JACKKNIFE BIAS REDUCTION FOR POLYCHOTOMOUS LOGISTIC REGRESSION

Despite theoretical and empirical evidence that the usual MLEs can be misleading in finite samples and some evidence that bias reduced estimates are less biased and more efficient, they have not seen a wide application in practice. One can obtain bias reduced estimates by jackknife methods, with or without full iteration, or by use of higher order terms in a Taylor series expansion of the log-likelihood to approximate asymptotic bias. We provide details of these methods for polychotomous logistic regression with a nominal categorical response. We conducted a Monte Carlo comparison of the jackknife and Taylor series estimates in moderate sample sizes in a general logistic regression setting, to investigate dichotomous and trichotomous responses and a mixture of correlated and uncorrelated binary and normal covariates. We found an approximate two-step jackknife and the Taylor series methods useful when the ratio of the number of observations to the number of parameters is greater than 15, but we cannot recommend the two-step and the fully iterated jackknife estimates when this ratio is less than 20, especially when there are large effects, binary covariates, or multicollinearity in the covariates. © 1997 by John Wiley & Sons, Ltd.     

Sudden Cardiac Death Despite a Functional Cardioverter‐Defibrillator: The Case for Early and Aggressive Therapy for Ventricular Tachycardia in Selected Patients

We present three cases within 11 months at a single institution of sustained VT that fell below the programmed detection rate of the patients’ implantable cardioverter‐defibrillators (ICDs), two of which continued until converting to an agonal VF that did not meet criteria for detection, and a third case that could not be successfully defibrillated after a prolonged period of VT. These episodes may be under‐recognized due to the dependence of device diagnostic storage on programming and the post‐mortem effort that is often required to review these events. Some patients, likely those with the most advanced heart failure, may not tolerate sustained ventricular tachycardia (VT) and may even die from ventricular arrhythmias without ever having a rhythm that meets detection criteria in a ventricular fibrillation (VF) zone.     

ALTERATIONS IN GAP JUNCTION PROTEIN EXPRESSION IN HUMAN BENIGN PROSTATIC HYPERPLASIA AND PROSTATE CANCER

PURPOSE: Gap junctions composed of connexin proteins have an essential role in intercellular communication and differentiation. Dysregulation of connexin expression is believed to have a role in carcinogenesis. The human prostate has been reported to express connexin 32 and 43. However, the expression pattern in prostate cancer is controversial, while to our knowledge connexin expression has not been reported in benign prostatic hyperplasia (BPH). To understand the potential involvement in prostate disease connexin 32 and 43 expression was evaluated in a series of normal prostate, BPH and prostate cancer specimens that were surgically removed due to bladder outlet obstruction. MATERIALS AND METHODS: Frozen sections of 23 normal, 43 BPH and 40 cancer involved prostates were evaluated for the presence, staining intensity and pattern of connexin 32 and 43 by immunocytochemical testing. RESULTS: In all specimens examined connexin 43 stain was punctate along the borders of the basal epithelial cells, whereas connexin 32 immunolocalized to luminal epithelial cells. In normal prostate connexin 43 and 32 were present in 87% and 65% of specimens, respectively, at low to moderate stain intensity. Importantly none of the normal samples were negative foreach connexin. In BPH specimens there was a marked increase in the incidence and intensity of connexin 43 and 32 immunostaining within epithelial cells. In addition, 23% of BPH samples showed strong connexin 43 expression in stromal cells. In contrast, connexin was decreased in prostate cancer specimens, of which 65% and 38% were negative for connexin 43 and 32, respectively, and 28% were negative for each type. In poorly differentiated tumors connexin 43 and 32 were present in only 10% and 40% of tumors, respectively, at low immunostaining intensity. CONCLUSIONS: In normal human prostate basal cells communicate via connexin 43 gap junctions, whereas luminal cells communicate via connexin 32 gap junctions. In BPH gap junctional intercellular communication is increased in epithelial and stromal cells, which may have a role in BPH pathogenesis. In prostate cancer gap junctional intercellular communication is decreased, is as indicated by decreased expression of connexin 43 and 32 with severe loss in poorly differentiated prostate cancer. These alterations in connexin expression may have a role in dedifferentiation and tumor progression.