Findings of a multifaceted study have provided new insight into the role of lipolytic pathways in metabolic health. Using gene sequencing in 24 people in an Amish population who had extremely high or low fasting triglycerides, investigators identified a frameshift deletion in exon 9 of the gene encoding hormone-sensitive lipase (LIPE) in one individual with high trigyceride concentrations. In a cohort of 2738 Amish individuals, 140 were heterozygous for the deletion and one was homozygous. Compared with non-carriers, carriers of the frameshift deletion had higher insulin resistance (p=0·005), serum triglyceride concentrations (p=0·003), and hepatic fat content (p<0·001), and lower HDL cholesterol concentrations (p<0·001). The prevalence of type 2 diabetes in heterozygous carriers was 11·4% compared with 6·8% in individuals who were not carriers (p=0·02). According to investigators, bariatric surgery might be associated with long-term remission of diabetes. Using data from the Swedish Obese Study, the researchers compared long-term those for 343 patients with type 2 diabetes who underwent surgery with outcomes for a control group of 260 obese patients with type 2 diabetes. At 2 years, a greater proportion of people in the surgery group were in diabetes remission than in the control group (72.3% vs 16.4%; unadjusted odds ratio 13·3, 95% CI 8·5–20·7). At 15 years follow-up, the proportion of patients in the surgery group who were in remission decreased (30·4%), but these patients were still more likely to be in remission than were those in the control group (odds ratio 6·3, 95% CI 2·1–18·9). Over a 20 year follow-up period, the surgery group was also less likely to develop macrovascular and microvascular complications. A study by Maahs and colleagues has shown differences in the management of children with type 1 diabetes from either the USA or Germany and Austria. Using data from the US T1DX registry (674 patients) and German–Austrian DPV registry (1948 patients), characteristics and clinical outcomes for children younger than 6 years with type 1 diabetes were compared. Mean HbA1c concentrations in children in T1DX were higher than those in individuals in DPV (8·2% vs 7·4%, p<0·001); notably a larger proportion of patients in DPV were receiving insulin pump therapy compared with patients from T1DX (74% vs 50%, p<0·001). The authors of a US retrospective cohort study have attempted to address whether people with type 2 diabetes who are failing on metformin monotherapy should receive insulin or a sulfonylurea as second-line therapy. Investigators ascertained rates of cardiovascular events and death in veterans started on metformin between 2001 and 2008 and who later received a sulfonylurea or insulin. Participants were observed until 2011. The analysis included 2436 patients receiving metformin plus insulin and 12 180 receiving metformin plus a sulfonylurea. There was increased risk of a composite of cardiovascular outcomes and all-cause mortality for individuals in the insulin group compared with individuals in the sulfonylurea group (42·7 vs 32·8 events per 1000 person-years; adjusted hazard ratio 1·30, 95% CI 1·07–1·58). The risk of death from any cause was lower for individuals in the sulfonylurea group than for those in the insulin group (22·7 vs 33·7 per 1000 person-years; adjusted hazard ratio 1·44, 95% CI 1·15–1·79). When added to metformin, GLP-1 receptor agonist albiglutide might result in greater improvements in HbA1c concentration than addition of DPP-4 inhibitor sitagliptin or sulfonylurea glimepiride. In the 104 week, double-blind, placebo and active controlled HARMONY 3 trial, patients with type 2 diabetes taking metformin were randomised to receive weekly albiglutide (n=302), daily sitagliptin (n=302), daily glimepiride (n=307), or placebo (n=101). Participants receiving albiglutide had greater reductions in HbA1c at 104 weeks than did those receiving sitagliptin (mean difference −0·4%; p=0·0001), glimepiride (–0·3%; p=0·0033), or placebo (–0·9%; p<0·0001). The rate of adverse events and serious adverse events was generally similar between the four groups. Results from a randomised clinical trial have provided some hope for young women with primary ovarian insufficiency. In the study of women with primary ovarian insufficiency aged between 18 and 42 years, investigators assessed the effect of treatment with transdermal estradiol, cyclical oral progesterone, and placebo (n=72) versus treatment with estradiol, progesterone, and transdermal testosterone (n=73) on bone mineral density. They also included a control group (n=70) of women with normal ovarian function. Although the trial was terminated early because of futility of testosterone on the primary outcome, women receiving estradiol, progesterone, and placebo had about a 2·45% gain in bone mineral density at the femoral neck compared with a 2·61% loss in control individuals (p=0·0002) during the 3 year follow-up.