Risk factors for early re-bleeding and associated hospitalization in patients with colonic diverticular bleeding.

AIM: The annual incidence of colonic diverticular bleeding is increasing, but treatments are not yet well established. Here we aimed to identify the risk factors for early re-bleeding and to determine the associated duration of hospitalization. METHOD: Records of 90 emergent patients with colonic diverticular bleeding between 1999 and May 2012 were retrospectively reviewed. They were divided into an early re-bleeding within 1 month group (n = 24) and a no re-bleeding group (n = 66) and we investigated the risk factors for early re-bleeding. In the former group, we calculated the time from the first haemostasis to early re-bleeding and the associated duration of hospitalization. RESULTS: Univariate analysis showed that there were significantly more patients with signs of shock (P = 0.00055) and active bleeding on the first colonoscopy after admission (P = 0.020) in the early re-bleeding group. Multivariate conditional logistic regression analysis using stepwise variable selection showed that signs of shock on admission (odds ratio, 5.23; 95% confidence interval, 1.84-14.90; P = 0.0019) remained statistically significant. All patients who re-bled without signs of shock (n = 7) and 16 of 17 with signs of shock re-bled within 126 h (5.25 days) of initial hospitalization. CONCLUSION: Shock was an independent risk factor for early re-bleeding. The associated duration of hospitalization was 6 days.

Changes in serum S100A12 and sRAGE associated with improvement of the PaO(2)/FiO(2) ratio following PMX-DHP therapy for postoperative septic shock.

BACKGROUND: Endotoxin (Et) adsorption therapy with a column of polymyxin B-immobilized fibers (PMX) is effective in improving the partial pressure of arterial oxygen/fraction of inspired oxygen ratio (PaO(2)/FiO(2) ratio) and increasing mean arterial blood pressure (MAP) in sepsis. S100A12 and soluble receptor for advanced glycation end product (sRAGE) are useful as early markers of acute lung injury. PURPOSE: To investigate the effect of improving the PaO(2)/FiO(2) ratio by PMX-direct hemoperfusion (PMX-DHP) on production of S100A12 and sRAGE. SUBJECTS AND METHODS: Sepsis patients after surgery for perforation of the lower gastrointestinal tract were adopted as the subjects. We retrospectively reviewed the cases of 20 patients on mechanical ventilation and continuous administration of norepinephrine. We recorded PaO(2)/FiO(2) ratio, MAP, and norepinephrine doses. S100A12, sRAGE, and Et levels were measured before and after PMX-DHP. RESULTS: The PaO(2)/FiO(2) ratio and MAP improved significantly after PMX-DHP (p < 0.05). S100A12 and Et decreased significantly after PMX-DHP (p < 0.05). No differences were observed in sRAGE. CONCLUSION: S100A12 is useful as a marker that reflected improvement in the PaO(2)/FiO(2) ratio after PMX-DHP. We consider PMX-DHP to be useful as adjunctive therapy for sepsis that reduces the Et and corrects the pathology in the early stage.

Preliminary study on glucose control with an artificial pancreas in postoperative sepsis patients.

BACKGROUND: Glucose control is essential to avoid hypoglycemia in postoperative patients. AIM: To conduct a preliminary examination to evaluate the feasibility of the use of an artificial pancreas for glucose control as well as the accuracy of assessment by the artificial pancreas of the insulin dose required. SUBJECTS AND METHODS: Glucose control using an artificial pancreas was undertaken in 8 postoperative sepsis patients. The blood glucose level was set at 80-150 mg/dl. Blood glucose levels over time, insulin dose requirements, and occurrence of hypoglycemia (≤40 mg/dl) were recorded for each patient. The patients were divided into 2 groups based on the total insulin dose they received over the 7 days (HG, n = 4: consisting of patients who required a higher insulin dose; LG, n = 4: patients who required a lower insulin dose). The data of the 2 groups were analyzed retrospectively. RESULTS: The blood glucose level before glucose control was 203.3 ± 9.9 mg/dl and could be controlled in all patients to within the target range. No hypoglycemia events were recorded for any of the patients. The insulin dose in the HG and LG groups was 21,824.8 ± 6,030.4 and 6,254.5 ± 3,402.3 mU/kg (p < 0.05). CONCLUSIONS: Accurate glucose control could be achieved with the artificial pancreas.