ABSTRACT
Abstract Introduction "Dizziness" is a common complaint in clinical practice that can occur with anyone. However, since the symptom is caused by a wide range of disorders, a general clinician usually faces some difficulty to detect the cause. Objective This study aimed to formulate and validate a simple instrument that can be used to screen and predict the most likely cause of dizziness in Thai outpatients. Methods This study was divided into two phases. Phase I included 41 patients diagnosed with common causes of dizziness to determine the algorithm and construct the "structural algorithm questionnaire version 1". In addition, to test and retest its content validity and reliability until the instrument had an acceptable level of both. Phase II of the study pertained to evaluating its accuracy in clinical trials, 150 patients with dizziness had a face-to-face interview while they were waiting for their medical appointment. Results The degree of agreement between the algorithm results and clinical diagnoses was within an acceptable level (κ = 0.69). Therefore, this algorithm was used to construct the structural algorithm questionnaire version 1. The content validity of the structural algorithm questionnaire version 1 evaluated by seven experts. The content validity index values of the questionnaire ranged from 0.71 to 1.0. The Cohen's kappa coefficient (κ) of intra-rater reliability of the structural algorithm questionnaire version 1 was 0.71. In clinical trials, 150 patients with dizziness had a face-to-face interview while they were waiting for their appointment. The overall agreement between their questionnaire responses and final diagnoses by specialists showed a moderate degree of clinical accuracy (κ = 0.55). Conclusions The structural algorithm questionnaire version 1 had a well-developed design and acceptable quality pertaining to both validity and reliability. It might be used to differentiate the cause of dizziness between vestibular and non-vestibular disorders, especially of outpatients with dizziness symptoms.
Resumo Introdução "Tontura' é uma queixa comum na prática clínica que pode ocorrer com qualquer pessoa. No entanto, como o sintoma pode ser causado por uma grande quantidade de distúrbios, o clínico geral normalmente enfrenta alguma dificuldade em detectar sua causa. Objetivo Formular e validar um instrumento simples que pode ser usado para rastrear e predizer a causa mais provável de tontura em pacientes ambulatoriais tailandeses. Método Este estudo foi dividido em duas fases. A fase I consistiu em determinar o algoritmo, usaram‐se 41 pacientes com diagnóstico de causa comum de tontura, depois construir o questionário de algoritmo estrutural versão 1 (structural algorithm questionnaire version 1) e testar e retestar a validade de seu conteúdo e sua confiabilidade até que o instrumento apresentasse um nível aceitável de ambos. A fase II do estudo consistiu em avaliar a precisão do instrumento em ensaios clínicos, 150 pacientes com tontura foram pessoalmente entrevistados enquanto aguardavam o atendimento médico. Resultados O grau de concordância entre os resultados do algoritmo e os diagnósticos clínicos ficou dentro de um nível aceitável (κ = 0,69). Portanto, esse algoritmo foi usado para construir o questionário de algoritmo estrutural versão 1. A validade de conteúdo do questionário foi avaliada por sete especialistas. Os valores do índice de validade de conteúdo do questionário variaram de 0,71 a 1,0. O coeficiente kappa de Cohen (κ) de confiabilidade intraexaminador foi de 0,71. Em estudos clínicos, 150 pacientes com tontura foram pessoalmente entrevistados enquanto aguardavam a consulta com o médico. A concordância geral entre as respostas ao questionário e os diagnósticos finais dos especialistas mostrou um grau moderado de acurácia clínica (κ = 0,55). Conclusões O questionário de algoritmo estrutural versão 1teve um desenho bem desenvolvido e qualidade aceitável no que diz respeito à validade e confiabilidade. Pode ser usado para diferenciar a causa da tontura entre distúrbios vestibulares e não vestibulares, especialmente em pacientes ambulatoriais com sintomas de tontura.
ABSTRACT
A clinical analysis of 6 patients with pathologically confirmed tonsillar tuberculosis was carried out retrospectively. The subjects comprised three men and three women, ranging in age from 20 to 74 years. All of the patients presented with a sore throat and 5 had lymphadenopathy. Ulcerations, masses and white patches characterized the tonsillar lesions; the pathological findings included caseous granuloma with positive acid-fast bacilli (AFB) in 5 patients and chronic granulomatous inflammation with negative AFB in one patient. Four of the six patients had pulmonary tuberculosis. The three patients who received complete treatment responded well. The presenting symptoms and abnormal tonsillar findings associated with tonsillar tuberculosis are similar to those of malignant tumors and therefore it is difficult to differentiate the two pathologies; moreover, tonsillar tuberculosis often occurs with pulmonary tuberculosis and AIDS and therefore, a chest X-ray and HIV-screening are recommended for all patients with tonsillar tuberculosis.