Fatores de risco associados à intervenção com injeção intravítrea de anti-VEGF em pacientes com edema macular diabético / Risk factors related to the intervention with intravitreal anti-VEGF injection in patients with diabetic macular edema

Resumo Objetivo: Propor um modelo de regressão logística para auxiliar na decisão de realização da injeção intravítrea (IIV) de anti-VEGF, a partir da quantificação e hierarquização dos fatores de risco que compõem o perfil dos indivíduos diabéticos. Métodos: Trata-se de estudo transversal, observacional e inferencial, realizado em três instituições da Paraíba, de julho de 2015 a setembro de 2016. O modelo de regressão logística foi utilizado para obtenção do modelo preditivo e os dados foram analisados no software R®. Resultados: Foram avaliados 80 pacientes com diabetes tipo 1 ou 2, maiores de 18 anos, dos quais 57,5% não tiveram indicação de IIV e 42,5% receberam indicação deste tratamento. No grupo com edema macular diabético (EMD), a média de idade foi de 60,65 anos, sendo 52,94% do sexo feminino. Ainda nesse grupo, a maioria apresentou retinopatia diabética não-proliferativa severa ou retinopatia proliferativa (79,41%). Foram identificados como fatores de risco para EMD: o indivíduo ser aposentado (OR=5,22; p-valor 0,05), ter histórico pessoal de retinopatia diabética (OR=20,27; p-valor 0,006) e de tratamento prévio com anti-VEGF (OR=23,23; p-valor 0,002). Conclusão: Os resultados da pesquisa evidenciaram que um indivíduo diabético com baixa visual e apresentando esses três fatores deve ser encaminhado o quanto antes ao especialista, pois possui, com 91,17% de acerto, risco de apresentar EMD com necessidade de IIV de anti-VEGF. Essa ferramenta pode servir como coadjuvante na tomada de decisão, sobretudo do não-retinólogo, a fim de encaminhar para diagnóstico e tratamento precoces os indivíduos com EMD, o que pode ser decisivo na prevenção da perda visual irreversível nesses pacientes.

Abstract Purpose: To propose a predictive model to aid in the decision to perform the intravitreal anti-VEGF injection, based on the risk factors quantification and hierarchy presented by diabetic patients. Methods: It is a cross-sectional, observational and inferential study carried out in three institutions in Paraíba from July 2015 to September 2016. The logistic regression model was used to obtain the predictive model and data were analyzed in R(r) software. Results: Eighty patients with type 1 or 2 diabetes, over 18 years of age, were included, 57.5% of whom had no indication of IIV and 42.5% received an indication of this treatment. In the group with diabetic macular edema (DME), the mean age was 60.65 years, of which 52.94% were female. In this group, the majority presented severe non-proliferative diabetic retinopathy or proliferative retinopathy (79.41%). The main risk factors for DME were: be retired (OR = 5.22, p-value0.05), had a personal history of diabetic retinopathy (OR = 20.27, p-value 0.006), and previous treatment with anti-VEGF (OR = 23.23, p-value 0.002). Conclusion: The results of the research showed that a diabetic patient with low visual acuity and presenting these three factors should be referred as soon as possible to the specialist, since he presents a risk of presenting DME with need for anti-VEGF IIV, with 91.17% of accuracy. This tool can serve as an adjunct to decision making, especially the nonretinologist, in order to refer individuals with EMD to early diagnosis and treatment, which may be crucial in preventing irreversible visual loss in these patients.

Evaluation of transition zone prostate cancer by Logistic regression of prostate imaging reporting and data system version 2 combined with prostate specific antigen / 中国医学影像技术

Objective To establish the Logistic regression model by reporting and data system version 2 (PI-RADS v2)and prostate specific antigen (PSA),and to evaluate the diagnostic efficiency in transition zone prostate cancer (PCa).Methods MRI and PSA data of 33 patients with PCa and 54 patients with non-PCa confirmed by pathology were analyzed retrospectively.The PI-RADS v2 was used to evaluate the risk of 2 groups (from low to high as 1 to 5 points).Total PSA (t-PSA),free to total PSA ratio (f-PSA/t-PSA),PSA density (PSAD) and PI-RADS v2 scores were compared between 2 groups.The Logistic regression models were established with parameters which were significantly different between 2 groups.The Logistic regression was divide into three protocols:PI-RADS v2-+ t-PSA (A),PI-RADS v2 + f-PSA/t-PSA (B),PI-RADS v2+PSAD (C).The ROC curves were constructed by the new parameters Logit (P) and PI-RADS v2 scores for assessing the diagnostic efficiency.Results The t-PSA,f-PSA/t-PSA,PSAD and PI-RADS v2 scores had significant differences between the 2 groups (all P＜0.01).Predictive multivariate model of A,B,C was established as Logit (P)=-8.682+1.507 PI-RADS v2+0.234 t-PSA (x2=65.993,P＜0.01),Logit(P)=-5.425+1.906 PI-RADS v2 13.921 f-PSA/t-PSA (x2 =65.993,P＜0.01),Logit(P)=-7.534+1.045 PI-RADS v2+13.318 PSAD (x2 =74.036,P＜0.01),their area underthe curve (0.945,0.919,0.960) were all higher than that of PI-RADS v2 score (0.861,all P ＜0.01).The protocol C had the best diagnostic efficiency,and the sensitivity and specificity were 87.88 ％ and 92.59 ％.The sensitivity and specificity of PI-RADS v2 score were 87.88％ and 77.78％.Conclusion The diagnostic efficiency of the Logistic regression model which includes the PI-RADS v2 score and PSA are superior to the PI-RADS v2 score alone for transition zone PCa,which can provide a reliable basis for patients whether need biopsy or not.

Logistic regression analysis of ultrasonography in diagnosis of malignant thyroid nodules / 中国介入影像与治疗学

Objective To investigate the value of conventional ultrasound and CEUS in diagnosis of thyroid nodules with Logistic regression models.Methods A total of 218 cases of thyroid nodules (74 cases of malignant,144 cases of benign nodules) confirmed by pathology were enrolled.The boundary,morphology,anteroposterior and transverse diameter ratio,microcalcification,internal echogenicity,blood distribution and enhanced pattern of nodules were observed and analyzed with univariate analysis.The Logistic regression model was established with parameters which were significantly different of those features,and the receiver operating characteristic curves (ROC) were constructed.Results Hypoechoic,irregular morphology,blurred boundary,anteroposterior and transverse diameter ratio≥ 1,microcalcifications,blood distribution (Ⅰ,Ⅱ),heterogeneous enhanced pattern and low enhanced were significantly prognostic factors (all P＜0.01).Irregular morphology,microcalcifications,heterogeneous enhanced and low enhanced were independent prognostic factors (all P＜0.05).The accuracy of Logistic regression model was 82.57％,and the area under ROC curve was 0.906.Conclusion The Logistic regression model of boundary,morphology,anteroposterior and transverse diameter ratio,microcalcifica tions,internal echogenicity,blood distribution and enhanced pattern can help to diagnose malignant thyroid nodules.

Fatores associados à mortalidade hospitalar na rede SUS do Rio Grande do Sul, em 2005: aplicação de modelo multinível / Factors associated with hospital mortality in Rio Grande do Sul SUS network in 2005: application of a Multilevel Model

OBJETIVO: Avaliar a mortalidade hospitalar por meio de análise multinível utilizando dados disponíveis no Sistema de Informações Hospitalares do Sistema Único de Saúde. MÉTODOS: Estudo transversal com dados de internações obtidas das Autorizações de Internação Hospitalar do Rio Grande do Sul no ano de 2005. A modelagem foi realizada por meio de regressão logística multinível, utilizando variáveis do nível individual (internações) e do nível contextual (hospitais). Analisou-se a variabilidade causada por variáreis individuais no nível hospitalar, bem como a participação do perfil dos hospitais na taxa de mortalidade hospitalar. RESULTADOS: A taxa bruta de mortalidade calculada para o conjunto de hospitais foi de 6,3 por cento. As variáveis uso de Unidade de Terapia Intensiva e idade foram os principais preditores para óbito hospitalar no nível individual. As variáveis de contexto que se relacionaram mais intensamente com o óbito hospitalar foram: porte do hospital, natureza jurídica e média de permanência. A chance de óbito em hospital de grande porte é 1,85 vezes a chance do hospital de pequeno porte e no hospital de médio porte é 1,69 vezes a chance do hospital de pequeno porte. Os hospitais públicos apresentam 67 por cento mais chances de óbito hospitalar do que os privados. CONCLUSÕES: O perfil hospitalar tem papel importante na mortalidade hospitalar do Sistema de Informações Hospitalares do Sistema Único de Saúde. A análise multinível deve ser empregada para a estimação da contribuição do perfil dos hospitais na mortalidade hospitalar.

OBJECTIVE: To use a multilevel analysis methodology to evaluate hospital mortality from the data available in the Hospital Information System of the National Unified Health System. METHODS: Cross-sectional study with data obtained from Authorization Forms for Hospital Admissions in Rio Grande do Sul, Brazil in 2005. The modeling was performed using multilevel logistic regression, with variables from the individual level (hospital admissions) and the context level (hospital profile). The variability originated from individual variables was analyzed as well as the participation of the profile of hospitals in the rate of hospital mortality. RESULTS: The crude death rate calculated for all hospitals was 6.3 percent. The variables "Use of Intensive Care Unit" followed by "Patient Age" were the main predictors for hospital death at the individual level. The context variables that were related most closely to hospital death (outcome) were: size of hospital, legal nature, and average length of stay. The OR for deaths at large hospitals was 1.85 times the odds for small hospitals and the OR for medium hospitals was 1.69 times the odds for small ones. The chance of deaths in public hospitals was 67 percent higher than in private ones. CONCLUSIONS: The hospital profile has an important role in hospital mortality in the Hospital Information System of the National Unified Health System. Multilevel analysis should be used to estimate the contribution of the profile of mortality in hospitals.