Upsala J Med Sci 79: 39-44, 1974 Influence of Various Anaesthetic Agents on the Formation and Stability of Haemostatic Plugs in the Rabbit Mesentery D. BERGQVIST, F. N. McKENZIE and K.-E. ARFORS Department of Experimental Medicine, Pharmacia AB, Uppsala, Sweden, Departmeni of Surgery, University Hospital, Uppsala, Sweden, and Department of Surgery, University of Aberdeen, Aberdeen, Scotland ABSTRACT The effect of five different anaesthetic agents (urethane, chloralose, pentobarbital, fentanyl-fluanisone and ether) on the formation and stability of haemostatic plugs in transected microvessels of the rabbit mesentery was studied. Ether significantly decreased the frequency of rebleeding through previously stable haemostatic plugs. There were no signifcant differences between the other four agents although there was a tendency to de- creased plug stability during pentobarbital anaesthesia. It is concluded that, despite some theoretical disad- vantages, urethane offers the satisfactory form of anaesthesia static plug formation using the paration. most convenient and for studies on haemo- rabbit mesenteric pre- INTRODUCTION Observations o n the response to microvascular transection or puncture in the mesentery of ana- esthetised animals has been extensively used as a means of studying haemostatic plug formation. Different workers using the mesenteric technique have employed a wide variety of anaesthetic agents. Thus barbiturates (11, 12, 14, 15, 19, 24, 25), chloralose (16, 17), urethane (1, 8, 14, 17), mor- phine (16) and ether ( 1 ) are among the different anaesthetic agents which have been used. The milieu in which platelets aggregate to form a haemostatic plug should appropriately be con- sidered prior to interpretation of results obtained using the method. This view is strengthened by recent reports of anaesthetic-induced changes in platelet reactivity in vivo (6, 20). A study of the effect of five commonly used anaesthetic agents on the formation and stability of haemostatic plugs was therefore undertaken to define the suitability of different forms of anaesthesia in studies of the initial haemostatic mechanism, using the rabbit 1 mesenteric preparation. MATERIALS AND METHODS M a t e r i a l New Zealand white rabbits (weight 2.5h0.5 kg), fed on a standard diet (Teknosan pellets, AB Ferrosan, Malmo, Sweden) were used. Five groups, each of 5 aminals, were studied. 1. Ethylcarbamate (Urethane, Kebo, Stockholm, Sweden) 2. Alpha-D-glucochloralose (Chloralose@, Merck, Darm- stadt, West Germany) 3 . Sodium pentobarbital (NembutaP, Abbott Labora- tories Ltd., North Chicago, USA) 4. Fentanyl+fluanisone (Hypnorm@ a d us. vet., AB Leo, Halsingborg, Sweden) 5. Diethylether (Aether ad narcosin, Skinska Bomulls- krutsfabriks AB, Dosjebro, Sweden). In a further 3 groups, each of 5 animals, the effects of anaesthesia with urethane, pentobarbital or ether on arterial blood p H , pC0, and PO, were studied. A n a e s t h e t i c t e c h n i q u e 1 . Urethane was administered i.v. as a 20% solution in 0.9% saline. Supplementary doses were given inter- mittently as required t o a total amount of 1.4-r-0.1 g/kg. 2. Chloralose anaesthesia was induced using 5% chlora- lose in 5% sodium borate given i.v. (2) and maintained by intermittent i . v . infusion of 0.8% chloralose in saline. This solution was kept at 30 to 40°C and continuously stirred and before injection it was filtered. Local ana- esthesia with 1% xylocaine (Astra Lakemedel AB, Soder- talje, Sweden) was needed for skin incision t o suppress the hyperexcitability on tactile stimulation after clora- lose infusion. The amount of chloralose used for in- duction was 68.0?16.4 mg/kg and for maintenance 3 5 . 3 2 3 0 . 5 mg/kg. 3 . Sodium pentobarbital was administered slowly i.v. a s a 12.5 mg per ml solution in physiological saline. Small fortification doses were given a s required. The total amount used was 36.226.9 mg/kg. 4. Neurolept analgesia was achieved by intermittent i.m. injection of Hypnorm (fentanyl 10 mglml and fluanisone 0.2 rngiml). The ,mount used was 0 . 8 2 t 0.20 mltkg. 5 . Ether anaesthesia was achieved by drip on open mask. Upsula J Med Sci 79 40 D . Bergqvisf et (11. The injection were given into the marginal ear vein. With each agent, anaesthesia was kept at a depth suffi- cient to inhibit the blink and paw reflex while main- taining regular respiration. The depth of anaesthesia was assessed every 5 minutes. Operative technique The rabbits were starved for about 12 hours before the experiment. Before anaesthesia, a central ear artery was cannulated (Silasticm, Dow Coming, Midland, USA) for sampling and blood pressure measurement before and during induction. After induction, blood pressure was measured through a Teflon cannula (AB Stille-Werner, Stockholm) placed in the femoral artery, using a strain gauge pressure transducer (Statham) and an Ultralette U V recorder (ABEM, Stockholm). Pressure measure- ments were not obtained during induction with ether. The heart rate was obtained from the pressure curves. The procedure for haemostatic experiments is de- scribed in detail elsewhere (3, 4). Briefly, mesenteric arterioles and venules, each divided according to size into two groups, 20-<40 pm and 40-<60 pm, were cleanly transected, using a fresh Gillette scalpel blade, shape E/I 1 . The primary haemostatic plug formation time, i.e. from transection to the first cessation of bleeding, and the time and frequency of rebleeding through previously stable haemostatic plugs were re- corded. The sum of the primary haemostatic plug formation time and the rebleeding times in one vessel gives the total haemostatic plug formation time. In each animal 4 transections in each of the two arteriolar subgroups and 5 transections in each of the two venular subgroups were made. Each transected vessel was ob- served for 15-20 minutes. Laboratory technique Blood from the central ear artery was taken for de- termination of haematocrit and coagulation time before and 30 minutes after the start of the experiment and at the end of the experiment. Haematocrit was measured in triplicate using a micro-haematocrit centrifuge (10000 g for 5 min; International Equipment Co., Boston, USA). Coagulation time on 1 ml whole blood was determined in duplicate in unsiliconised glass tubes. At the same time intervals the bleeding time after cutting small peripheral ear veins, 0.2-0.5 mm, was measured. The mean value of four bleeding times was calculated on each occasion. In the 15 animals used for acid-base studies blood samples were obtained through a cannula inserted into the central ear artery before and at 15, 30, 60, 90 and 120 minutes after the induction of anaesthesia. The animals were other- wise prepared in the same way as the rabbits used for the haemostatic plug experiments. pH, pC0, and PO, were measured with an acid-base analyzer (model 213, Instrumentation Laboratory Inc., Lexington, USA). Statistical methods The haemostatic plug formation times from transected mesenteric vessels have a skew distribution (4). For statistical purposes the original values were rendered Upsala J Med Sci 79 more normally distributed by logarithmic transformation. The mean logarithmic haemostatic plug formation time for each vessel type in each group was calculated and the group means compared by one-way-layout analysis of variance (23). When this test showed significant differences between the groups, simultaneous 95% con- fidence limits were calculated for differences between the groups according to the T- and S-methods of Scheffe (22). The frequency of rebleeding was calculated for both vessel types and subgroups together for each animal and the differences in the frequencies com- pared using the rank sum test (9). This test avoids assumptions on the nature of the frequency distribu- tion of the data. The significance of changes in haema- tocrit, coagulation time, bleeding time, pH, pCO,, PO, and blood pressure was examined by the Student [-test. RESULTS The depth of anaesthesia was kept at an even level by frequent reassessment of the blink reflex, paw reflex and the rate and depth of respiration. Despite this, two animals in the chloralose group died of respiratory failure. They were excluded from the study. Under cloralose anaesthesia the rabbits tended to be hyperexcitable with excessive reaction to noise and exaggerated tendon reflexes. With neurolept analgesia, there is no real ana- esthesia, only profound sedation and analgesia. The values of pH, pCOz and pOz in three groups studied are shown in Table I . There was a significant decrease of pH at 15 minutes in the ether group, a slight but significant increase of pOz at 90 minutes in the urethane group and a significant decrease of pOz at 15 minutes in the pentobarbital group. During induction the mean arterial pressure increased between 20 and 30 mm Hg in animals anaesthetised with urethane and chloralose, some- what less after sodium pentobarbital and not at all after neurolept analgesia (Fig. 1). Thereafter, blood pressure in the urethane group decreased and remained constant at the control value while after chloralose, arterial pressure remained approxi- matively 10 mmHg above the control value. The immediate increase in pressure after sodium pento- barbital injection was followed by a progressive decline throughout the period of observation. After both neurolept analgesia and ether anaesthesia a gradual fall below control values was observed. The changes in arterial blood pressure within and between the different groups were not statisti- cally significant. The heart rate increased significantly after in- Influence of unaesthetic ugents on huemostatic plugs 41 Table 1 . Effect of urethane, sodium pentobarbital and Mean value and S.D. of 5 experiments in each group on arterial p H , p C 0 , and PO, s,p