Upsala J Med Sci 100: 125-1 36, 1995 Effect of Indomethacin on Thrombin-Induced Pulmonary Edema in the Rat Chul Min Ahn,1,3 H a a n Sandler,I Thomas Wegener2 and Tom Saldeenl Depurtments of 'Forensic Medicine and 'Lung Medicine, Uppsala Univrrsit?;, U p p a l a , Sweden. 3Department of' Internul Medicine, Yunsei Universit?; College of Medicine, Seoul, Korea ABSTRACT The preventive effect of indomethacin on thrombin-induced pulmonary edema was studied in rats. Administration of thrombin caused a significant increase in lung weight, wet weight to dry weight ratio (WWDW), and relative lung water content. During infusion of thrombin, mean pulmonary arte- ry pressure rose and mean systemic artery pressure fell, Pa02 decreased progressively and there was a continuous rise in pH and PaC02. An inhibitor of cyclooxygenase, indomethacin, at a dose of 1 m g k g body weight, induced a signifi- cant further increase in lung weight (p<0.05), and a tendency towards an increase in WW/DW and water content compared with animals given thrombin alone. Treatment with indomethacin, however, counteracted the elevated pulmonary artery pressure occurring in the early phase after thrombin infu- sion, but not that in the late phase. Systemic artery pressure was not affected by indomethacin. The increases in pH and PaCOz after thrombin infusion were attenuated and remained stable almost at baseline level after indomethacin administration. Tndomethacin did not prevent the hypoxemia indu- ced by thrombin infusion. In conclusion, although indomethacin prevented the early increase in pulmonary artery pressure due to thrombin and the decrease in pH and the increase in PaC02, it caused lung vascular permeabili- ty to protein to increase more than with thrombin alone. INTRODUCTION In both the dog ( 2 ) and the rat (6), pulmonary microembolism induced by infusion of thrombin during inhibition of fibrinolysis has been found to induce pulmonary insufficiency with similarities to the cli- nical adult respiratory distress syndrome. Infusion of thrombin results in systemic hypotension, acute pulmonary hypertension, and increased pulmonary vascular permeability to protein with subsequent pulmonary edema (12). The pathophysiology of thrombin-induced microembolism seems to involve the activity of cyclooxygenase-derived arachidonate metabolites (3). Evidence for this assumption is mostly based on studies using various cyclooxygenase inhibitors. Among these arachidonate metabo- 125 lites, thromboxane is known to be associated with sequences of events that follow infusion of thrombin in the rat (20), and with the development of thrombin-induced pulmonary edema in sheep (8,9). lncre- ased permeability to protein may also result from entrapment of fibrin (6) and leukocytes (2 1 , s ) in the lung. The cyclooxygenase inhibitor indomethacin is known to inhibit prostaglandin synthesis (24) and the release of thromboxane A2 in guinea pig lung parenchymal strips (26). It also inhibits several leu- kocyte activities in vitro such as neutrophil migration (4), aggregation (13), and adherence (251, and also the myeloperoxidase-H~02-C1 system of neutrophils (22). The aim of the present study was to test the effects of indomethacin on thrombin-induced changes in pulmonary artery and systemic artery pressure and in arterial blood gas exchange variables, and on pulmonary edema induced by thrombin, in rats. MATERIALS AND METHODS Animals Fifty-two male Sprague-Dawley rats (Alab, Stockholm, Sweden) were used. They were divided into three treatment groups: saline controls (n=8), thrombin plus AMCA (n=21), and thrombin plus AMCA plus indomethacin (n=23). All rats weighed between 200 and 250 g and had free access to food (Ewos rat pellet) and tap water. Mate rials Bovine thrombin (Topostashe') was kindly supplied by Hoffman La Roche, Switzerland. It was dis- solved in physiological saline to aconcentration of 100 IU/ml, and kept at - 20' C until used in theexpe- riment. The fibrinolytic inhibitor tranexamic acid (trans-4-aminomethyl-cyclohexane-carboxylic acid, AMCA) was purchased from Kabi, Stockholm, Sweden. Indomethacin (Confortid') was suppli- ed by Dumex AIS, Copenhagen, Denmark. Pentobarbital (Inactin') was obtained from Byk-Gulden, FRG. S-2238, a chromogenic substrate for detection of antithrombin activity, was obtained from Kabi, Stockholm, Sweden. Metliods The rats were anesthetized intraperitoneally with 125 m g k g of pentobarbital. They were placed in a supine position and tracheostomized. A tracheal cannula (PE 240, Clay Adams, Becton Dickson & Co., USA) was inserted for airway support and an abdominal cannula (Portex; outer diameter 0.80 mm, HYTHE, Kent, England) was inserted into the peritoneal cavity for AMCA injection and maintenance of anesthesia. All rats breat- hed air spontaneously through a tracheostomy, and the body temperature was kept constant at 38" C with an electric pad throughout the experimental period. Intravenous injection sites were prepared in one of the saphenous veins, using polyethylene cathe- ters (Portex; outer diameter 0.80 mm, HYTHEi, Kent, England) and 26 gauge needles for injection of indomethacin and thrombin. Indomethacin was dissolved in sterile distilled water to a final concentra- 126 tion of 2 mg/mL. The solution was slowly administered through a saphenous vein 30 min prior to the infusion of thrombin. Pulmonary microembolization was induced essentially as described previously (16) and modified as described below. In brief, 15 rnin prior to thrombin administration rats were injected through the abdominal catheter with 200 m g k g of AMCA to prolong fibrin entrapment in the lungs and increase pulmonary damage (2). Three hundred sixty I U k g of bovine thrombin was injected manually by the intravenous route over a period of 10 min, using a stopwatch and a 1 .O ml disposable syringe. The doses of indomethacin and thrombin used in this study was based on the findings in a preliminary study, in which animals treated with indomethacin did not tolerate the thrombin dose used in previous studies (500 IUkg). The throrn- bin dose was therefore reduced to 360 IUkg. The rats were killed 90 min after termination of thrombin infusion and the lungs were excised, gently cleaned with gauze, weighed, and dried at 37' C in a warm incubator for approximately 72 h. The lungs were reweighed until their weight was constant. Calculation of wet weight to dry weight ratio ( W W D W ) and water content The relative lung water content was calculated in the left lung as described previously (17). Water con- tent (%) = WW - D W / W W x 100. Pressure measurements Pulmonary artery pressure (PAP) was recorded via a SilasticR catheter (Dow Corning Co., USA; outer diameter 0.025 inch) inserted into the right jugular vein and then gently advanced to the pulmonary artery. Systemic artery pressure (SAP) was recorded via a polyethylene catheter (PE 50) inserted into the femoral artery. The catheters were connected to Statham transducers and the pressure tracing was performed throughout the experimental period. The position of the catheters was checked from the pressure tracings and confirmed when the rats were killed. Rats in which catheters were not properly positioned were excluded from further analysis. Arterial blood gas studies Blood samples of 100 pL were drawn into heparinized microcapillary tubes from the femoral artery and analyzed immediately with a blood gas analyzer (Instrumental Laboratory System 1302, Italy). The analyses were carried out at 37' C. Antithrombin activity Antithrombin activity was measured in vitro by means of a chromogenic substrate, S-2238. One vial of S-2238 was dissolved in 15 mL sterile water to a concentration of 0.75 nmoVL and 50 pL+ of test plasma or standard was diluted with 3.0 ml working buffer solution. Bovine thrombin 53(nkat) was reconstitu- ted with 1.5 mL sterile water. At assay, 400 pL of diluted test plasma or standard, and indomethacin in different concentrations were incubated at 37' C for 3-6 min. Then 100 pL of thrombin was added and incubated at 37' C for 30 min. S-2238 was thereafter added, mixed well, and incubated at 37O C for exactly 30 s. The reaction was stopped by addition of 300 pL of 50 % acetic acid, which was mixed 127 immediately. The absorbance of the sample was read against distilled water in a photometer at 405 nm. Statistical analyses Statistical analyses of differences between groups were performed with the Wilcoxon-White two- sample ranks test. Two-way analysis of variance was used to determine the significance of changes in mean PAP (MPAP), mean SAP (MSAP), and arterial blood gas exchange from the baseline value after infusion of thrombin. The baseline value was the value just prior to thrombin infusion. Student’s unpaired t test was used for comparisons between groups at equivalent time periods. A p value of < 0.05 was conside- red significant in all cases. RESULTS Changes in lung weight, WW/DU: and water content Time Schedule for Experiments Time schedule (Min) -30 -1 5 0 10 100 Saline AMCA Thrombi n Sacrifice 200 mg/kg 360 IU/kg i .v. i.p i .v. lndomethacin AMCA Thrombin Sacrifice 1 mg/kg 200 mg/kg 360 IU/kg i.v. i.p. i .v. The experiment was performed on 23 rats according to the protocol shown in Figure I . The results are presented in Figure 2 . Rats injected with thrombin + AMCA showed a significant increase in lung weight, WW/DW, and water content. Rats pretreated with indomethacin had a higher lung weight than those not treated with indomethacin (p<0.05). W / D W and the water content in the indomethacin- treated rats were slightly higher but not significantly different from those in rats given thrombin alone. 128 3 2.5 2 A v 01 E m 1.5 P = -I 1 0.5 0 Lung weight 10 9 6 7 * # 6 3 g 5 4 3 2 1 0 WW/DW * * T Water content * * @ Saline wntrd Saline &Thrumbin lndo ((I Thrombin Fig 2. Preventive effect of indomethacin, 1 mgkg, on changes in lung weight, WWlDW ratio, and relative lung water con- tent induced by thrombin inhsion in rats. * indicatesp