A linear adjustment-based approach to posterior drift in transfer learning.

We present new models and methods for the posterior drift problem where the regression function in the target domain is modelled as a linear adjustment, on an appropriate scale, of that in the source domain, and study the theoretical properties of our proposed estimators in the binary classification problem. The core idea of our model inherits the simplicity and the usefulness of generalized linear models and accelerated failure time models from the classical statistics literature. Our approach is shown to be flexible and applicable in a variety of statistical settings, and can be adopted for transfer learning problems in various domains including epidemiology, genetics and biomedicine. As concrete applications, we illustrate the power of our approach (i) through mortality prediction for British Asians by borrowing strength from similar data from the larger pool of British Caucasians, using the UK Biobank data, and (ii) in overcoming a spurious correlation present in the source domain of the Waterbirds dataset.

Protein Convertase Subtilisin/Kexin Type 9 Monoclonal Antibodies (PCSK9mab) in Clinical Practice at Secondary Care - Real World Multicentre Experience.

Background and Aims Protein convertase subtilisin/Kexin type 9 monoclonal antibodies (PCSK9mab) are a novel addition to the therapeutic options for managing hyperlipidemia. Various guidelines have advocated the addition of these agents if the target low-density lipoprotein-cholesterol ( LDL-C) is not achieved by maximum lipid-lowering therapy. They have shown a robust and consistent reduction in LDL-C in clinical trials. However, the translation of these results in a real-world setting is limited and confined mainly to tertiary lipid centers. This service evaluation aimed to assess their efficacy in a real-world outpatient setting of secondary care centers. Methods Data was collected retrospectively from four hospitals in the North-West of England. Patients were required to attend a lipid clinic for follow-up investigations to continue with the prescription of PCSK9mab. Results A total of 175 patients were identified. Efficacy outcomes were measured in 169 patients. 6 discontinued the agent within 3 months of initiation and were excluded from the efficacy outcomes. 19.5% (n=33) had confirmed familial hypercholesterolemia. 61% (n=103) of the patients were intolerant to statins. 53.2% (n=90) of the patients have been prescribed Alirocumab. Mean LDL-C reduction was 50.6% at 6-month which was sustained at 48.9% at 12 months. There was no difference in % reduction of LDL-C between Alirocumab and Evolocumab. LDL-C reduction was more significant in patients who were on concomitant statins. 9.1% of patients experienced side effects, and 5.1% discontinued the PCSK9mab during treatment. Conclusion The efficacy of lipid reduction and the side effect profile of PCSK9mab from these secondary care services are similar to randomized clinical trials and real-world observational studies from tertiary lipid centers.

Non-diabetic ketoacidosis: a case series and literature review.

The genesis of ketone bodies by organisms is a protective mechanism. This metabolic process helps organisms to survive acute metabolic derangements in times of nutrient deficiency. When prolonged, ketogenesis leads to ketoacidosis, which is a potentially life-threatening metabolic disorder due to the accumulation of keto-acids in the body. The most common cause is diabetic ketoacidosis, though starvation ketoacidosis and alcoholic ketoacidosis are not uncommon. The presentation of all ketoacidotic states is similar-being generally unwell, abdominal pain, rapid and shallow breathing, vomiting and dehydration. Non-diabetic ketoacidotic states are very commonly overlooked due to relative unawareness among the clinicians, leading to misdiagnosis and thereby inappropriate management culminating in added mortality and morbidity. We describe here six cases of alcoholic and starvation ketoacidosis, review the literature currently available and discuss the common pitfalls in managing such cases.

Modulation of Small Artery Function by Insulin in Young Women: Role of Adiposity.

OBJECTIVES: Vascular dysfunction is common in obesity. Insulin can directly modulate arterial function, but its role is unclear in obesity. We examined the influence of adiposity on direct effects of insulin on human artery responses. METHODS: 22 healthy women were stratified by median BMI into lower (LA) (n=11) and higher adiposity (HA) (n=11). Small arteries from gluteal biopsies were tested for contractile responses to Noradrenaline (NA), the endothelium-dependent dilator Carbachol and the endothelium-independent dilator sodium nitroprusside were examined before and after incubation with 100 mU/ml human insulin. RESULTS: Contractile responses were similar in the two groups. Insulin reduced NA-induced contraction in HA [3.5 (2.4-4.6) vs. 2.4 (1.4-3.4) mN/mm: p=0.004] but not those from LA [4.1 (2.8-5.3) vs. 3.7 (2.5-5.0) mN/mm: p=0.33]. Endothelium-dependent dilation (EDD) was significantly reduced in arteries from women in the HA (34.7 (18.8-50.6%)) compared to those from women in the LA (62.3 (46.2- 78.4); p=0.013). Insulin improved EDD (change in maximal dilation before/after insulin (%)) in arteries from the HA (37.7 (18.0 to 57.3) but not the LA (6.3 (-6.5 to 19.1), p=0.007. CONCLUSION: Reduced EDD evident in arteries from HA subjects improve by incubating in insulin. Hyperinsulinaemia may be necessary in maintaining endothelial function in obesity.

PPARγ Agonistic Activity of Sulphonylureas.

BACKGROUND: Sulphonylureas (SU) are known to cause weight gain. Some investigators have reported increased insulin sensitivity with some sulphonylurea agents. OBJECTIVE: To review available evidence of SU agents having PPARγ agonist activity. METHODS: We searched online databases of PubMed®, Embase®, Google Scholar® and Web of Science® as per current guidance, published in English, between 1st January 1970 and 31st December 2017. The search found 6 articles. RESULTS: None of the 1st generation SU drugs have any demonstrable PPARγ agonist activity. Most of the 2nd generation SU agents had a positive correlation between their concentration and PPARγ agonist activity except Gliclazide. The demonstrated PPARγ agonist activity was maximum in experiments with Glimepiride and Gliquidone and was seen in these in-vitro experiments at concentrations which were pharmacologically achievable in-vivo. The PPARγ agonist activity may be responsible for some sideeffect of the SU agents as weight gain. On the contrary, the clinical efficacy of the thiazolidinediones could theoretically be reduced when used in combination with the SUs with significant PPARγ agonist activity. CONCLUSION: The PPARγ agonist activity demonstrated in vitro experiments may have clinical connotations.

Drawing inferences for high-dimensional linear models: A selection-assisted partial regression and smoothing approach.

Drawing inferences for high-dimensional models is challenging as regular asymptotic theories are not applicable. This article proposes a new framework of simultaneous estimation and inferences for high-dimensional linear models. By smoothing over partial regression estimates based on a given variable selection scheme, we reduce the problem to low-dimensional least squares estimations. The procedure, termed as Selection-assisted Partial Regression and Smoothing (SPARES), utilizes data splitting along with variable selection and partial regression. We show that the SPARES estimator is asymptotically unbiased and normal, and derive its variance via a nonparametric delta method. The utility of the procedure is evaluated under various simulation scenarios and via comparisons with the de-biased LASSO estimators, a major competitor. We apply the method to analyze two genomic datasets and obtain biologically meaningful results.

Risk Factors for Failure of Direct Current Cardioversion in Patients with Type 2 Diabetes Mellitus and Atrial Fibrillation.

INTRODUCTION: Type 2 diabetes mellitus (T2DM) is a well-recognised risk factor for cardiovascular disease and the prevalence of atrial fibrillation (AF) is higher among patients with T2DM. Direct current cardioversion (DCCV) is an important management option in persistent AF. We sought to determine independent risk factors for immediate and short-term outcomes of DCCV for treatment of AF in patients with T2DM. METHODS: Retrospective outcome analysis of DCCV for persistent AF in 102 T2DM patients compared with 102 controls. RESULTS: DCCV was successful in 68 (66.6%) people with T2DM compared to 86 (84.3%) in the control group (P = 0.003). After initial successful cardioversion, only 38 (37.2%) T2DM patients remained in sinus rhythm compared to 63 (61.8%) in the control group (P = 0.007) at a median follow-up of 74.5 days (IQR 69.4-77.4). Multiple logistic regression analysis showed that the presence of T2DM (P = 0.014), digoxin use (P = 0.01), statin use (P = 0.005), left-atrial size (P = 0.01), and LV ejection fraction (P = 0.008) were independent risk factors for immediate DCCV failure. T2DM (P = 0.034) was an independent risk factor for AF relapse. Among patients with T2DM, previous DCCV (P = 0.033), digoxin use (P = 0.035), left-atrial size (P = 0.01), LV ejection fraction (P = 0.036), and HbA1c (P = 0.011) predicted immediate failure of DCCV whilst digoxin use (P = 0.026) was an independent risk factor for relapse of AF. CONCLUSION: T2DM, higher HbA1c, digoxin treatment, and structural and functional cardiac abnormalities are independent risk factors for immediate DCCV failure and AF relapse.

Effect of ranolazine on glycaemia in adults with and without diabetes: a meta-analysis of randomised controlled trials.

Background: Ranolazine is an antianginal drug reported to have hypoglycaemic effects. Objectives: To assess the effect of ranolazine versus placebo on glycaemic control for adults with and without diabetes. Methods: A systematic search of seven databases was conducted to identify all randomised controlled trials that compared the effect of ranolazine versus placebo on haemoglobin A1c (HbA1c) and/or fasting plasma glucose (FPG) and/or incidence of hypoglycaemia. We used mean differences in HbA1c and FPG to express intervention effect estimates and analysed the data with random-effects model for meta-analyses using Revman 5.3. Results: We identified seven trials including 6543 subjects to assess the effect of ranolazine on HbA1c and/or FPG. A separate trial that included 944 subjects was included to assess the effect of ranolazine on hypoglycaemia. The change in HbA1c for all patients was -0.36% (95% CI -0.57% to -0.15%; p=0.0004, I2=78%). In patients with diabetes, the change in HbA1c was -0.41% (95% CI -0.58% to -0.25%; p<0.00001, I2=65%). There was no significant difference in FPG between ranolazine and placebo groups (-2.58 mmol/L, 95% CI -7.02 to 1.85; p=0.25; I2=49%) or incidence of hypoglycaemia between ranolazine and placebo groups (OR 1.70, 95% CI 0.89 to 3.26; p=0.61, I2=0%). Conclusions: Our meta-analytic findings support the fact that ranolazine improves HbA1c without increasing the risk of hypoglycaemia. It therefore has a potential of having an additional benefit of improving glycaemic control in patients with chronic stable angina and diabetes.

ASYMPTOTICS FOR CHANGE-POINT MODELS UNDER VARYING DEGREES OF MIS-SPECIFICATION.

Change-point models are widely used by statisticians to model drastic changes in the pattern of observed data. Least squares/maximum likelihood based estimation of change-points leads to curious asymptotic phenomena. When the change-point model is correctly specified, such estimates generally converge at a fast rate (n) and are asymptotically described by minimizers of a jump process. Under complete mis-specification by a smooth curve, i.e. when a change-point model is fitted to data described by a smooth curve, the rate of convergence slows down to n1/3 and the limit distribution changes to that of the minimizer of a continuous Gaussian process. In this paper we provide a bridge between these two extreme scenarios by studying the limit behavior of change-point estimates under varying degrees of model mis-specification by smooth curves, which can be viewed as local alternatives. We find that the limiting regime depends on how quickly the alternatives approach a change-point model. We unravel a family of 'intermediate' limits that can transition, at least qualitatively, to the limits in the two extreme scenarios. The theoretical results are illustrated via a set of carefully designed simulations. We also demonstrate how inference for the change-point parameter can be performed in absence of knowledge of the underlying scenario by resorting to subsampling techniques that involve estimation of the convergence rate.

Fast estimation of regression parameters in a broken-stick model for longitudinal data.

Estimation of change-point locations in the broken-stick model has significant applications in modeling important biological phenomena. In this article we present a computationally economical likelihood-based approach for estimating change-point(s) efficiently in both cross-sectional and longitudinal settings. Our method, based on local smoothing in a shrinking neighborhood of each change-point, is shown via simulations to be computationally more viable than existing methods that rely on search procedures, with dramatic gains in the multiple change-point case. The proposed estimates are shown to have [Formula: see text]-consistency and asymptotic normality - in particular, they are asymptotically efficient in the cross-sectional setting - allowing us to provide meaningful statistical inference. As our primary and motivating (longitudinal) application, we study the Michigan Bone Health and Metabolism Study cohort data to describe patterns of change in log estradiol levels, before and after the final menstrual period, for which a two change-point broken stick model appears to be a good fit. We also illustrate our method on a plant growth data set in the cross-sectional setting.

Evidence that multiple defects in lipid regulation occur before hyperglycemia during the prodrome of type-2 diabetes.

BACKGROUND: Blood-vessel dysfunction arises before overt hyperglycemia in type-2 diabetes (T2DM). We hypothesised that a metabolomic approach might identify metabolites/pathways perturbed in this pre-hyperglycemic phase. To test this hypothesis and for specific metabolite hypothesis generation, serum metabolic profiling was performed in young women at increased, intermediate and low risk of subsequent T2DM. METHODS: Participants were stratified by glucose tolerance during a previous index pregnancy into three risk-groups: overt gestational diabetes (GDM; n = 18); those with glucose values in the upper quartile but below GDM levels (UQ group; n = 45); and controls (n = 43, below the median glucose values). Follow-up serum samples were collected at a mean 22 months postnatally. Samples were analysed in a random order using Ultra Performance Liquid Chromatography coupled to an electrospray hybrid LTQ-Orbitrap mass spectrometer. Statistical analysis included principal component (PCA) and multivariate methods. FINDINGS: Significant between-group differences were observed at follow-up in waist circumference (86, 95%CI (79-91) vs 80 (76-84) cm for GDM vs controls, p<0.05), adiponectin (about 33% lower in GDM group, p = 0.004), fasting glucose, post-prandial glucose and HbA1c, but the latter 3 all remained within the 'normal' range. Substantial differences in metabolite profiles were apparent between the 2 'at-risk' groups and controls, particularly in concentrations of phospholipids (4 metabolites with p ≤ 0.01), acylcarnitines (3 with p ≤ 0.02), short- and long-chain fatty acids (3 with p<  = 0.03), and diglycerides (4 with p ≤ 0.05). INTERPRETATION: Defects in adipocyte function from excess energy storage as relatively hypoxic visceral and hepatic fat, and impaired mitochondrial fatty acid oxidation may initiate the observed perturbations in lipid metabolism. Together with evidence from the failure of glucose-directed treatments to improve cardiovascular outcomes, these data and those of others indicate that a new, quite different definition of type-2 diabetes is required. This definition would incorporate disturbed lipid metabolism prior to hyperglycemia.

Small artery function 2 years postpartum in women with altered glycaemic distributions in their preceding pregnancy.

GDM (gestational diabetes mellitus) is associated with later adverse cardiovascular risk. The present study examined the relationship between glycaemia during pregnancy and small artery function and structures approx. 2 years postpartum. Women were originally enrolled in the HAPO (Hyperglycaemia and Adverse Pregnancy Outcome) study from which they were classified by their glycaemic distribution during pregnancy as controls (in the lower half of the distribution), UQ (upper quartile; in the UQ of the glycaemic distribution) or having had overt GDM. Subcutaneous arteries from a gluteal fat biopsy taken at follow-up 2 years later were examined using wire myography. Small artery structure, stiffness and vasoconstrictor responses were similar across groups. Maximal endothelium-dependent dilation in response to carbachol was impaired in arteries from both GDM (43.3%, n=8 and P=0.01) and UQ (51.7%, n=13 and P=0.04) women despite generally 'normal' current glycaemia (controls, 72.7% and n=8). Inhibition of NOS (nitric oxide synthase) significantly reduced maximum endothelium-dependent dilation in controls but had no effect on arteries from UQ and GDM women, suggesting impaired NOS activity in these groups. Endothelium-independent dilation was unaffected in arteries from previous GDM and UQ women when compared with the control group. Multiple regression analysis suggested that BMI (body mass index) at biopsy was the most potent factor independently associated with small artery function, with no effect of current glycaemia. Overweight women with either GDM or marginally raised glycaemia during pregnancy (our UQ group) had normal vascular structure and stiffness, but clearly detectable progressively impaired endothelium-dependent function at 2 years follow-up. These results suggest that vascular pathology, which may still be reversible, is detectable very early in women at risk of decline into Type 2 diabetes mellitus.

Impact of a single one-to-one education session on glycemic control in patients with diabetes.

BACKGROUND: Education is one of the pivotal aspects of diabetes care. The impact of education has been found to be efficacious in the short term, but tends to lose efficacy in the long term. The hypothesis tested here was that one-to-one education would confer knowledge that would be reflected in metabolic improvement in this group of participants with diabetes. METHODS: Thirty-nine patients with type 2 diabetes mellitus and poor diabetes control attended a one-to-one diabetic and dietetic education session with a diabetic specialist nurse and a diabetes specialist dieticican. Glycemic control was assessed by measuring serum glycosolated hemoglobin (Hb(A1c)) measurements before the session and at 6 and 12 months afterwards. RESULTS: The Hb(A1c) levels fell significantly in the whole group to 8.0 ± 0.5% (P < 0.05) at 6 months and to 8.3 ± 0.7% (P < 0.05) at the final visit from a baseline of 9.2 ± 0.5%. Compared to female patients, male patients had a similar drop in Hb(A1c) at 6 months of 1.1%, which persisted until the final visit, when the drop was -1.3% vs 0.3% (P < 0.05). CONCLUSIONS: A single episode of one-to-one diabetic and dietetic education to subjects with poor diabetic control is effective in improving short- and long-term diabetic control for up to 1 year. Male patients were found to have a better response than female patients.

Metastatic breast carcinoma presenting with profound hypocalcemia.

Hypocalcemia is a rare complication of malignancy. We present the case of a 30-year-old lady presenting with a grand mal seizure due to profound hypocalcemia. She was subsequently found to be hypoparathyroid and was diagnosed with metastatic breast carcinoma. Treatment with goserelin and exemestane produced a significant reduction in her tumor load and a correction of her hypocalcemia, which was initially refractory to treatment. We believe this to be the first case of metastatic breast carcinoma actually presenting with hypocalcemia and feel that clinicians should be aware of this rare complication.

Vildagliptin in clinical practice: a review of literature.

Vildagliptin is the second member of the DPP-IV inhibitor class of drugs licensed for the treatment of type 2 diabetes mellitus (T2DM). The novel action of these drugs has promoted a new outlook in the pathobiology of T2DM. This review undertakes to examine the clinical studies published to date, with the aim of evaluating the position of vildagliptin among the drugs that are now available to treat this common dysmetabolic state.

Inference for constrained estimation of tumor size distributions.

SUMMARY: In order to develop better treatment and screening programs for cancer prevention programs, it is important to be able to understand the natural history of the disease and what factors affect its progression. We focus on a particular framework first outlined by Kimmel and Flehinger (1991, Biometrics, 47, 987-1004) and in particular one of their limiting scenarios for analysis. Using an equivalence with a binary regression model, we characterize the nonparametric maximum likelihood estimation procedure for estimation of the tumor size distribution function and give associated asymptotic results. Extensions to semiparametric models and missing data are also described. Application to data from two cancer studies is used to illustrate the finite-sample behavior of the procedure.

Maternal and neonatal influences on, and reproducibility of, neonatal aortic pulse wave velocity.

Aortic pulse wave velocity (aPWV), a noninvasive measure of vascular stiffness, is an independent predictor of cardiovascular disease both before and in overt vascular disease. Its characteristics in early life and its relationship to maternal factors have hardly been studied. To test the hypothesis that infant aPWV was positively related to maternal anthropometry and blood pressure (BP) at 28 weeks gestation, after adjusting for neonatal anthropometry and BP, 148 babies born in Manchester were measured 1 to 3 days after birth. A high reproducibility of aPWV, assessed in 30 babies within 3 days of birth, was found with a mean difference between occasions of -0.04 m/s (95% CI: -0.08 to 0.16 m/s). Contrary to our hypothesis, a significant inverse relation was found between neonatal aPWV (mean: 4.6 m/s) and maternal systolic BP (mean: 108.9 mm Hg; r=-0.57; 95% CI: -0.67 to -0.45) but not maternal height nor weight. Neonatal aPWV was positively correlated with birth length, birth weight, and systolic BP. In multiple regression, neonatal aPWV remained significantly inversely associated with maternal systolic BP (adjusted beta coefficient: -0.032; 95% CI: -0.040 to -0.024; P<0.001), after adjustment for maternal age, birth weight, length, and neonatal BP (all independently and positively related to aPWV) and for gestational age, maternal weight, and height (unrelated). These results suggest that infant aPWV may be a useful index of infant vascular status, is less disturbing to measure than infant BP, and is sensitive to the gestational environment marked by maternal BP.