Avaliação da incidência, prevalência e progressão de varizes esofagianas em pacientes pediátricos cirróticos.

Abstract

Introduction: The upper gastrointestinal bleeding (UGB) secondary to the rupture of esophageal varices (EV) and gastric varices (GV) is the leading cause of mortality and morbidity in the children with cirrhosis. It’s diagnosis is made with gastrointestinal edoscopy. The high-risk varices (HRV) are submitted to endoscopic variceal band ligation. The correct identification of the pacient with HRV and of the frequency of screening and surveillance of EV would spare a number of children of invasive procedures.and allow the primary prophylaxis of bleeding, keeping a cost effective approach. Objectives: To evaluate the factors associated with the presence of HRV in the pediatric cirrhotic patient. To evaluate the incidence of HRV and the progression of esophageal varices with small caliber and the factors associated with them in the cirrhotic pediatric patients at the ambulatory of pediatric hepatology of the Hospital das Clínicas UFMG. To evaluate the safety of the protocol current instituted in the hospital. Patients and Methods: The study included pediatric patients diagnosed with cirrhosis at Hospital das Clínicas UFMG from 2004 to 2018, that were submitted to screenin and surveillance of EV with endoscopy. The sample was nonprobabilistic and composed of 158 children. According to protocol the patients with portal hypertension and without UGB were screened with endoscopy and in those with EV of small caliber the surveillance endoscopy were repeated yearly, those without EV the surveillance endoscopy were repeated at two year intervals. The clinical condition and the laboratory tests were registered in a interval of three months from the endoscopy. The factors associated with the presence and the emergence of HRV were analyzed. The avaliation of the factors associated with the presence of HRV were made with diagnostic tests. Investigation of intervenient factors for emergence of HRV was obtained using uni- and multivariate analysis through logistic regression. Kaplan-Meier curves and Cox model were used to evaluate survival rates free of HRV. Results: In the analysis of the non invasive factors associated with HRV, the best predictor was the Risk Score, when used the cutoff value of -1,2 showed an sensitivity of 0,90 and specificity of 0,52, with positive predictive value of 31% and negative predictive value of 96%. The association of Risk Score and APRI had a sensitivity of 0,97 and specificity of 0,34. There was 23 (18%) new cases of HRV, with higher progression among the patients with biliary atresia (BA) when compared with patients with autoimmune hepatites. The presence of EV of small caliber in the first endoscopy was related with the occurrence of HRV (OR:3,43; IC95%: 1,324 – 8,902; p:0,011), greater impact was found in the BA subgrup (OR:7,309; IC95%: 1,444 – 37,006; p:0,016). In the Cox regression when the first endoscopy was perfomed before the age of six, there was a risk 15,5 times greater of emerging with HRV (p<0,001). The presence of small varices in the first endoscopy was associated with a risk 3,70 times greater (p=0,003). There were three UGB in the period with no deaths. Conclusion: In our cohort the patient with Risk Score less than -1,2 and APRI less than 1,4 showed a low risk of HRV and could be spared of endoscopy. The intervals between endoscopy proposed for adults are safe in children but can result in a high number of unnecessary exams. The EV progress in the cirrohtic children, but in diferente pace according to etiology of liver disease and the presence of small caliber EV. As with adults stratifying risk and individualizing care seens to be the best approach for children with cirrhosis and portal hypertension.