Evaluation of normality and reproducibility parameters of scintigraphy with ^sup 99m^Tc-MAA in the diagnosis of intrapulmonary vascular dilatations

• Published in: Annals of nuclear medicine 2015-01, Vol.29 (1), p.46
• Main Authors: de Queirós, Andréa Simone, Siqueira, Brandão, Simone Cristina, Soares, Macedo, Liana Gonçalves, Ourem, Maira Souto, Mota, Vitor Gomes, Leite, Luiz Arthur, Calheiros, Lopes, Edmundo Pessoa, Almeida, Domingues, Ana Lúcia, Coutinho
• Format: Article
• Language: English
• ISSN: 0914-7187; 1864-6433
• DOI: 10.1007/s12149-014-0915-9

The formation of intrapulmonary vascular dilations (IPVD) is the key event for the onset of hepatopulmonary syndrome, vascular changes secondary to portal hypertension that leads to hypoxemia. The diagnosis of IPVD can be made by contrasted transthoracic echocardiography or scintigraphy with technetium-macroaggregated albumin--(^sup 99m^Tc-MAA)--that is a sensitive and specific diagnostic method and quantifies the IPVD magnitude. However, its procedure and diagnostic indices are not yet standardized and well defined in health services. The aims of this study were to define normality values and evaluate the inter- and intra-observer reproducibility degree of diagnostic indexes of IPVD through ^sup 99m^Tc-MAA scintigraphy. Cross-sectional study was conducted at the Clinical Hospital, Federal University of Pernambuco (HC-UFPE) between July and December 2012. Fifteen patients with hepatosplenic schistosomiasis and nine patients without liver or heart disease (control group) were assessed. After clinical assessment, ultrasound and echocardiography, patients underwent ^sup 99m^Tc-MAA scintigraphy, and a relative brain uptake value exceeding 6 % or systemic uptake value exceeding 11 % was considered diagnostic of IPVD. Each assessment was performed by two independent observers. To analyze the results of the normal group, the nonparametric Bootsptrap method simulation model combined with the Monte Carlo method was used and to analyze inter- and intra-observer reproducibility indexes, the kappa and intra-class correlation coefficient were used. In normal subjects, the average brain uptake of ^sup 99m^Tc-MAA was 7.9 ± 0.01 % and systemic uptake was 12.4 ± 0.03 %, with low dispersal rates for both measures. The intra-observer agreement was 100 %, with kappa index of 1.0 (p < 0.0001), suggesting a perfect agreement. The inter-observer agreement was also 100 % (kappa = 1.0, p < 0.0001) for brain uptake; however, systemic uptake showed kappa = 0.25 (p = 0.07), which features tolerable concordance. The intra-class correlation was excellent for both uptake indexes. The normality values were slightly higher than those reported in studies from other countries. The demographic characteristics of the Brazilian population, the small number of patients or different methodologies can be the causes of such differences. ^sup 99m^Tc-MAA scintigraphy showed excellent reproducibility.[PUBLICATION ABSTRACT]