Let {Xn, n≥1} be a sequence of independent and identically distributed random variables, taking non-negative integer values, and call Xn a δ-record if Xn>max {X1, …, Xn−1}+δ, where δ is an integer constant. We use martingale arguments to show that the counting process of δ-records among the first n observations, suitably centered and scaled, is asymptotically normally distributed for δ≠0. In particular, taking δ=−1 we obtain a central limit theorem for the number of weak records.

To prevent the overdiagnosis and overtreatment of prostate cancer (PC), therapeutic strategies have been established such as active surveillance and focal therapy, as well as methods for clarifying the diagnosis of high-grade prostate cancer (HGPC) (defined as a Gleason score ≥7), such as multiparametric magnetic resonance imaging and new markers such as the 4Kscore test (4KsT). By means of a pilot study, we aim to test the ability of the 4KsT to identify HGPC in prostate biopsies (Bx) and compare the test with other multivariate prognostic models such as the Prostate Cancer Prevention Trial Risk Calculator 2.0 (PCPTRC 2.0) and the European Research Screening Prostate Cancer Risk Calculator 4 (ERSPC-RC 4). Fifty-one patients underwent a prostate Bx according to standard clinical practice, with a minimum of 10 cores. The diagnosis of HGPC was agreed upon by 4 uropathologists. We compared the predictions from the various models by using the Mann-Whitney U test, area under the ROC curve (AUC) (DeLong test), probability density function (PDF), box plots and clinical utility curves. Forty-three percent of the patients had PC, and 23.5% had HGPC. The medians of probability for the 4KsT, PCPTRC 2.0 and ERSPC-RC 4 were significantly different between the patients with HGPC and those without HGPC (p≤.022) and were more differentiated in the case of 4KsT (51.5% for HGPC [25-75 percentile: 25-80.5%] vs. 16% [P 25-75: 8-26.5%] for non-HGPC; p=.002). All models presented AUCs above 0.7, with no significant differences between any of them and 4KsT (p≥.20). The PDF and box plots showed good discriminative ability, especially in the ERSPC-RC 4 and 4KsT models. The utility curves showed how a cutoff of 9% for 4KsT identified all cases of HGPC and provided a 22% savings in biopsies, which is similar to what occurs with the ERSPC-RC 4 models and a cutoff of 3%. The assessed predictive models offer good discriminative ability for HGPCs in Bx. The 4KsT is a good classification model as a whole, followed by ERSPC-RC 4 and PCPTRC 2.0. The clinical utility curves help suggest cutoff points for clinical decisions: 9% for 4KsT and 3% for ERSPC-RC 4. This preliminary study should be interpreted with caution due to its limited sample size.

Clinical under staging occurs in 40% to 60% of patients who undergo radical prostatectomy for prostate cancer. To decrease under staging several methods of predicting pathological stage preoperatively have been developed based on statistical logistic regression analysis and neural networks. To our knowledge none has been validated in our homogeneous regional patient population to date. We created logistic regression and neural network models, and implemented and adapted them into our practice. We also compared the 2 methods to determine their value and practicality in daily clinical practice. We present the results of our novel approach for predicting pathological staging of prostate adenocarcinoma. Between 1986 and 1999, 600 white men from the Aragon region of Spain underwent surgery for prostate cancer; of whom 468 were selected for study. Predictive study variables included patient age, clinical stage, biopsy Gleason score and preoperative prostate specific antigen (PSA). The predicted result included in analysis was organ confined or nonorgan confined disease. Data were analyzed by multivariate logistic regression and a supervised neural network (multilayer perceptron and radial basis function). Results were compared by comparing the areas under the receiver operating characteristics curves. We generated 5 logistic regression models. The model created with clinical staging, Gleason biopsy score and PSA distributed in 5 categories (p <0.001) with an area under the receiver operating characteristics curve of 0.840 proved to be most predictive of pathological stage. Similarly of the 6 neural network models evaluated the radial basis function model, which included age, clinical stage, Gleason biopsy score and preoperative PSA distributed in 5 categories with an area under the curve of 0.882, proved the most predictive but not superior to the logistic regression model. The difference in the area under the curves in the 2 chosen models was 0.042 (p = 0.1). It is possible to generate useful predictive models of organ confined disease using logistic regression or neural networks with high indexes of clinical and statistical validity. However, using these variables neural networks did not prove to be better than logistic regression analysis. Therefore, better predictive variables must be identified, preferably nonlinear characteristics with respect to the probability of organ confined tumor, to generate better predictive models using neural networks.

Objective: To investigate the association between pregestational body mass index (BMI), total gestational weight gain (GWG) and/or trimester-specific weight gain (GWGT) with adverse maternal or perinatal outcomes (AMPOs). Material and methods: Maternal clinical characteristics and pregnancy and perinatal outcomes were used to predict AMPOs. The predictive ability of BMI, GWG or GWGT for AMPOs was analyzed using the area under the curve (AUC). Logistic regression models in a univariate and multivariate analysis were performed to estimate the odds ratios (OR) and 95% confidence intervals (CI) to predict maternal outcomes (pregnancy induced hypertension, preeclampsia or gestational diabetes mellitus) and perinatal outcomes (small for gestational age, large for gestational age, 5-min Apgar score, admission to neonatal intensive care unit or umbilical cord pH < 7.15). Results: Women with AMPOs (n = 293) were younger with higher rate of nulliparity (p < 0.001) and with lower height (p = 0.018) as compared to controls (n = 134). In the univariate study, GWGT in third trimester was associated with double risk of pregnancy induced hypertension (OR 2.00; 95% CI, 1.01–3.97). Nonetheless, third-trimester GWG and total GWG have a negative relationship with gestational diabetes mellitus OR 0.32 (95% CI, 0.18–0.58) and OR 0.35 (95% CI, 0.21–0.59), respectively. Women with greater overall and in second trimester GWG have lower risk of having SGA neonates, OR 0.62 (95% CI, 0.39–0.98) and OR 0.60 (95% CI, 0.37–0.98), respectively. In the multivariate study, pregestational BMI is strongly related to the development of preeclampsia and the area under the curve (AUC) of the combination of pregestational BMI and total weight gain was 0.832 (95% CI, 0.63–0.81) for preeclampsia and 0.719 (95% CI, 0.71–0.94) for gestational diabetes mellitus. Conclusion: Our results suggest than timing of gestational weight gain influence in maternal and perinatal outcomes. Pregestational BMI is a determinant of preeclampsia, maternal weight gain in the third trimester is a determinant of pregnancy induced hypertension and the increase in total GWG reduces the risk of gestational diabetes mellitus and small for gestational age.

Objective: To assess the predictive ability of the ultrasound estimated percentile weight (EPW) at 35 weeks of pregnancy to predict adverse perinatal outcomes (APOs) at term delivery according to 5 fetal growth standards, including population, population-customized, and international references. Methods: This was a retrospective cohort study of 9,585 singleton pregnancies. Maternal clinical characteristics, fetal ultrasound data obtained at 35 weeks and pregnancy and perinatal outcomes were used to calculate EPWs to predict APOs according to: the customized and noncustomized (NC) Miguel Servet University Hospital (MSUH), the customized Figueras, the NC INTERGROWTH-21st, and the NC World Health Organization (WHO) international standards. APOs were defined as the occurrence of cesarean or instrumental delivery for nonreassuring fetal status, 5-min Apgar score < 7, arterial cord blood pH <7.10, or stillbirth. The predictive ability of EPW for APOs was analyzed using the area under the curve (AUC), and sensitivities were calculated for different false-positive rates (FPRs). Results: For a 10% FPR, detection rates for total APOs ranged between 12.7% with the customized MSUH (AUC 0.52; 95% CI 0.50–0.55) and 14.4% with the NC MSUH standard (AUC 0.55; 95% CI 0.53–0.57) for EPW by ultrasound; and from 22.0% with the customized MSUH standard (AUC 0.60; 95% CI 0.58–0.63) to 27.8% with the NC WHO (AUC 0.65; 95% CI 0.63–0.68) for EPW at delivery. Conclusions: The predictive capacity of the EPW for APOS is limited and similar, by both ultrasound and at delivery, for the 5 growth standards, without significant differences between customized and NC standards.

Objectives: To implement the use of nomograms in clinical practice showing how to choose thresholds in nomograms’ predictions to select risk groups. To validate and compare the predictive ability and clinical utility of the H ospital U niversitario ‘ M iguel S ervet’ ( HUMS ) and the updated P artin T ables 2012 ( PT ‐2012) nomograms to predict organ‐confined disease ( OCD ) after radical prostatectomy ( RP ). Patients and Methods: Cohort of 1285 patients with prostate cancer treated with RP at I nstituto V alenciano de O ncología ( IVO ) between 1986 and 2011. The predictive value of the nomograms was assessed by means of calibration curves, discrimination ability (area under the receiver operating characteristic ( ROC ) curve ( AUC ) and probability density functions). The clinical utility was evaluated through V ickers’ decision curves and thresholds were chosen through probability density functions. Results: The calibration curves showed a minimal underestimation in low probabilities (50%) in the HUMS nomogram and a regular minimal overestimation in the PT ‐2012. Their AUC of 0.7285 (95% confidence interval [ CI ] 0.7010–0.7559) and 0.7288 (95% CI 0.7013–0.7562) respectively, show an adequate discrimination ability for both predictive models in the IVO cohort. The decision curves show similar net benefits for both models. In this study we advocate for a threshold of 53% for the identification of OCD . Conclusions: The HUMS ‐nomogram and the PT ‐2012 predictions of OCD confirm their utility in a contemporary cohort of patients. Patients with a probability of OCD >53% should be classified as OCD , helping physicians to better counsel their patients. A selection of adequate thresholds, as presented in this paper, makes nomograms more accessible tools.