Glycogen storage disease type III (GSDIII) is a rare disorder of glycogenolysis due to AGL gene mutations, causing glycogen debranching enzyme deficiency and storage of limited dextrin. Patients with GSDIIIa show involvement of liver and cardiac/skeletal muscle, whereas GSDIIIb patients display only liver symptoms and signs. The International Study on Glycogen Storage Disease (ISGSDIII) is a descriptive retrospective, international, multi-centre cohort study of diagnosis, genotype, management, clinical course and outcome of 175 patients from 147 families (86 % GSDIIIa; 14 % GSDIIIb), with follow-up into adulthood in 91 patients. In total 58 AGL mutations (non-missense mutations were overrepresented and 21 novel mutations were observed) were identified in 76 families. GSDIII patients first presented before the age of 1.5 years, hepatomegaly was the most common presenting clinical sign. Dietary management was very diverse and included frequent meals, uncooked cornstarch and continuous gastric drip feeding. Chronic complications involved the liver (hepatic cirrhosis, adenoma(s), and/or hepatocellular carcinoma in 11 %), heart (cardiac involvement and cardiomyopathy, in 58 % and 15 %, respectively, generally presenting in early childhood), and muscle (pain in 34 %). Type 2 diabetes mellitus was diagnosed in eight out of 91 adult patients (9 %). In adult patients no significant correlation was detected between (non-) missense AGL genotypes and hepatic, cardiac or muscular complications. This study demonstrates heterogeneity in a large cohort of ageing GSDIII patients. An international GSD patient registry is warranted to prospectively define the clinical course, heterogeneity and the effect of different dietary interventions in patients with GSDIII.Electronic supplementary materialThe online version of this article (doi:10.1007/s10545-016-9932-2) contains supplementary material, which is available to authorized users.

SummaryThe presence of the 20210A allele of the prothrombin (PT) gene has recently been shown to be a risk factor for venous thromboembolism. This is probably mediated through increased plasma prothrombin levels. The aim of this study was to compare the prevalence of the prothrombin 20210A allele in control subjects and in subjects with recognised thrombophilia and to establish whether the additional inheritance of the PT 20210A allele is associated with an increased risk of venous thromboembolism. 101 subjects with a history of venous thromboembolism and diagnosed as having either factor V Leiden (R506Q) or heritable deficiencies of protein C, protein S or antithrombin were studied. The prevalence of the PT 20210A allele in this group was compared with the results obtained for 150 control subjects. In addition, the relationships were examined between genetic status and the number of documented thromboembolic episodes, and between plasma prothrombin levels and possession of the PT 20210A allele. 8 (7.9%) of the 101 patients were also heterozygous for the PT 20210A allele. This compares with 0.7% in the control subjects (p = 0.005). After exclusion of patients on warfarin, the mean plasma prothrombin of 113 subjects without 20210A was 1.09 U/ml, as compared with 1.32 U/ml in 8 with the allele (p = 0.0002). Among the 101 patients with either factor V Leiden, protein S deficiency, protein C deficiency or antithrombin deficiency, the age adjusted mean (SD) number of venous thromboembolic episodes at diagnosis was 3.7 (1.5) in those with the PT 20210A allele, as compared with 1.9 (1.1) in those without (p = 0.0001). We have demonstrated that the prevalence of the PT 20210A allele is significantly greater in subjects with venous thrombosis and characterised heritable thrombophilia than in normal control subjects and that the additional inheritance of PT 20210A is associated with an increased risk of venous thromboembolism. We have also confirmed that plasma prothrombin levels are significantly greater in subjects possessing the PT 20210A compared with those who do not.

No abstract

Spatially or temporally dense polling remains both difficult and expensive using existing survey methods. In response, there have been increasing efforts to approximate various survey measures using social media, but most of these approaches remain methodologically flawed. To remedy these flaws, this article combines 1,200 state-level polls during the 2012 presidential campaign with over 100 million state-located political tweets; models the polls as a function of the Twitter text using a new linear regularization feature-selection method; and shows via out-of-sample testing that when properly modeled, the Twitter-based measures track and to some degree predict opinion polls, and can be extended to unpolled states and potentially substate regions and subday timescales. An examination of the most predictive textual features reveals the topics and events associated with opinion shifts, sheds light on more general theories of partisan difference in attention and information processing, and may be of use for real-time campaign strategy.