Diss. (sammanfattning) Stockholm : Karol. inst. / Härtill 5 uppsatser.

No description available.

This research develops a detailed, anatomically-based model of the human pulmonary circulatory system from the large scale arterial and venous vessels, to the microcirculatory alveolar-capillary unit. Flow is modelled through these networks enabling structure-function simulations to be conducted to increase our understanding of this complex system.Voronoi meshing is applied in a novel technique to represent the three-dimensional structure of the alveoli, and the corresponding capillary plexus intimately wrapped over the alveolar surface. This technique is used to create the alveolar-capillary structure of a single alveolar sac, closely representing the geometry measured in anatomical studies.A Poiseuille type flow solution technique is implemented within the capillary geometry. The solution procedure incorporates calculations of red and white blood cell transit time frequencies. Novel predictions of regional microcirculatory blood cell transit in the anatomically-realistic alveolar-capillary model compare well with experimental measures.An anatomically-based finite element model of the arterial and venous vessels, down to the level of their accompanying respiratory bronchioles, is created using a combination of imaging and computational algorithms, which includes generation of supernumerary vessels. Large arterial and venous vessels and lobar geometries are derived from multi-detector row x-ray computed tomography (MDCT) scans. From these MDCT vessel end points a volume-filling branching algorithm is used to generate the remaining blood vessels that accompany the airways into the MDCT-derived host volume. An empirically-based algorithm generates supernumerary blood vessels - unaccompanied by airways that branch to supply the closest parenchymal tissue. This new approach produces a model of pulmonary vascular geometry that is far more anatomically-realistic than previous models in the literature.A reduced form of the Navier-Stokes equations are solved within the vascular geometries to yield pressure, radius, and velocity distributions. Inclusion of a gravitational term in the governing equations allows application of the model in investigating the relative effects of gravity, structure, and posture on regional perfusion.Gravity is shown to have a lesser influence on blood flow distribution than suggested by earlier experimental studies, and by comparison between different model solutions the magnitude of the gravitational flow gradient is predicted. This study clearly demonstrates the significant role that symmetric vascular branching has in determining the distribution of blood flow. The influence of branching geometry is revealed by solution in symmetric, human, and ovine vascular models.

This research develops a detailed, anatomically-based model of the human pulmonary circulatory system from the large scale arterial and venous vessels, to the microcirculatory alveolar-capillary unit. Flow is modelled through these networks enabling structure-function simulations to be conducted to increase our understanding of this complex system.Voronoi meshing is applied in a novel technique to represent the three-dimensional structure of the alveoli, and the corresponding capillary plexus intimately wrapped over the alveolar surface. This technique is used to create the alveolar-capillary structure of a single alveolar sac, closely representing the geometry measured in anatomical studies.A Poiseuille type flow solution technique is implemented within the capillary geometry. The solution procedure incorporates calculations of red and white blood cell transit time frequencies. Novel predictions of regional microcirculatory blood cell transit in the anatomically-realistic alveolar-capillary model compare well with experimental measures.An anatomically-based finite element model of the arterial and venous vessels, down to the level of their accompanying respiratory bronchioles, is created using a combination of imaging and computational algorithms, which includes generation of supernumerary vessels. Large arterial and venous vessels and lobar geometries are derived from multi-detector row x-ray computed tomography (MDCT) scans. From these MDCT vessel end points a volume-filling branching algorithm is used to generate the remaining blood vessels that accompany the airways into the MDCT-derived host volume. An empirically-based algorithm generates supernumerary blood vessels - unaccompanied by airways that branch to supply the closest parenchymal tissue. This new approach produces a model of pulmonary vascular geometry that is far more anatomically-realistic than previous models in the literature.A reduced form of the Navier-Stokes equations are solved within the vascular geometries to yield pressure, radius, and velocity distributions. Inclusion of a gravitational term in the governing equations allows application of the model in investigating the relative effects of gravity, structure, and posture on regional perfusion.Gravity is shown to have a lesser influence on blood flow distribution than suggested by earlier experimental studies, and by comparison between different model solutions the magnitude of the gravitational flow gradient is predicted. This study clearly demonstrates the significant role that symmetric vascular branching has in determining the distribution of blood flow. The influence of branching geometry is revealed by solution in symmetric, human, and ovine vascular models.

This research develops a detailed, anatomically-based model of the human pulmonary circulatory system from the large scale arterial and venous vessels, to the microcirculatory alveolar-capillary unit. Flow is modelled through these networks enabling structure-function simulations to be conducted to increase our understanding of this complex system.Voronoi meshing is applied in a novel technique to represent the three-dimensional structure of the alveoli, and the corresponding capillary plexus intimately wrapped over the alveolar surface. This technique is used to create the alveolar-capillary structure of a single alveolar sac, closely representing the geometry measured in anatomical studies.A Poiseuille type flow solution technique is implemented within the capillary geometry. The solution procedure incorporates calculations of red and white blood cell transit time frequencies. Novel predictions of regional microcirculatory blood cell transit in the anatomically-realistic alveolar-capillary model compare well with experimental measures.An anatomically-based finite element model of the arterial and venous vessels, down to the level of their accompanying respiratory bronchioles, is created using a combination of imaging and computational algorithms, which includes generation of supernumerary vessels. Large arterial and venous vessels and lobar geometries are derived from multi-detector row x-ray computed tomography (MDCT) scans. From these MDCT vessel end points a volume-filling branching algorithm is used to generate the remaining blood vessels that accompany the airways into the MDCT-derived host volume. An empirically-based algorithm generates supernumerary blood vessels - unaccompanied by airways that branch to supply the closest parenchymal tissue. This new approach produces a model of pulmonary vascular geometry that is far more anatomically-realistic than previous models in the literature.A reduced form of the Navier-Stokes equations are solved within the vascular geometries to yield pressure, radius, and velocity distributions. Inclusion of a gravitational term in the governing equations allows application of the model in investigating the relative effects of gravity, structure, and posture on regional perfusion.Gravity is shown to have a lesser influence on blood flow distribution than suggested by earlier experimental studies, and by comparison between different model solutions the magnitude of the gravitational flow gradient is predicted. This study clearly demonstrates the significant role that symmetric vascular branching has in determining the distribution of blood flow. The influence of branching geometry is revealed by solution in symmetric, human, and ovine vascular models.

This research develops a detailed, anatomically-based model of the human pulmonary circulatory system from the large scale arterial and venous vessels, to the microcirculatory alveolar-capillary unit. Flow is modelled through these networks enabling structure-function simulations to be conducted to increase our understanding of this complex system.Voronoi meshing is applied in a novel technique to represent the three-dimensional structure of the alveoli, and the corresponding capillary plexus intimately wrapped over the alveolar surface. This technique is used to create the alveolar-capillary structure of a single alveolar sac, closely representing the geometry measured in anatomical studies.A Poiseuille type flow solution technique is implemented within the capillary geometry. The solution procedure incorporates calculations of red and white blood cell transit time frequencies. Novel predictions of regional microcirculatory blood cell transit in the anatomically-realistic alveolar-capillary model compare well with experimental measures.An anatomically-based finite element model of the arterial and venous vessels, down to the level of their accompanying respiratory bronchioles, is created using a combination of imaging and computational algorithms, which includes generation of supernumerary vessels. Large arterial and venous vessels and lobar geometries are derived from multi-detector row x-ray computed tomography (MDCT) scans. From these MDCT vessel end points a volume-filling branching algorithm is used to generate the remaining blood vessels that accompany the airways into the MDCT-derived host volume. An empirically-based algorithm generates supernumerary blood vessels - unaccompanied by airways that branch to supply the closest parenchymal tissue. This new approach produces a model of pulmonary vascular geometry that is far more anatomically-realistic than previous models in the literature.A reduced form of the Navier-Stokes equations are solved within the vascular geometries to yield pressure, radius, and velocity distributions. Inclusion of a gravitational term in the governing equations allows application of the model in investigating the relative effects of gravity, structure, and posture on regional perfusion.Gravity is shown to have a lesser influence on blood flow distribution than suggested by earlier experimental studies, and by comparison between different model solutions the magnitude of the gravitational flow gradient is predicted. This study clearly demonstrates the significant role that symmetric vascular branching has in determining the distribution of blood flow. The influence of branching geometry is revealed by solution in symmetric, human, and ovine vascular models.

This research develops a detailed, anatomically-based model of the human pulmonary circulatory system from the large scale arterial and venous vessels, to the microcirculatory alveolar-capillary unit. Flow is modelled through these networks enabling structure-function simulations to be conducted to increase our understanding of this complex system.Voronoi meshing is applied in a novel technique to represent the three-dimensional structure of the alveoli, and the corresponding capillary plexus intimately wrapped over the alveolar surface. This technique is used to create the alveolar-capillary structure of a single alveolar sac, closely representing the geometry measured in anatomical studies.A Poiseuille type flow solution technique is implemented within the capillary geometry. The solution procedure incorporates calculations of red and white blood cell transit time frequencies. Novel predictions of regional microcirculatory blood cell transit in the anatomically-realistic alveolar-capillary model compare well with experimental measures.An anatomically-based finite element model of the arterial and venous vessels, down to the level of their accompanying respiratory bronchioles, is created using a combination of imaging and computational algorithms, which includes generation of supernumerary vessels. Large arterial and venous vessels and lobar geometries are derived from multi-detector row x-ray computed tomography (MDCT) scans. From these MDCT vessel end points a volume-filling branching algorithm is used to generate the remaining blood vessels that accompany the airways into the MDCT-derived host volume. An empirically-based algorithm generates supernumerary blood vessels - unaccompanied by airways that branch to supply the closest parenchymal tissue. This new approach produces a model of pulmonary vascular geometry that is far more anatomically-realistic than previous models in the literature.A reduced form of the Navier-Stokes equations are solved within the vascular geometries to yield pressure, radius, and velocity distributions. Inclusion of a gravitational term in the governing equations allows application of the model in investigating the relative effects of gravity, structure, and posture on regional perfusion.Gravity is shown to have a lesser influence on blood flow distribution than suggested by earlier experimental studies, and by comparison between different model solutions the magnitude of the gravitational flow gradient is predicted. This study clearly demonstrates the significant role that symmetric vascular branching has in determining the distribution of blood flow. The influence of branching geometry is revealed by solution in symmetric, human, and ovine vascular models.

Made available in DSpace on 2015-07-22T20:49:44Z (GMT). No. of bitstreams: 0 Previous issue date: 2010-06-30. Added 1 bitstream(s) on 2015-08-11T03:26:07Z : No. of bitstreams: 1 Publico-12636.pdf: 1313292 bytes, checksum: 9dc5695df095889d7733e20c113824b4 (MD5) / Objetivo: Comparar os valores absolutos da CVF e VEF1 de uma amostra randomizada da populacao de Sao Paulo com mais de 40 anos de idade com os valores previstos das equacoes mais utilizadas no Brasil. Comparar os valores em percentuais do previstos da CVF e VEF1 das equacoes selecionadas com os da equacao PLATINO. Avaliar a mudanca de estadio da DPOC ao se utilizar as diferentes equacoes de referencia. Material e Metodos: Estudo retrospectivo, com 178 pessoas normais da amostra de moradores da regiao metropolitana de Sao Paulo, com idade .40 anos, proveniente do estudo PLATINO. Para a comparacao da CVF e do VEF1, foram subtraidos dos valores previstos a partir de cada equacao os valores absolutos obtidos de cada individuo, considerando uma diferenca de ate 0,15L, como definida pela ATS. Tambem foi realizada a diferenca entre os valores do percentual do previsto da CVF e VEF1 das equacoes com os da equacao PLATINO, com diferenca . 3% (ATS). O diagnostico de DPOC foi definido pela relacao VEF1/CVF < 0,70 pos-broncodilatador e a comparacao do estadiamento foi realizada utilizando a variavel VEF1 em porcentagem do previsto de cada equacao, tendo como referencia o percentual previsto da equacao PLATINO. As equacoes foram aplicadas de acordo com os seus limites de idade e altura. Teste t-student, Qui-quadrado e coeficiente de correlacao intraclasse foram utilizados para avaliar o comportamento da CVF e VEF1 nas diferentes equacoes e os dados foram apresentados em media, desvio padrao, erro padrao e intervalo de confianca de 95%. Resultados: Observamos que as equacoes que apresentam menor diferenca entre o valor previsto e o valor absoluto para CVF e VEF1, em ambos os sexos, foram Roca et al. (n=151), Pereira et al. 2007 (n=176), Enright et al (n=43) e PLATINO (n=178). A equacao Knudson et al (n=177). subestimou os valores absolutos para CVF e VEF1. Na comparacao com a equacao PLATINO, em percentual do previsto para CVF e VEF1, a equacao que apresentou menor diferenca foi a NHANES III (n=172), exceto para CVF masculina, que apresentou grande variabilidade. Foram avaliados 152 pacientes com DPOC, estadiados pelo VEF1, segundo a equacao PLATINO, observando-se mudanca no estadiamento em 29,6% dos pacientes em relacao as equacoes, tendo a equacao Knudson et al. apresentado o maior numero de alteracoes. Conclusao: Existe uma grande variabilidade nos valores previstos entre as equacoes de referencia. As equacoes que apresentaram menor variabilidade e melhor coeficiente de correlacao intraclasse foram Roca et al., NHANES III caucasiano et al., Enright et al. e PLATINO, recomendando-se o seu uso para a populacao brasileira. Existe mudanca no estadio da DPOC quando diferentes equacoes sao aplicadas, sugerindo-se o uso de uma unica equacao durante o tratamento do mesmo paciente para evitar eventuais alteracoes no tratamento. / Objective: To compare the absolute values of FVC and FEV1 found in a random sample of a Brazilian population over 40 years of age with values of equations most used in Brazil. To compare the values in percentage of predicted of FVC and FEV1 of the selected equations with the PLATINO equation ones. To evaluate the change in COPD staying when using different reference equations. Material and Methods: We conducted a retrospective study from a representative sample of residents in the metropolitan region of Sao Paulo, aged .40 years, from the PLATINO study. For comparison of the FVC and FEV1, were subtracted from the predicted values of each equation the absolute values obtained from each individual, considering the equation inadequate wherever there was a difference of up to 0.15 L as defined by ATS. We also evaluated the difference between the values of FVC and VEF1 in percentage of predicted of the equations with the PLATINO equation, considering as a limit a difference .3% (ATS). The diagnosis of COPD was defined as FEV1/FVC <0.70 after bronchodilator, and the comparison of staging was performed using the FEB1 in percentage of predicted in each equation, with reference to the percentage expected from the PLATINO equation. T-student test, chi-squared and the intraclass correlation coefficient were used to evaluate the performance of FVC and FEV1 in different equations and data are presented as mean, standard deviation, standard error and confidence interval of 95%. Results: After evaluating 178 healthy subjects, we observed that the equations that show less difference between the predicted value and the absolute value for both FEV1 and FVC in both sexes were Roca et al., Pereira et al. 2006, Enright et al. and PLATINO. The Knudson et al. equation underestimated the absolute values for both FVC and FEV1. In comparison with PLATINO equation, the percentage of predicted for FVC and FEV1, the equation showing the lowest difference was NHANES III, except for FVC male, who showed a wide variability. We evaluated 152 patients with COPD, observing staging changes in 29.6% in relation to the difference equations, being Knudson et al. the equation that showed the greatest number of changes. Conclusion: There is a great variability among the reference equations. The equations showing less variability and better intraclass correlation coefficient were Roca et al., Caucasian NHANES III et al., Enright et al. and PLATINO, which are the ones that should be recommended to be used for the Brazilian population. There is a change in the COPD staging when different equations are applied, suggesting that the use of a single equation for the treatment of the patient would avoid possible alterations in treatment. / TEDE / BV UNIFESP: Teses e dissertações

Orientador: Lourenço Sbragia Neto / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Ciencias Medicas / Made available in DSpace on 2018-08-12T11:31:54Z (GMT). No. of bitstreams: 1 Nassr_AzizeCristinaCapelli_D.pdf: 6642584 bytes, checksum: 60fb43046c34664559a19ebb364b282b (MD5) Previous issue date: 2008 / Resumo: A Hérnia Diafragmática Congênita (HDC) é um defeito da formação do músculo diafragma que incide em aproximadamente 1:2500 nascidos vivos e apresenta altos índices de mortalidade fetal e neonatal decorrentes da hipoplasia e da hipertensão pulmonares. Este defeito pode ser induzido experimentalmente em ratas grávidas administrando o herbicida nitrofen que causa HDC em 24% dos fetos. A análise microscópica do pulmão da HDC demonstra a presença de hipolasia pulmonar além de alveolização e vascularização alterada. Um dos fatores de crescimento envolvidos no desenvolvimento vascular é o VEGF (vascular endothelial growth factor) e seus receptores, no entanto ainda não se conhece como a expressão desta glicoproteína e de seus receptores varia ao longo do desenvolvimento pulmonar fetal nesta doença. Utilizando o modelo experimental de HDC induzido pelo nitrofen (2,4-dicloro-4'nitrodifenil éter) investigamos o grau de hipoplasia pulmonar e por meio de análise imunoistoquímica, comparamos a expressão dos receptores para o VEGF em três fases do desenvolvimento pulmonar, pseudoglandular, canalicular e sacular de fetos de ratos normais e com HDC. Dividimos o experimento em ratas da raça Sprague-Dawley em três grupos: controle externo (CE), exposto ao óleo de oliva (OO) e expostas ao nitrofen com e sem HDC. Estudamos quatro grupos de 20 fetos cada em cinco dias gestacionais (DG) diferentes 17,5, 18,5, 19,5, 20,5 e 21,5. As variáveis morfológicas estudadas foram: peso corporal (PC), peso pulmonar total (PPT), peso do pulmão esquerdo (PPE), relação PPT/PC, volume pulmonar total (VPT) e volume do pulmão esquerdo (VPE). As variáveis histométricas estudadas foram: parênquima pulmonar (Par), espaço aéreo (EA), densidade do parênquima (DAP) e volume do parênquima do pulmão esquerdo. A avaliação imumohistoquímica foi realizada por meio da contagem de pontos de receptor de VEGFR-1 e 2. Obtivemos 37 % (100/270) de HDC nas ratas expostas ao nitrofen, todas variáveis morfológicas e histométricas indicam diminuição dos resultados no grupo nitrofen com e sem HDC em relação aos demais, mas que se acentuam mais ainda no grupo HDC. Essas alterações são mais evidentes a partir dos DG 18,5 e 19,5. A imunomarcação para os receptores VEGFR-1 aumentou nos grupos nitrofen e foram progressivamente maiores no grupo nitrofen com HDC (p<0,005) que os fetos dos grupo CE e OO a partir do dia gestacional 17,5, fase pseudoglandular com pico máximo no dia gestacional 19,5. O mesmo ocorreu com os receptores de VEGFR-2 a partir do dia gestacional 17,5, fase pseudoglandular até o dia 21,5 fase sacular do desenvolvimento pulmonar. Concluímos que o modelo é valido e que os fetos expostos ao nitrofen com e sem HDC apresentam hipoplasia pulmonar primária sendo mais acentuada nos fetos portadores de HDC. O mesmo resultado ocorre com imunomarcação para os receptores de VEGFR-1 e 2 que foram maiores na HDC. / Abstract: The Congenital Diaphragmatic Hernia (CDH) is a defect in the embryogenesis of the diaphragm with an incidence of 1:2500 liveborns and high fetal and neonatal mortality due to pulmonary hypoplasia and hypertension. This defect can be experimentally induced in fetuses of pregnant rats by the administration of Nitrofen, an herbicide that causes CDH in 24% of the fetuses. The histology of lungs in CDH shows pulmonary hipoplasia and not only the alveolarization but also the vascularization are affected. These changes lead to a high neonatal mortality because of the thickening of the middle layer of the arterioles causing pulmonary hypertension. One of the factors involved in the growth of the arterioles is VEGF (vascular endothelial growth factor) and its receptors; however, it is not known how the expression of this glycoprotein and its receptors change during lung development in this disease. In Brazil, the experimental model has never been tested. So, we tested the model and verified the degree of pulmonary hipoplasia and, using imunohystochemistry, we compared the expression of the receptor of VEGF in three different stages of lung development, pseudoglandular, canalicular and saccular, of normal rat fetuses and fetuses with CDH. Female Sprague-Dawley rats were divided in three groups: external control (EC), exposed to olive oil (OO) and exposed to nitrofen (N). We studied four groups - EC, OO, N with CDH and N without CDH - with 20 fetuses in each five different gestational days (GD) 17,5, 18,5, 19,5, 20,5, 21,5. The morphologic variables studied were: body weight (BW), total lung weight (TLW), left lung weight (LLW), relationship TLW/BW, total lung volume (TLV) and left lung volume (LLV). The hystometric variables studied were: lung parenchyma (LP), air space (AS), left lung parenchyma density (PD) and left lung parenchyma volume (PV). The immunohystochemistry variables were: points positive and negative for the receptor for VEGF 1 and 2. We had 37% (100/270) of CDH frequency in the fetuses exposed to nitrofen. All the morphological and hystometrical variables show a reduction in the nitrofen group with and without CDH, which were more pronounced in the group of fetuses with CDH. These changes are more evident from the GD 18,5 and 19,5 on. The receptors VEGFR-1 e 2 are increased in the nitrofen groups with and without CDH, but this increase is higher in the fetuses with CDH. We conclude that the model is valid and that the fetuses exposed to nitrofen with and without CDH show primary pulmonary hypoplasia that is more pronounced in CDH, the same is also observed in the receptors of VEGFR-1 and 2. / Doutorado / Pesquisa Experimental / Doutor em Cirurgia

Introdução: A hemodiluição normovolêmica aguda (HNA) é estratégia utilizada para diminuir a necessidade de transfusão em procedimentos cirúrgicos. O hematócrito alvo, a solução empregada e os riscos da HNA são controversos e, embora as ações sobre os sistemas nervoso e cardiocirculatório tenham sido extensamente pesquisados, há poucos estudos sobre implicações pulmonares da HNA. O objetivo deste estudo é avaliar os efeitos pulmonares da HNA comparando amido hidroxietílico e Ringer lactato. Métodos: Vinte e cinco porcos foram monitorizados e cateterizados. Utlizou-se anestesia intravenosa contínua com propofol, cetamina, fentanil e pancurônio. Após estabilização da anestesia, os animais foram randomizados: Grupo Controle (n=8), Grupo HES (amido hidroxietílico, n=7) e Grupo RL (Ringer lactato, n=10). Os animais dos grupos II e III foram submetidos a HNA até hematócrito pré-estabelecido de 15%. A retirada do sangue foi acompanhada da expansão plasmática com amido hidroxietílico ou Ringer lactato na proporção de 1:1 ou 3:1 respectivamente. Medidas de mecânica respiratória e oxigenação sangüínea foram realizadas antes da retirada de sangue (T0), ao término da hemodiluição (T1) e após uma e duas horas (T2 e T3). Ao final foram realizadas biópsias pulmonares. Os dados foram submetidos à análise de variância para medidas repetidas (ANOVA) seguidas pelo teste de Tukey. Resultados: Em relação aos parâmetros do grupo controle, não houve diferença significante durante todo o procedimento. O grupo Ringer lactato demonstrou diminuição da complacência estática e aumento da diferença alvéolo-arterial de O2 (DAa-O2) em relação ao controle (*P<0,05). A análise histopatológica evidenciou atelectasias e alargamento da membrana basal na barreira alvéolo-capilar no grupo Ringer lactato. O grupo amido hidroxietílico apresentou comportamento semelhante ao grupo controle. Conclusões: O estudo sugere que na HNA acentuada, o amido hidroxietílico é superior na preservação estrutural do pulmão, evidenciado pormanter a complacência pulmonar e a oxigenação, quando comparada à realizada com solução de Ringer lactato / Introduction: Acute normovolemic hemodilution (ANH) is a common procedure to decrease the needs of blood transfusion during a variety of surgical procedures. There are controversies regarding ANH such as the target hematocrit, the solution to be used and its risks. Although its effects in heart and brain have been extensively studied; investigations concerning its pulmonary effects are lacking. The aim of this experiment was to compare the pulmonary effects of ANH with hydroxyethyl starch and lactated Ringer\'s. Methods: Twenty five pigs were instrumented for standard hemodynamic monitoring. Intravenous anesthesia consisted continuous infusion of propofol, ketamine, fentanyl and pancuronium. After anesthesia stabilization animals were randomized: Control group (n=8), HES group (hydroxyethyl starch, n=7) and LR group (lactated Ringer\'s, n=10). Animals of Control Group were anesthetized and observed. Group Animals of group HES and LR were submitted to ANH to reach a pre-established hematocrit around 15%. Blood withdrawal was accomplished simultaneously with plasma expansion performed with hydroxyethyl starch (1:1) or with lactated Ringer\'s (3:1). Pulmonary mechanics and oxygen transport and were measured before blood withdrawal (T0), at the end of hemodilution (T1), and one and two hours after the end of the hemodilution (T2, T3). Lungs biopsies were performed at the end. Data were submitted to analysis of Variance for repeated measures followed by the Tukey test. Results: In regard to all parameters of Control group, there were no significant differences during the whole procedure. The LR group demonstrated a decrease in compliance and an increased in (a-A)DO2 (*P<0,05). Hystopathological lung analysis revealed moderate to serious collapses as well as basement membrane enlargement in group LR. The HES data were similar to that found in control group. Conclusion: These results suggest that in severe ANH, HES preserves pulmonary structure and seems to less interfere with pulmonary mechanics and oxygenation indexes when compared to lactated Ringer\'s