The use of periosteal autografts to resurface osteochondral defects was investigated in 10 horses (2 to 3 years old), and the repair tissue was characterized morphologically. Middle carpal joint arthrotomies were made, and osteochondral defects were induced bilaterally on the distal articular surface of each radial carpal bone. Each defect measured approximatively 1 cm2 and extended 3 mm into the subchondral bone plate. Residual subchondral bone plate of control and principal defects was perforated by drilling. A sterile fibrin adhesive was made by mixing a fibrinogen component and a thrombin component. A periosteal autograft was harvested from the proximal portion of the tibia and was glued onto the recipient osseous surface, with its cambium facing the joint cavity. Control defects were glued, but not grafted. Horses were walked 1 hour daily on a walker, starting at postoperative week 7 and continuing for 9 weeks. Sixteen weeks after the grafting procedure was done, carpal radiography was performed, after which horses were euthanatized. Quality of repair tissue of control and grafted defects was evaluated and compared grossly, histologically, and histochemically. Using a reticule, the proportions of various repair tissue types filling each defect were quantitated. Seven weeks after the grafting procedure was done, bilateral arthroscopy revealed synovial adhesions and marginal pannus formation in control and grafted defects. None of the autograft was found floating unattached within the respective middle carpal joints. At 16 weeks, the gross appearance of most grafted and nongrafted defects was similar, and repair was dominated by a fibrous pannus. In 4 grafted defects, bone had formed either concentrically within the defect or eccentrically in the fibrous adhesions between the defect and the joint margin. Histologically, all grafted and nongrafted defects were repaired similarly by infiltration of a mixture of fibrous tissue, fibrocartilage, and bone. Fibrous tissue was the predominant tissue in most defects and its mean proportion was 56 and 59% in the grafted and nongrafted defects, respectively. Fibrocartilaginous tissue in the deeper layers approximated 20%, and woven bone at the base of the defect was 20% in all defects. Histochemically, difference in staining for proteoglycans was not observed between grafted and nongrafted defects. Little remaining original periosteal graft tissue was evident at the defect sites. The only distinguishing feature of grafted defects was the presence of islands of bone formation either at the defect site (n = 2 horses), or in somewhat dorsally displaced tissue that was incorporated in fibrous adhesions (n = 2 horses). It was concluded that use of periosteal autograft did not improve the healing of osteochondral defects of the distal portion of the radial carpal bone. The repair tissue produced in grafted and nongrafted defects was similar and was principally fibrous in nature.

The effects of method of seminal collection and a diuretic on retrograde flow of spermatozoa into the urinary bladder of rams were examined. In experiment 1, semen and urine were collected from 8 rams during the nonbreeding season. Prior to seminal collection, all rams were given furosemide and a sample of urine was obtained during micturition. Semen was then collected from each ram with an artificial vagina or by electroejaculation in alternate weeks for 4 weeks, and the urine released during the first postseminal collection micturition was collected in 4 consecutive samples. The volume of electroejaculates was larger (P < 0.0001) than the volume of ejaculates, but the total number of spermatozoa in the electroejaculate or in the ejaculate were not different (P > 0.1). Urine obtained before seminal collection was azoospermic or contained few, nonmotile spermatozoa (mean +/- SD = 0.053 +/- 0.114 x 10(6)/ml). The adjusted spermatozoal concentration (mean +/- SD = 1.630 +/- 2.258 X 10(6)/ml) in the urine collected after seminal collection was 31 times higher (P < 0.0001) and there were motile spermatozoa in most (97%) of the samples. The spermatozoal concentration in sequential samples of urine was not different (P > 0.1) between samples and was not affected (P > 0.1) by the method of seminal collection. There was a trend, approaching significance (P = 0.052), for an effect of method of seminal collection on the percentage of retrograde flow. Retrograde flow ranged from 0.21 to 19.38% when semen was collected with an artificial vagina and from 0.03 to 94.60% when semen was collected by electroejaculation and varied (P = 0.02) among rams within the 2 methods of seminal collection. In experiment 2, the 8 rams used in experiment 1 were given injections of 0.9% physiologic saline solution or furosemide in alternate weeks prior to seminal collection with an artificial vagina. Furosemide increased (P = 0.009) the volume of urine voided during the first postejaculation micturition, but did not influence (P > 0.1) the time from exposure of rams to the teaser to ejaculation, seminal characteristics, number of spermatozoa in the urine, or the percentage of retrograde flow. There was a trend (P < 0.1) for more rams to have motile spermatozoa in the postejaculation urine after treatment with furosemide. Administration of furosemide prior to seminal collection facilitates the noninvasive collection of pre- and postejaculation samples of urine for the determination of retrograde flow.

Bilateral, midshaft metacarpal osteotomies were performed in 11 sheep and bilateral, midshaft radial osteotomies were performed in 7 sheep. The lesions were repaired with bone plates. One of each pair of plates was luted with polymethylmethacrylate and all screws were tightened uniformly with a torque screwdriver. Sheep were allowed unrestricted exercise after surgery. At 8 weeks, 10 of 11 sheep with metacarpal osteotomies were sound and both osteotomies were healing. Seven were lame on the limb with the unluted plate during the first 3 weeks; 4 were never lame on either limb. The screws of the unluted plates were significantly (P < 0.01) looser at 8 weeks than those in the luted plates. All of the sheep with radial osteotomies were lame in the limb with the unluted plate. Four of 7 sheep had overt loosening of the unluted plates. One sheep only had mild screw loosening with continued alignment of the osteotomy. Two of 7 sheep fractured the radius with the luted plate; these 2 sheep were lame in the limb with the unluted plate and were using the limb with the luted plate vigorously. Excluding the 2 sheep with fractures, all had substantially more screw loosening in the unluted plate. Histologically, there were no discernible differences in the vascularity or porosity of the bone under the luted vs the unluted plates. The only adverse consequence of the luting technique was introduction of a small amount of polymethylmethacrylate into the osteotomy gap in 5 bones.

The effects of IV administered amiodarone, a class-III antiarrhythmic agent, on myocardial contractility, early myocardial relaxation, and hemodynamic variables were evaluated in normal canine hearts and those with infarcts. In the normal canine heart, amiodarone had important, but relatively mild, depressant effects on left ventricular contractility (assessed by maximal positive first derivative of left ventricular pressure (+dP/dt(max)) and maximal elastance (Emax)) and heart rate when given IV at a dose of 10 mg/kg of body weight. An effect on contractility or active relaxation (assessed by maximal negative first derivative of left ventricular pressure(-dP/dt(max)) and the time constant of isovolumic pressure decrease) was not identified with smaller doses. Myocardial infarction itself caused a predictable and marked depressant effect on myocardial contractility, as indicated by decreases in +dP/dt(max) ejection fraction, Emax, and-dP/dt(max), and elevation in end diastolic pressure. Additional depressive effects on contractility and active relaxation resulted when 10 mg of amiodarone/kg was administered to dogs with myocardial infarction and these effects were sufficient to worsen acute myocardial infarction-induced heart failure. Significant changes attributable to heart rate alone could not be identified. On the basis of our findings, we suggest that amiodarone administered IV should be used with caution in dogs with compromised ventricular function.

To compare the effects of milk stasis and milk flow on Brucella abortus infection of the mammary gland under the same systemic conditions, primiparous goats (n = 5) were inoculated IV with B abortus on the day of parturition, and suckling by their neonates was restricted to one mammary gland. Goats were euthanatized and necropsied at 3 weeks after inoculation, and milk, mammary glands, and supramammary lymph nodes were evaluated by bacteriologic, histologic, and immunoenzymatic staining techniques. Nonnursed mammary glands had high titers of brucellae in milk, moderate interstitial mastitis, and brucellar antigen in macrophages located primarily in alveolar and ductal lumina. Brucellae often filled the macrophage cytoplasm. In contrast, nursed mammary glands had fewer brucellae in milk, minimal inflammatory changes, and no detectable brucellar antigen in histologic sections. Hyperplastic changes were only seen in supramammary lymph nodes draining nonnursed mammary glands; these contained more brucellae than lymph nodes draining nursed mammary glands. These studies show that milk stasis may be the sole cause of increased susceptibility of nonnursed mammary glands to B abortus infection.

Erythrocyte insulin receptor binding measurements were evaluated in 8 dogs with spontaneous hyperadrenocorticism. These dogs had normal serum glucose concentration, with normal to high serum insulin concentration (range, 45 to 1,400 pmol/L; normal, 40 to 170 pmol/L). Dogs with hyperadrenocorticism had significant (P < 0.01) decrease in mean +/- SEM percentage of maximal binding for erythrocyte insulin receptors (2.25 +/- 0.21%), compared with results in 11 clinically normal pet dogs (4.29 +/- 0.42%). The decrease in erythrocyte receptor binding was attributed to significant (P < 0.01) decrease inhigh-affinity receptor sites in dogs with hyperadrenocorticism (14.5 +/- 2.8), compared with clinically normal dogs (31.2 +/- 4.3). Significant differences in receptor affinity were not apparent between the 2 groups. Percentage of maximal binding for erythrocyte insulin receptors for dogs with hyperadrenocorticism was inversely correlated with serum insulin concentration (r = - 0.85, P < 0.01). Results indicate that the observed decrease in erythrocyte insulin receptor binding could contribute to insulin resistance and hyperinsulinemia associated with hyperadrenocorticism. Alternatively, decreased binding of insulin receptors in animals with hyperadrenocorticism may result from down-regulation secondary to hyperinsulinemia itself caused by insulin resistance at a postreceptor site (decreasedresponsiveness).

Hepatic abscesses were induced experimentally in 5 steers by inoculating Fusobacterium necrophorum via ultrasonography-guided, percutaneous catheterization of the portal vein. Hepatic ultrasonography was performed to determine the onset and progression of abscessation. Blood samples were collected before and after inoculation for performing leukocyte counts and hepatic function tests. Ultrasonographic evidence of liver abscesses was observed as early as 3 days after inoculation. Abscesses appeared as hyperechoic centers (cellular debris and pus) surrounded by hypoechoic or anechoic areas (fluid). Increases in rectal temperature, leukocyte counts, fibrinogen, globulin, bilirubin, gamma-glutamyltransferase, and sorbitol dehydrogenase concentrations were detected. Hepatic dysfunction was evidenced by decrease in serum albumin concentration and low sulfobromophthalein clearance. The ultrasonographic diagnosis of abscesses correlated well with necropsy findings.

Nineteen purebred Beagles of various ages (4, 5, 13,and 47 weeks) were inoculated with North American Trypanosoma cruzi isolates obtained from an opossum (Tc-O), armadillo (Tc-A), or a dog (Tc-D). Dogs were grouped on the basis of clinical outcome of infection. During the acute stage of disease, dogs of group 1 (n = 7 inoculated with Tc-O or Tc-A) died or were euthanatized because of the severity of disease. Dogs of group 2 (n = 5 inoculated with Tc-O or Tc-A) developed acute disease, but survived to develop chronic disease. Dogs of group 3 (n = 7Tc-D-inoculated dogs) developed neither acute nor chronic disease. Dogs of group 4 (n = 4-2 dogs 13 weeks old and 2 dogs 47 weeks old) served as noninoculated controls. Clinical signs associated with severe acute myocarditis developed in dogs of groups 1 and 2 between postinoculation day (PID) 15 and 28. Generalized lymphadenopathy and lymphocytosis were observed in all dogs of groups 1, 2, and 3 between PID 14 and 17. Serum alanine transaminase and aspartate transaminase activities and urea nitrogen concentration were high, and glucose concentration was low prior to death of dogs in group 1. Serum activities of isoenzymes of creatine kinase were significantly (P < 0.05) high in only 1 dog (group 1), whereas serum lactate dehydrogenase isoenzyme activities were not significantly high in any dog. Parasitemia was detected by examination of thick blood smears as early as PID 3, peaked by PID 17 in most dogs, and was not detected by PID 33 in dogs of groups 1 and 2. Parasitemia was documented by blood culture results in dogs of groups 2 and 3 at various times throughout the study. Dogs infected at an older age generally had lesser degree of parasitemia and higher survival rate than did dogs infected at a younger age. Dogs of group 2 did not manifest clinical signs of disease for 27 to 120 days prior to onset of chronic disease. Ventricular-based arrhythmias and exercise intolerance developed in all dogs of group 2 at various times by PID 120. Two dogs developed signs of biventricular heart failure.

We investigated small-volume (5 ml/kg) 7% NaCl in 6% dextran 70 (HS/D70) as an alternative to large-volume (60 ml/kg) 0.9% NaCl for treatment of experimentally induced canine gastric dilatation-volvulus (GDV) shock. The stomach was surgically displaced and then distended with an intragastric balloon in 11 dogs anesthetized with pentobarbital. All dogs were subjected to GDV for 180 minutes before partial decompression and resuscitation. Hemodynamic values, blood gas values, and plasma volume were measured during control, shock, and resuscitation periods. Resuscitation started with 1 group (n = 6) receiving 5 ml of HS/D70/kg, IV, over 5 minutes, and the other group (n = 5) receiving 60 ml of 0.9% NaCl/kg, IV, over 60 minutes. Both groups received a surgical maintenance dosage (20 ml/kg/h of 0.9% NaCl after initial resuscitation. Resuscitative effects of small-volume HS/D70 were similar to large-volume 0.9% NaCl during the first hour of treatment; however, cardiac output was significantly higher in the HS/D70 group for the last 2 hours of resuscitation. Changes in heart rate, left ventricular pressure change, and systemic vascular resistance appeared to be responsible for improved perfusion. Mixed venous oxygen partial pressure data supported improved perfusion in the HS/D70 group. Packed cell volume remained higher in the HS/D70 group, indicating less hemodilution and improved oxygen delivery. Resuscitation of this GDV-induced shock model was better sustained with small-volume HS/D70, compared with conventional large-volume 0.9% NaCl.

In vitro adherence to intestinal epithelial cells by enterotoxigenic Escherichia coli strains bearing K88, K99, F41, or 987P adhesins and of their variants not bearing adhesins (K88-, K99-, or F41-) was investigated in European Large White and Chinese Meishan pigs. Possible relationship between adherence and virulence was also examined. The K88-positive (K88+) strain strongly adhered to intestinal epithelial cells from 26 of 28 Large White pigs. This strain had previously been found to be highly virulent for Large White pigs. The only surviving pig was of nonadherent phenotype, and cells from 4 dehydrated moribund pigs had strong adherence. By contrast, the same K88+ strain found previously to have little pathogenicity for Meishan pigs adhered with variable intensity to cells from 17 of 23 Meishan pigs; correlation was not evident between adherence and virulence. The K99+F41+ strain of porcine origin and the F41+ strain generally adhered strongly to cells from 24 and 23 Meishan pigs, respectively, and to cells from 25 of 26 Large White pigs. Correlation was not found between adherence and virulence for the 2 strains. A K99+F41+ strain of bovine origin adhered to cells from 20 of 22 Meishan and 22 of 23 Large White pigs, and a K99-F41+ variant adhered to cells from 19 of 23 Meishan and 23 of 24 Large White pigs. The adhesin-negative variants never adhered to intestinal epithelial cells. Strain 987 known not to readily produce 987P adhesin after in vitro growth never adhered to cells during the test. Results indicated that K88, K99, and F41 adhesins were responsible for in vitro adherence and, except for the K88+ strain in Large White pigs, adherent phenotype was not a sufficient condition to make a pig susceptible to enterotoxigenic E coli. The contribution of physiologic factors and their genetic origin to the degree of resistance of the individual is not yet completely understood for every enterotoxigenic E coli strain and breed of pig.