Search

Background and Aims: TC-325 (Hemospray, Cook Medical, Winston-Salem, NC) is an inorganic hemostatic powder recently approved by the U.S. Food and Drug Administration. This study aimed to examine the effectiveness, safety, and predictors of TC-325 failure in a large real-life cohort. Methods: This was a retrospective study conducted at 21 Spanish centers. All patients treated with TC-325 until September 2018 were included. The primary outcome was treatment failure, defined as failed intraprocedural hemostasis or recurrent bleeding within the first 30 postprocedural days. Secondary outcomes included safety and survival. Risk and predictors of failure were assessed via competing-risk models. Results: The cohort comprised 261 patients, of whom 219 (83.9%) presented with upper gastrointestinal bleeding (GIB). The most common causes were pepticulcer (28%), malignancy (18.4%), and therapeutic endoscopy-related GIB (17.6%). TC-325 was used as rescue therapy in 191 (73.2%) patients. The rate of intraprocedural hemostasis was 93.5% (95% confidence interval [CI], 90%-96%). Risks of TC-325 failure at postprocedural days 3, 7, and 30 were 21.1%, 24.6%, and 27.4%, respectively. On multivariate analysis, spurting bleeding (P = .004), use of vasoactive drugs (P = .02), and hypotension (P = .008) were independent predictors of failure. Overall 30-day survival was 81.9% (95% CI, 76%-86%) and intraprocedural hemostasis was associated with a better prognosis (adjusted hazard ratio, 0.29; P = .006). Two severe adverse events were noted. Conclusion: TC-325 was safe and effective for intraprocedural hemostasis in more than 90% of patients, regardless of the cause or site of bleeding and its use as rescue therapy. In this high-risk cohort treated with TC-325, the 30-day failure rate exceeded 25% and was highest with spurting bleeding or hemodynamic instability.

Fundamento y objetivos: la automonitorización de glucemia capilar en pacientes con diabetes tipo 2 podría indicar el grado de control metabólico. Los objetivos del presente trabajo fueron evaluar en qué momento del día las glucemias capilares mostraban mejor correlación con las cifras de HbA1c y establecer puntos de corte de glucemia capilar para predecir cifras de HbA1c por encima o por debajo de 7,5%. Diseño: estudio descriptivo transversal. Material y métodos: a 42 pacientes con diabetes tipo 2 se les realizaron 3 perfiles glucémicos capilares domiciliarios de 6 puntos (antes y 2 horas después de desayuno, comida y cena), el día anterior, 4 semanas antes y 8 semanas antes de la determinación de una cifra de HbA1c. Resultados: las glucemias que mostraban mejor correlación con las cifras de HbA1c eran las de antes de desayuno (r=0,44; p=0,0047) y las de antes de cena (r=0,44; p=0,0041). La curva ROC de glucemia de antes de desayuno para predecir HbA1c con un punto de corte de 7,5% tuvo un área bajo la curva de 0,694 (IC 95%: 0,524 - 0,833); el valor óptimo de glucemia elegido fue de 145 mg/dl. La curva ROC de glucemia de antes de cena tuvo un área bajo la curva de 0,739 (IC 95%: 0,574 - 0,866); el valor glucémico elegido fue de 141 mg/dl. Conclusiones: las glucemias que mejor correlacionan con HbA1c son las de antes de desayuno y cena. Dadas las correlaciones débiles entre HbA1c y glucemia capilar, la automonitorización debe considerarse orientativa, no sustitutiva, de la HbA1c. Introduction and objectives: self monitoring of blood glucose in patients with type 2 diabetes could reflect the quality of metabolic control. The objective was to evaluate in which moments of the day capillary blood glucose showed the best correlation with glycated hemoglobin (HbA1c) values and to establish capillary blood glucose cut-off in order to predict HbA1c values higher and lower than 7.5%. Design: cross-over, descriptive study. Material and methods: three capillary blood glucose profiles were practised at home to 42 patients with type 2 diabetes mellitus: one of them on the day before HbA1c determination and the other ones four and eight weeks before that HbA1c determination. Each profile included six checking points, before and two hours after breakfast, lunch and dinner. Results: the best correlation between capillary blood glucose and HbA1c was obtained before breakfast (r=0,44; p=0.0047) and before dinner (r=0.44; p=0.0041). The ROC curve of blood glucose obtained before breakfast to predict HbA1c with 7.5% cut-off had an area under graph of 0.694. The optimal blood glucose value choosed was 145 mg/dl.The ROC curve of blood glucose obtained before dinner had an under graph area of 0.739; the blood glucose value choosed was 141 mg/dl. Conclusions: blood glucoses that showed the best correlation with HbA1c values were the ones before breakfast and dinner. Due to poor correlation between HbA1c and capillary blood glucoses, selfmonitoring blood glucose should be considered orientative and not sustitutive of HbA1c determination.