SciELO - Scientific Electronic Library Online

vol.52 issue2Preanesthetic clinical problems of morbidly obese patients submitted to bariatric surgery: comparison with non-obese patiensMarked hypercapnia during cardiopulmonary bypass for myocardial revascularization: case report author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand



Related links


Revista Brasileira de Anestesiologia

Print version ISSN 0034-7094

Rev. Bras. Anestesiol. vol.52 no.2 Campinas Mar./Apr. 2002 



Histopathologic study on the effects of tenoxicam with bidistilled water or with 0.9% sodium chloride in rabbits venous endothelium*


Estudio histopatológico del efecto del tenoxican con agua bidestilada o con cloreto de sodio a 0,9% en el endotélio venoso de conejos



Taylor Brandão Schnaider, TSA, M.D.I; Alcino Lázaro da Silva, M.D.II; Míriam de Fátima Brasil Engelman, M.D.III; Yara Juliano, M.D.IV; Neil Ferreira Novo, M.D.V; Gabrielle Sormanti Schnaider, M.D.VI; Caroline Sormanti Schnaider, M.D.VII

IProfessor Titular do Departamento de Clínica Cirúrgica da Faculdade de Medicina - Universidade do Vale do Sapucaí (UNIVAS)
IIProfessor Titular da Disciplina de Cirurgia do Aparelho Digestivo da Faculdade de Medicina - Universidade Federal de Minas Gerais (UFMG)
IIIProfessora Adjunta do Departamento de Ciências Fisio-morfológicas e Patologia da Faculdade de Medicina (UNIVAS)
IVProfessora Adjunta da Disciplina de Bioestatística da Escola Paulista de Medicina - Universidade Federal do Estado de São Paulo (UNIFESP)
VProfessor Adjunto da Disciplina de Bioestatística da Escola Paulista de Medicina (UNIFESP)
VIME de Pediatria (UNICAMP)
VIIDoutoranda da Faculdade de Medicina (UNIVAS)





BACKGROUND AND OBJECTIVES: After exposure to indomethacin, human umbilical vein endothelial cells have shown increased procoagulant activity. Thrombosis in rabbit auricular veins has been observed after injection of tenoxicam with its diluent or of the diluent alone 2. Human studies evaluating venous endothelium after tenoxican injection were not found in the literature. This study aimed at evaluating whether tenoxicam with 0.9% sodium chloride changes the venous endothelium of rabbits as it is observed when associated to its diluent (bidistilled water).
METHODS: The study involved 90 rabbits (2000 and 3500 grams) divided in two groups: Control group, which received 0.9% NaCl; Experimental group, which received tenoxicam (20 mg) with bidistilled water or with 0.9% NaCl. A constant volume of 2 ml was administered to both groups. Anesthesia was induced with acepromazine maleate, ketamine hydrochloride and xylazine hydrochloride, and the puncture of right and left caudal auricular veins was performed with a 27G butterfly needle. Animals were confined for 6 hrs, 12 hrs, and 24 hrs, when they were once more anesthetized and sacrificed, with extraction of the auriculae at their base, followed by microscopic venous study.
RESULTS: Thrombosis was observed in 19.4% of the Experimental Group after the administration of tenoxicam with bidistilled water and in 22.2% after the administration of tenoxicam with 0.9% sodium chloride. In the Control group, which has only received 0.9% sodium chloride, no thrombosis was observed.
CONCLUSIONS: It was possible to conclude that tenoxicam, either with bidistilled water or 0.9% sodium chloride, has induced thrombosis in the veins it was injected.

Key words: ANALGESICS, Anti-inflammatory: tenoxicam; ANIMAL: rabbit


JUSTIFICATIVA Y OBJETIVOS: En estudio con células endoteliales de venas umbilicales humanas expuestas a la indometacina, fue observado aumento de la actividad pró-coagulante. Estudio en conejos comprobó la presencia de trombosis en las venas auriculares después de la administración de tenoxicam con su diluyente o de su diluyente aislado. No fueron encontrados estudios en la literatura consultada que hayan evaluado el endotélio venoso después de la administración de tenoxicam, en seres humanos. El objetivo de esta pesquisa fue evaluar se el tenoxicam con cloreto de sodio a 0,9% (NaCl a 0,9%) provoca alteraciones en el endotélio venoso de conejos, como las observadas cuando asociado a su diluyente (agua bidestilada).
MÉTODO: Noventa conejos (2.000 - 3.500 g) fueron distribuidos aleatoriamente en dos grupos: Control, con administración de NaCl a 0,9%; Experimento, con tenoxicam (20 mg) asociado a agua bidestilada o al NaCl a 0,9%. El volumen inyectado en los dos grupos fue constante de 2 ml. La anestesia fue inducida con maleato de acepromazina, clorhidrato de cetamina y clorhidrato de xilazina, siendo la punción de las venas auriculares caudales derecha e izquierda realizada con aguja tipo mariposa 27G. Los animales fueron mantenidos en el biotério por 6 h, 12 h e 24 h, nuevamente anestesiados y sometidos a eutanasia, siendo entonces realizada exéresis de las aurículas en su base y posterior evaluación microscópica de las venas.
RESULTADOS: Se observó trombosis en el grupo Experimento, en una porcentaje de 19,4% después de administración del tenoxicam con agua bidestilada y 22,2% después administración del tenoxicam con NaCl a 0,9%. En el grupo Control, en que fue inyectado, solamente NaCl a 0,9%, ninguna de las venas presento trombosis.
CONCLUSIONES: Los resultados encontrados permiten concluir que el tenoxicam, con agua bidestilada o con solución de cloreto de sodio a 0,9%, produjo trombosis en las venas en que fue inyectado.




After observing in our anesthesiology clinic some patients referring signs and symptoms of vasculytis (pain, heat, redness) after direct intravenous tenoxicam with bidistilled water (without commercial diluent), we have searched the literature but were unable to find studies on this event in humans.

Nygard et al., as a consequence of the association of non-steroid anti-inflammatory drugs and intestinal microvascular injuries, have studied coagulation abnormalities in human umbilical vein endothelial cells exposed to indomethacin, bacterial lipopolysaccharides or their association, after 1 h, 6 h, 12 h and 24 h. The authors have found the following results: lipolysaccharides have significantly induced procoagulant activity at 6 h, 12 h and 24 h, with a significant increase in 6-keto-PGF1a, while the increase in PGE2 production was less pronounced; with indomethacin alone, the procoagulant activity was time-dependent; with Indomethacin and lipopolysaccharides, procoagulant response was more significant than that produced by cultures with just indomethacin or lipopolysaccharides 1.

Schnaider et al., in experimental studies with rabbits, have hystopathologically evaluated the presence or absence of thrombosis in the venous endothelium after the administration of tenoxicam with its commercial diluent or its diluent alone. Authors have found 19.4% thrombosis in both cases 2, but this study has raised a question: was thrombosis caused by tenoxicam or by bidistilled water used as a diluent.

This study aimed at hystopathologically evaluating whether tenoxicam with 0.9% NaCI would change rabbits venous endothelium similarly to what is seen when this anti-inflammatory is associated to bidistilled water.



After the Animal Research Ethical Committee approval, 90 white, New Zealand, male rabbits (Oryctolagus cuniculus), aged above 10 weeks, weighing 2000 to 3500 grams, were randomly distributed in two groups: Experimental and Control. The Experimental group was subdivided in 3 subgroups of 24 rabbits each, which were called E6, E12 and E24, corresponding respectively to 6 h, 12 h and 24 h of observation and randomly receiving tenoxicam/diluent or tenoxicam/0.9% NaCI in the right or left caudal auricular vein. The Control group was also subdivided in 3 subgroups of 6 rabbits each, which were called subgroups C6, C12 and C24, corresponding respectively to 6 h, 12 h and 24 h of observation and receiving 0.9% NaCI in the right or left caudal auricular vein.

There were no food and water restrictions until beginning of the experiment, when animals were weighed and anesthetized with muscular 1% acepromazine maleate (2, ketamine hyodrochloride (35 and 2% xilazine hydrochloride (5 in the gluteal region.

Ten minutes after anesthesia, animals were placed in the supine position and chest and pelvic limbs were tied with strings to the table’s edges.

After 15 minutes, anesthetic depth was tested by response of right pelvic limb to a painful stimulus in the skin fold between digits II and III.

Next, antisepsis was performed with 70% alcohol and animals were covered with sterilized drapes. Right and left caudal auricular veins were punctured with a 27G butterfly needle. The following drugs were randomly injected: tenoxicam/diluent or tenoxicam/0.9% NaCI in the Experimental group and 0.9% NaCI in the Control group (Figure 1). Puncture sites were then marked with a “Texta fine line 7000” pen. Substances (2 ml) were previously packaged in 3 ml syringes and were injected at the speed of 1 ml for 30 seconds.

After the evolution time scheduled for each subgroup (6 h, 12 h or 24 h), animals were again anesthetized by the same technique used in the experiment, and sacrificed with 3 ml intracardiac 10% potassium chloride. Next, a triangular cut was performed at the auricles rostral portion and a stitch, with 000 cotton thread was placed at the distal portion of the right auricle. Auricles were then resected at their base and placed in 10% neuter buffered formalin after macroscopic observation (Figure 2). Containers with surgical specimens were identified by group, subgroup and animal’s number. Specimens were place in wax and 5 sections with 5 micrometers of width each were performed at 1 cm from the puncture site, proximal to the auricle, using a rotating mocrotome with disposable blade (Leica RM-2025). Sections were then placed in thin plates, died with hematoxylin-eosin (H-E) and covered by glass slides. Plates were analyzed by optical microscopy (Lambda LQT-2) with original enhancements of 40, 100 and 400 times, digitized (Samsung SHC-410 NAD Camera) and written in 3 ½ diskettes.

Chi-square partition test for 2xN tables was used to compare, for each group, 6 h, 12 h and 24 h of euthanasia with regards to thrombosis. The same test was applied to compare total thrombosis incidence in tenoxicam/diluent and tenoxicam/0.9% NaCI Control groups, regardless of the euthanasia time. Significance level was 0.05 or 5% (a £ 0.05) for all tests and a bold star (*) was used to highlight significant values.



To analyze the incidence of thrombosis, statistical analysis was performed in two stages. In the first stage, Chi-square test for 2xN tables was used to compare euthanasia periods of 6 h, 12 h or 24 h, for each group (Tenoxicam/0.9% NaCI, Tenoxicam/Diluent and 0.9% NaCI) without significant differences among periods.

In the second stage, the same test was used to compare total thrombosis of Tenoxicam/0.9% NaCI and Tenoxicam/Diluent and 0.9% NaCI groups, regardless of the euthanasia period, observing that Tenoxicam/0.9% NaCI and Tenoxicam/Diluent groups thrombosis incidence was significantly higher than in the 0.9% NaCI group.

Results are shown in table I, expressed in figure 3 and figure 4 and illustrated in figure 5.



Regardless of the triggering aggression, endothelial injuries suffer a major thormbogenic influence. When there is an endothelial cell injury, blood gets in touch with the adjacent collagen tissue which induces platelet aggregation. Next, fribin forms a tridimensional fibrous net which traps erythrocytes, leucocytes and platelets originating thrombi, which may grow and even block blood flow 3.

Platelets activation factor, most potent ulcerogenic agent, may be another cause for increased intestinal patency, by mucosal vascularization exchanges induction 4 or by acting on neutrophils 5. Its local production may be directly stimulated by NSAIDs or by the development of an inflammatory response 6.

Nygard et al. have observed in human umbilical vein endothelial cells exposed to Indomethacin, that the synthesis of anti-thrombotic eicosanoids was totally inhibited by this NSAID, leading them to conclude that Indomethacin induces procoagulant activity in these cell cultures 1.

Schnaider et al., studying caudal auricular veins of rabbits (Oryctolagus cuniculus) after tenoxicam administration with its commercial diluent, have observed the presence of thrombosis in 19.4% of veins 2.

In a study by Nygard et al. with human umbilical vein endothelial cell cultures, procoagulant activity was time-dependent 1, whereas in the study by Schnaider et al., no statistically significant differences were found in rabbits among evaluation times 2.

The endothelial dysfunction alone, without exposure to the intima layer collagen, is enough to promote thrombosis. In general, the vascular endothelium has a non-thrombogenic surface due to the presence of surface heparan sulphate, a heparin-related glycosaminoglycan which, as heparin, is an anti-thrombin III co-factor. So, the endothelium plays an essential role in preventing platelet aggregation and intravascular coagulation. When aggressed, the endothelium induces heparan and tissue factor activation, no longer producing nitric oxide and prostacyclins, but producing endothelin. The two former agents, in addition to being vasodilators, also inhibit platelet aggregation and fibrin fixation to the white thrombus, while endothelin is vasoconstrictor and stimulates platelet aggregation and the coagulation cascade 7.

Thrombosis may be explained by the inhibition of anti-thrombotic eucosanoids, thus inducing procoagulant activity. This would be an alternative pathway for cycloxygenase inhibition with local platelet activation factor or leucotriens production.

Another hypothesis to explain thrombosis after NSAIDs administration would be the solution’s pH and its relation with drug transportation in the cell membrane. Bidistilled water has a pH of 6.34, 0.9% sodium chloride of 5.67, tenoxicam/diluent of 8.49 and tenoxicam/0.9% NaCI of 8.40. Cell membranes have polarized and hydrophilic surfaces with receptor glucoproteins and a wide variety of structural proteins, such as transmembrane. Drugs molecular weight, physical-chemical properties and solubility may influence their absorption. Lipophylic solutions with a pH close to neutrality easily penetrate the membrane. Tenoxicam, due to its alkaline pH, would be transported in the membrane, probably helped by a trasmembrane protein, directly inducing endothelial dysfunction or by means of reactive substances generated during metabolism.

In our study, thrombosis was observed in the Experimental group both after tenoxicam plus bidistilled water administration (19.4%) and after tenoxicam plus 0.9% NaCI (22.2%). In comparing the incidence of thrombosis in 6 h, 12 h and 24 h euthanasia periods for each group separately, no significant differences were found among periods, as shown in table I.

The conclusion was that tenoxicam, both with bidistilled water and 0.9% sodium chloride, produces thrombosis in the veins used for its administration.



01. Nygard G, Hudson M, Mazure G et al - Procoagulant and prothrombotic responses of human endothelium to indomethacin and endotoxin in vitro. Relevance to non-steroidal anti-inflammatory drug enteropathy. Scand J Gastroenterol, 1995;30:25-32.        [ Links ]

02. Schnaider TB, Andrade CHV, Juliano Y et al - Estudo histopatológico do efeito do tenoxicam e do seu diluente no endotélio venoso, em coelhos. Acta Cir Bras, 2000;15:118-123.        [ Links ]

03. Cotran RS, Kumar V, Robbins S - Robbins Patologia Estrutural e Funcional, 5ª Ed, Rio de Janeiro, Guanabara Koogan, 1996;84-109.        [ Links ]

04. Whittle BJR, Morishita T, Ohya Y et al - Microvascular actions of platelet-activating factor on rat gastric mucosa and submucosa. Am J Physiol, 1986;251:G772-G778.        [ Links ]

05. Wallace JL, Keenan CM, Granger DN - Gastric ulceration induced by nonsteroidal anti-inflammatory drugs is a neutrophil-dependent process. Am J Physiol, 1990;259: G462-G467.        [ Links ]

06. Wallace JL - Lipid mediators of inflammation in gastric ulcer. Am J Physiol, 1990;21:G1-G11.        [ Links ]

07. Rang HP, Dale MM, Ritter JM - Farmacologia, 2ª Ed, Rio de Janeiro, Guanabara Koogan, 2001;255-269.        [ Links ]



Correspondence to
Dr. Taylor Brandão Schnaider
Av. Francisca R de Paula, 289
37550-000 Pouso Alegre, MG

Submitted for publication July 19, 2001
Accepted for publication October 08, 2001



* Received from Faculdade de Medicina de Pouso Alegre, MG. Doutorado realizado no Curso de Pós-Graduação em Técnica Cirúrgica e Cirurgia Experimental da Faculdade de Medicina da Universidade Federal de Minas Gerais (UFMG)