 X 10(-5)/locus/generation. This average does not encompass mutation that does not result in a change in electrophoretic mobility or that results in loss of enzyme activity. It is noteworthy that this estimate is in satisfactory agreement with a recent estimate by Neel and Thompson [Neel, J. V. & Thompson, E. A. (1978) Proc. Natl. Acad. Sci. USA 75, 1904--1908] of the mutation pressure necessary to maintain the number of "private" genetic polymorphisms being encountered in Amerindian tribes, if selection is not a factor.

Estimation of expected number of rare alleles of a locus and calculation of mutation rate.

An approach is described for the estimation of the number of rare variants in the population from the number in a sample drawn at random from the population. This quantity is used to derive an estimate of the mutation rate. The data required are the number of rare variants in the sample and the distribution of offspring within a population of well-defined size with little or no immigration. Application of this approach to data on 28 loci assayed in the Yanamamo, a tribe of South American Indians, yields an average mutation rate of 0.1 approximately 0.2 x 10(-5). Determination of this figure is subject to several assumptions concerning the nature of the rare variants and the structure of the population. Violation of these assumptions will generally result in the underestimate of the true mutation rate.

Estimation of the number of sex alleles and queen matings from diploid male frequencies in a population of Apis mellifera.

The distribution of diploid males in a population of Apis mellifera was obtained by direct examination of the sexual phenotypes of the larvae. Using these data, estimates are derived for the number of sex alleles and the number or matings undergone by the queen. The number of sex alleles is estimated to be 18.9. The estimate is larger than previous ones, which have ranged between 10 and 12. However, the increase in the number of sex alleles can be explained by the large effective population number for our data. The best estimator of the number of matings by a queen is a maximum likelihood type that assumes a prior distribution on the number of matings. For the data presented here, this estimate is 17.3. This estimate is compared to others in the literature obtained by different approaches.

A consideration of the chi-square test of Hardy--Weinberg equilibrium in a non-multinomial situation.

The multinomial Dirichlet distribution was used to study the effect of correlation between observations in a sample on the frequency of rejection of the Hardy--Weinberg Law when in fact it was true in the population being sampled. It was shown that the usual X2 goodness-of-fit test of Hardy--Weinberg is very sensitive to non-multinomial sampling. In view of the lack of statistical power of the test to detect deviations due to in breeding, it is likely that whenever H--W is rejected using samples of size 100 or less, the underlying causation is sample correlation rather than failure of the H--W law to be true. Related to these findings is, of course, the effect of pooling heterogeneous frequencies or, in the case of contingency tables, Simpson's paradox (see Simpson, 1951).

A model for analysis of population structure.

Arguments have been presented for the appropriateness of a multinomial Dirichlet distribution for describing single-locus genotypic frequencies in a subdivided population. This distribution is defined as a function of allele frequency, the average (over the entire population) inbreeding coefficient and the correlation between genotypes within a subdivision. Alternative parameterizations and their genetic interpretations are given.-We then show how information from a sample drawn from this subdivided population, in the absence of pedigrees, can be combined with the multinomial Dirichlet model to form a likelihood function. This likelihood function is then used as the basis for estimation and testing hypotheses concerning the genetic parameters of the model. Comparisons of this approach to the alternative procedure of Cockerham (1969) and (1973) are made using human data obtained from Tecumseh, Michigan and Monte Carlo simulations.-Finally, implications of these results to statistical inference and to mutation rates are presented.

The population genetics of parthenogenetic strains of Drosophila mercatorum : I. One locus model and statistics.

A one locus model has been developed to describe parthenogenetic populations restoring diploidy by central fusion, terminal fusion and gamete duplication. It was found that in the absence of selection all populations become homozygous. With selection, however, it is possible to maintain heterozygotes and homozygotes. The conditions required to yield such an equilibrium are a function of (1) the proportions of the various diploid restoring mechanisms (2) linkage to the kinetochore and (3) the intensity of selection. The model was then used to derive one-generation likelihood functions. These likelihoods were used in deriving estimation procedures for the frequency of gamete duplication which is important in forming isogenic lines and for the probability of a heterozygous female giving rise to a heterozygous zygoid. Next, n-generation likelihood functions with and without selection were calculated. These were used to estimate the selection coefficient and to derive two tests of the hypothesis of no selection versus the hypothesis of selection. The first test is a locally best test in the vicinity of no selection, and the second an "odds" for the hypotheses using a prior distribution on the selection coefficient.