To develop a machine-based algorithm from clinical and demographic data, physical activity and glucose variability to predict hyperglycaemic and hypoglycaemic excursions in patients with type 2 diabetes on multiple glucose lowering therapies who fast during Ramadan. Thirteen patients (10 males and three females) with type 2 diabetes on 3 or more anti-diabetic medications were studied with a Fitbit-2 pedometer device and Freestyle Libre (Abbott Diagnostics) 2 weeks before and 2 weeks during Ramadan. Several machine learning techniques were trained to predict blood glucose levels in a regression framework utilising physical activity and contemporaneous blood glucose levels, comparing Ramadan to non-Ramadan days. The median age of participants was 51 years (IQR 49-52); median BMI was 33.2 kg/m² (IQR 33.0-35.9) and median HbA1c was 7.3% (IQR 6.7-7.8). The optimal model using physical activity achieved an R² of 0.548 and a mean absolute error (MAE) of 30.30. The addition of electronic health record (ehr) information increased R² to 0.636 and reduced MAE to 26.89 and the time of the day feature further increased R² to 0.768 and reduced MAE to 20.55. Combining all the features together resulted in an optimal XGBoost model with an R² of 0.836 and MAE of 17.47. This model accurately estimated normal glucose levels in 2584/2715 (95.2%) readings and hyperglycaemic events in 852/1031 (82.6%) readings, but fewer hypoglycaemic events (48/172 (27.9%)). The optimal XGBoost model prioritized age, gender, BMI and HbA1c followed by glucose levels and physical activity. Interestingly, the blood glucose level prediction by our model was influenced by use of SGLT2i. XGBoost, a machine learning AI algorithm achieves high predictive performance for normal and hyperglycaemic excursions, but has limited predictive value for hypoglycaemia in patients on multiple therapies who fast during Ramadan.

Objective: To compare pregnancy outcomes in patients with early versus usual gestational diabetes mellitus (GDM).Design: A retrospective cohort study.Settings: The Women’s Hospital, Hamad Medical Corporation, Qatar.Participants: GDM women who attended and delivered in the Women’s Hospital, between January and December 2016. GDM was diagnosed based on the 2013-WHO criteria. The study included 801 patients; of which, 273 E-GDM and 528 U-GDM. Early GDM (E-GDM) and usual GDM (U-GDM) were defined as GDM detected before and after 24 weeks’ gestation, respectively.Outcomes: Maternal and neonatal outcomes and the impact of timing of GDM-diagnosis on pregnancy outcomes.Results: At conception, E-GDM women were older (mean age 33.5±5.4 vs 32.0±5.4 years, p<0.001) and had higher body mass index (33.0±6.3 vs 31.7±6.1 kg/m², p=0.0059) compared with U-GDM. The mean fasting, and 1-hour blood glucose levels were significantly higher in E-GDM vs U-GDM, respectively (5.3±0.7 vs 4.0±0.7 mmol/L, p<0.001 and 10.6±1.7 vs 10.3±1.6 mmol/L, p<0.001). More patients in the U-GDM were managed on diet alone compared with E-GDM (53.6% vs 27.5%, p<0.001). E-GDM subjects gained less weight per week compared with U-GDM (0.02±0.03 vs 0.12±0.03 kg/week, p=0.0274). Maternal outcomes were similar between the two groups apart from a higher incidence of preterm labour (25.5% vs 14.4%; p<0.001) and caesarean section (52.4% vs 42.8%; p=0.01) in E-GDM vs U-GDM, respectively. After correction for covariates; gestational age at which GDM was diagnosed was associated with increased risk of macrosomia (OR 1.06, 95% CI 1.00 to 1.11; p<0.05) and neonatal hypoglycaemia (OR 1.05, 95% CI 1.00 to 1.11; p<0.05).Conclusion: Our data support the concept of early screening and treatment of GDM in high-risk patients. More data are needed to examine the optimal time for screening.

Diabetes first detected during pregnancy is currently divided into gestational diabetes mellitus (GDM) and diabetes mellitus (DM)- most of which are type 2 DM (T2DM). This study aims to define the prevalence and outcomes of diabetes first detected in pregnancy based on 75-gram oral glucose tolerance test (OGTT)using the recent WHO/IADPSG guidelines in a high-risk population. This is a retrospective study that included all patients who underwent a 75 g (OGTT) between Jan 2016 and Apr 2016 and excluded patients with known pre-conception diabetes. The overall prevalence of newly detected diabetes in pregnancy among the 2000 patients who fulfilled the inclusion/exclusion criteria was 24.0% (95% CI 22.1–25.9) of which T2DM was 2.5% (95% CI 1.9–3.3), and GDM was 21.5% (95% CI 19.7–23.3). The prevalence of newly detected diabetes in pregnancy was similar among the different ethnic groups. The T2DM group was older (mean age in years was 34 ±5.7 vs 31.7±5.7 vs 29.7 ±5.7, p<0.001); and has a higher mean BMI (32.4±6.4 kg/m2 vs 31.7±6.2 kg/m2 vs 29.7± 6.2 kg/m2, p< 0.01) than the GDM and the non-DM groups, respectively. The frequency of pre-eclampsia, pre-term delivery, Caesarean-section, macrosomia, LGA and neonatal ICU admissions were significantly higher in the T2DM group compared to GDM and non-DM groups. Diabetes first detected in pregnancy is equally prevalent among the various ethnic groups residing in Qatar. Newly detected T2DM carries a higher risk of poor pregnancy outcomes; stressing the importance of proper classification of cases of newly detected diabetes in pregnancy.

Psoriasis is characterized by hyperproliferation and defective differentiation of keratinocytes (KCs). Patients with psoriasis are at a high risk of developing diabetes and cardiovascular diseases. The debate on the genetic origin of psoriasis pathogenesis remains unresolved due to lack of suitable in vitro human models mimicking the disease phenotypes. In this study, we provide the first human induced pluripotent stem cell (iPSC) model for psoriasis carrying the genetic signature of the patients. iPSCs were generated from patients with psoriasis (PsO-iPSCs) and healthy donors (Ctr-iPSCs) and were efficiently differentiated into mature KCs. RNA sequencing of KCs derived from Ctr-iPSCs and PsO-iPSCs identified 361 commonly upregulated and 412 commonly downregulated genes. KCs derived from PsO-iPSCs showed dysregulated transcripts associated with psoriasis and KC differentiation, such as HLA-C , KLF4 , chemokines, type I interferon-inducible genes, solute carrier family, IVL , DSG1 , and HLA-DQA1 , as well as transcripts associated with insulin resistance, such as IRS2 , GDF15 , GLUT10 , and GLUT14 . Our data suggest that the KC abnormalities are the main driver triggering psoriasis pathology and highlights the substantial contribution of genetic predisposition in the development of psoriasis and insulin resistance.

Rates of obesity are increasing world-wide with an estimated 1billion people projected to be obese by 2030 if current trends remain unchanged. Obesity currently considered one of the most significant associated factors of non-communicable diseases poses the greatest threat to health. Diabetes mellitus is an important metabolic disorder closely associated with obesity. It is therefore expected that with the increasing rates of obesity, the rates of diabetes in pregnancy will also be rising. This disorder may pre-date pregnancy (diagnosed or undiagnosed and diagnosed for the first time in pregnancy) or may be of onset in pregnancy. Irrespective of the timing of onset, diabetes in pregnancy is associated with both fetal and maternal complications. Outcomes are much better if control is maximised. Early diagnosis, multidisciplinary care and tailored management with optimum glycaemic control is associated with a significant reduction in not only pregnancy complications but long-term consequences on both the mother and offspring. This review brings together the current understanding of the pathogenesis of the endocrine derangements that are associated with diabetes in pregnancy how screening should be offered and management including pre-pregnancy care and the role of newer agents in management.

Key content: Lipid and triglyceride levels increase in pregnancy and do not pose problems for most women. However, pregnancy‐specific factors and genetic aberrations, especially mutations, may result in supraphysiological hypercholesterolaemia (HC) and severe hypertriglyceridaemia (sHTG). HC and sHTG are associated with complications in pregnancy, including acute pancreatitis, hyperviscosity syndrome and pre‐eclampsia. Abnormally high levels of lipids and triglycerides also affect fetal growth and the onset of gestational diabetes. The clinical presentation and diagnosis of HC and sHTG are varied. Management options include lifestyle and dietary restriction/modification, omega‐3, bile‐acid sequestrants, fenofibrate, statins and plasmapheresis. Learning objectives: To understand the physiological changes that occur in lipids in pregnancy and how these are modulated. To appreciate the importance of severe HC and HTG in pregnancy, including their effects on maternal morbidity and mortality. To understand the various approaches to management and the importance of multidisciplinary management. Ethical issues: Should every woman be screened for HTG and HC during pregnancy? How safe are statins and other treatments for lipid disorders in pregnancy? Should women with very severe HTG be discouraged from getting pregnant?