Sensitivity and specificity of the Gastrointestinal Short Form Questionnaire in diagnosis of gastroesophageal reflux disease

Introduction: Gastrointestinal Short Form Questionnaire (GSFQ) is a questionnaire for gastroesophageal reflux disease (GERD) diagnosis, with a version in Spanish language, not yet compared to an objective test. Aims: To establish GSFQ diagnostic performance against 24-hour pH monitoring carried out in two tertiary care hospitals. Methods: Consecutive adult patients with typical GERD symptoms (heartburn, regurgitation) referred for pH monitoring fulfilled the GSFQ (score range 0-30, proportional to probability of GERD). Diagnosis of GERD was established when acid exposure time in distal esophagus was superior to 4.5% or symptom association probability was greater than 95%. Receiver-operator characteristic (ROC) curves were calculated and best cut-off score determined, with corresponding sensitivity, specificity and likelihood ratios (LR) (95% confidence interval for each). Results: One hundred and fifty-two patients were included (59.9% women, age 47.9 ± 13.9; 97.4% heartburn; 71.3% regurgitation). pH monitoring was abnormal in 65.8%. Mean GSFQ score was 11.2 ± 6. Area under ROC was 56.5% (47.0-65.9%). Optimal cut-off score was 13 or greater: sensitivity 40% (30.3-50.3%), specificity 71.2% (56.9-82.9%), positive LR 1.39 (0.85-2.26) and negative LR 0.84 (0.67-1.07). Exclusion of questions 1 and 3 of the original GSFQ, easily interpreted as referred to dyspepsia and not GERD, improved only marginally the diagnostic performance: AUROC 59.1%. Conclusion: The GSFQ does not predict results of pH monitoring in patients with typical symptoms in a tertiary care setting.

Nivel de correlación en la evaluación de la intubación traqueal efectuada por anestesiólogos docentes y residentes de anestesiología

Introducción. La educación médica ha experimentado numerosos cambios. Uno de éstos es el uso de la autoevaluación para establecer la progresión de los estudiantes. Hasta donde sabemos, en la realidad chilena no existen hasta el momento estudios sobre la validez y confiabilidad del uso de la autoevaluación en el proceso educacional. Sujetos y métodos. Las intubaciones traqueales efectuadas durante cuatro meses por los residentes de anestesia de la Pontificia Universidad Católica de Chile se evaluaron utilizando el instrumento 'observación directa de procedimientos y habilidades' (DOPS). Una vez intubado el paciente, el anestesiólogo a cargo efectuó la evaluación inmediata de la técnica de intubación del residente. A su vez, empleando el mismo instrumento, el residente efectuó una autoevaluación del mismo procedimiento. El nivel de acuerdo entre anestesiólogo y residente se estableció mediante el índice de correlación intraclase (ICC). Se realizó una encuesta de opinión sobre el método entre los participantes. Resultados. Se evaluaron 585 intubaciones. Participaron 31 residentes y 30 anestesiólogos. El ICC general fue 0,423. Éste disminuyó a medida que aumentaba el nivel de experiencia de los estudiantes, y alcanzó el nivel más bajo en los residentes de tercer año, con un ICC de 0,39. Anestesiólogos y residentes manifestaron satisfacción con el método, pero poca experiencia previa. Conclusiones. El nivel de concordancia entre las evaluaciones y autoevaluaciones fue bajo. La falta de experiencia previa de los participantes pudo influir en los resultados. Para utilizar adecuadamente instrumentos de autoevaluación, éstos deberían implementarse desde etapas tempranas de la formación.

Endoscopic band ligation without resection in selected patients for small and superficial upper gastrointestinal tract lesions

Background and aim: The aim of this study was to evaluate the efficacy of endoscopic band ligation (EBL) in carefully selected patients who would benefit from this method of resection. Methods: Patients with early upper gastrointestinal and small (< 15 mm) lesions treated with EBL (Duette® Multi-Band Mucosectomy) were prospectively recruited and retrospectively analyzed between 2010 and 2015. All cases were discussed in a multidisciplinary cancer committee and it was concluded that, owing to patient conditions, surgery was not possible and that not conducting histology would not change the clinical management. A first endoscopic control with biopsies was planned at 4-8 weeks. If there was no persistence of the lesion, new controls were programmed at 6 and 12 months. Results: The group (n = 12) included 5 esophagus lesions (adenosquamous carcinoma, n = 1; carcinoma squamous, n = 2; adenocarcinoma, n = 2); 4 gastric lesions (high grade dysplasia, n = 1; adenocarcinoma, n = 2; neuroendocrine tumor [NET], n = 1), and 3 duodenal lesions (NETs) (n = 3). The mean tumor diameter was 9.6 ± 2.8 mm (range 4-15). Only one minor adverse event was described. At first follow-up (4-8 weeks), there was 91.6% and 75% of endoscopic and histological remission, respectively. At 6-month follow-up there was 70% of both endoscopic remission and negative biopsies. And at 12 months, there was 100% and 75% of endoscopic and histological remission, respectively. Persisting lesions were T1 cancers. The median follow-up was 30.6 months. Conclusion: EBL without resection is an easy and safe technique that should be considered in patients with multiple morbidities and small superficial UGI lesions.