CONTEXT AND OBJECTIVE:Hyperinsulinemic hypoglycemia is newly recognized as a rare but important complication after Roux-en-Y gastric bypass (GB). The etiology of the syndrome and metabolic characteristics remain incompletely understood. Recent studies suggest that levels of incretin hormones are increased after GB and may promote excessive beta-cell function and/or growth.PATIENTS AND METHODS:We performed a cross-sectional analysis of metabolic variables, in both the fasting state and after a liquid mixed-meal challenge, in four subject groups: 1) with clinically significant hypoglycemia [neuroglycopenia (NG)] after GB surgery, 2) with no symptoms of hypoglycemia at similar duration after GB surgery, 3) without GB similar to preoperative body mass index of the surgical cohorts, and 4) without GB similar to current body mass index of the surgical cohorts.RESULTS:Insulin and C-peptide after the liquid mixed meal were both higher relative to the glucose level achieved in persons after GB with NG compared with asymptomatic individuals. Glucagon, glucagon-like peptide 1, and glucose-dependent insulinotropic peptide levels were higher in both post-GB surgical groups compared with both overweight and morbidly obese persons, and glucagon-like peptide 1 was markedly higher in the group with NG. Insulin resistance, assessed by homeostasis model assessment of insulin resistance, the composite insulin sensitivity index, or adiponectin, was similar in both post-GB groups. Dumping score was also higher in both GB groups but did not discriminate between asymptomatic and symptomatic patients. Notably, the frequency of asymptomatic hypoglycemia after a liquid mixed meal was high in post-GB patients.CONCLUSION:A robust insulin secretory response was associated with postprandial hypoglycemia in patients after GB presenting with NG. Increased incretin levels may contribute to the increased insulin secretory response.

Chronic steroid treatment is known to impair the hypothalamic-pituitary adrenal axis (HPA) but the need to assess HPA function prior to withdrawal of steroid therapy in post-transplant patients has not been uniformly accepted. We evaluated the status of the HPA axis in 48 kidney or kidney-pancreas transplant patients who were considered for possible discontinuation of glucocorticoid therapy using a recently validated dynamic test of HPA integrity, the low-dose (1 microg) adrenocorticotropin (ACTH) test. HPA suppression was detected in 29 (60%) of the patients, four of which had severe hypoadrenalism prohibitive of steroid withdrawal. Neither the duration of steroid treatment nor 8:00 am serum cortisol was a useful marker of hypoadrenalism. 8:00 am cortisol in subjects with normal HPA reserve and subjects with partial hypoadrenalism overlapped considerably but levels <5 microg/dL were indicative of severe hypoadrenalism. Pre-withdrawal diagnosis of partial hypoadrenalism allowed the identification of subjects requiring no further steroid replacement under regular daily circumstances. However glucocorticoid supplementation was prescribed in the event of stress such as infection, exceptional effort, trauma or surgery. Individuals with partial HPA impairment, but not patients with severe HPA suppression, improved upon retesting 3 months later. Patients exhibiting normal response to 1 mcg ACTH enjoyed an uneventful course following steroid withdrawal. Since hypoadrenalism is extremely common in post-transplant patients, we recommend the use of the low-dose ACTH test as a convenient method to identify patients with various degrees of hypoadrenalism prior to steroid withdrawal.

INTRODUCTION:Estrogen therapy (ET), tamoxifen and raloxifene are associated with a two- to three-fold increased risk of venous thrombosis (VT); however, the mechanisms by which each drug increases venous thrombosis propensity are not fully understood. The objectives of this investigation were to compare the effects of these three treatments on hemostasis in a head to head randomized placebo-controlled trial.PATIENTS/METHODS:Ninety-four postmenopausal women were assigned to receive oral estrogen (conjugated equine estrogen [CEE] 0.625 mg, n=23), tamoxifen 20 mg (n=24), raloxifene 60 mg (n=24) or placebo (n=23) daily for 6 months. Blood samples were analyzed for procoagulant factors (prothrombin, factors VII [fVII], VIII [fVIII], IX [fIX] and XI [fXI], D-dimer and von Willebrand factor [vWf]), anticoagulant factors (antithrombin [AT], total and free protein S, protein C and activated protein C [APC] resistance) and fibrinolytic factors (thrombin activatable fibrinolysis inhibitor [TAFI] and plasminogen activator inhibitor-1 [PAI-1]), at baseline and at 6 months of treatment.RESULTS:Estrogen increased factor VII and D-dimer, and decreased antithrombin, total and free protein S and PAI-1. Changes with tamoxifen were distinct from estrogen with increases in factors VIII, IX, vWf and free protein S, and decreases in AT, total protein S, protein C and plasminogen activator inhibitor-1. Raloxifene produced similar effects as tamoxifen, but did not increase factor IX or decrease protein C.CONCLUSIONS:Estrogen, tamoxifen and raloxifene affected hemostasis favoring procoagulation and impairing anticoagulation. The biochemical effects of the selective estrogen receptor modulators (SERMs) were distinct from those of estrogen and differed only subtly from each other.