Sample size and statistical power in the small-animal analgesia literature.

OBJECTIVE: To document the power and required sample sizes to achieve certain treatment objectives in the veterinary analgesia literature. METHODS: Pubmed's MEDLINE database and selected journals were searched. Only publications produced between 1994 and 2004 that reported 'no difference' between experimental groups in the abstract, results or conclusion sections and those that were randomised, prospective and blinded were reviewed. The data reported in the publications were then subjected to power analyses to determine the power and necessary sample size (to achieve a power of 0.8) to allow detection of 20 per cent, 50 per cent and 80 per cent treatment effects. RESULTS: Twenty-two studies provided sufficient data for analysis. Five out of 22 (23 per cent) had sufficient power to detect a 20 per cent treatment effect, 12 of 22 (54 per cent) had sufficient power to detect a 50 per cent treatment effect and 18 of 22 (82 per cent) had sufficient power to detect an 80 per cent treatment effect. The mean number of animals required per group to document a 20 per cent, 50 per cent and 80 per cent treatment effect were 90, 15 and 7, respectively. CLINICAL SIGNIFICANCE: Publications that report no significant difference between analgesic regimens may have committed a Type II error. The reader may inappropriately conclude that there is no difference between treatments when there may, in fact, be a superior analgesic regimen. Clinical practice based on the principles of evidence-based medicine could therefore result in suboptimal care for patients.

Measurement of velocity with a kinematic system versus a photocell system in the collection of canine ground reaction forces.

Velocities obtained from a five photocell system were compared to velocities of nine anatomical points on a handler and canine subject as reported by a kinematic system over the same distance. There was not a statistically significant difference between the velocities of the markers on the dogs' occipital protuberance and interscapular region compared with the velocity as reported by the photocell system. The average velocities of the three markers on the forelimb of the dogs and three markers on the handler's leg and one on the sacrum had statistically different values than the photocell system. Given these results, photocell systems with the same configuration in this study can be trusted to report accurate trunk velocities of canine subjects during the collection of ground reaction forces.

Measurement differences between three versus five photocells during collection of ground reaction forces in dogs.

The differences between velocities and accelerations obtained from three and five photocells were examined when obtaining ground reaction force (GRF) data in dogs. Ground reaction force data was collected 259 times from 16 different dogs in two experimental phases. The first phase compared velocities and accelerations reported by the two systems based on trials accepted by the three photocell system. The second phase accepted trials based on data from five photocells. Three photocell data were calculated mathematically in the second phase in order to compare the values of both systems. The velocity and acceleration values obtained from each system were significantly different (at the hundredth of a meter per second). Differences in measured values did not result in acceptance of data by the three photocell system that would not have been acceptable with the five photocell system (false positives), but did result in rejection of acceptable data by the three photocell system (11% false negative rate). Given the small differences between the two systems, GRF data collected should not be significantly different, though the three photocell system is less efficient in gathering data due to the number of trials rejected as false negatives.

Effect of durotomy on functional outcome of dogs with type I thoracolumbar disc extrusion and absent deep pain perception.

Spinal cord durotomy is performed as a diagnostic aid in determining spinal cord structural integrity, and this may be useful as an indicator of prognosis in cases with loss of deep pain perception (DPP). It has been suggested that a durotomy may relieve intramedullary compression but there is some debate about the therapeutic value. The purpose of this study was to compare ambulatory outcome of dogs that had loss of DPP treated with hemilaminectomy with durotomy versus hemilaminectomy without durotomy. Medical records of 81 dogs diagnosed with type I thoracolumbar IVD were reviewed. Dogs were included in the study if DPP was absent upon initial neurological examination and surgical decompression via hemilaminectomy was performed. Of the 81 cases, 48 were included in this study. The number of dogs that recovered ambulatory function were compared between durotomy and non-durotomy groups with a chi-squared test (p < 0.05). No differences were found. The findings of this study suggest that durotomy is useful as a diagnostic modality and that performing a durotomy does not significantly affect post-operative recovery of voluntary motor function.

Variability of vertical ground reaction forces collected with one and two force plates in healthy dogs.

OBJECTIVE: To compare peak vertical force (PVF) and vertical impulse (VI) data collected with one and two force plates during the same collection time period in healthy dogs at a trot. ANIMALS: Seventeen healthy client-owned adult dogs. METHODS: Vertical ground reaction force (GRF) data were collected in a crossover study design, with four sessions on two consecutive days, and then two weeks apart (days 1, 2, 15, and 16) using both one and two force plates collection methods. A repeated measures model analysis of variance (ANOVA) was used to test for differences in force plate PVF, VI, and average time per trial (ATT) between days, weeks, and systems (1 plate versus 2 plates). Coefficients of variation for PVF and VI were also calculated separately by forelimbs and hindlimbs, plates, day, and week. RESULTS: The time required to obtain a valid trial was significantly longer using a single force plate when compared with two force plates. Comparing GRF data for all dogs, significant differences in PVF data were found between one and two force plates, however, these differences were diminutive in absolute magnitude, and of unknown clinical importance. Examination of the coefficients of variation for PVF and VI during the different collection periods yielded similar results. CONCLUSIONS: Use of two force plates decreased trial repetition and collection time. Vertical GRF data had a similar coefficient of variation with either one or two force plates collection techniques in healthy dogs.

Comparison of cortical bone and serum concentrations of clindamycin achievable by direct local infusion and intravenous administration.

Tibial cortical bone and serum concentrations of clindamycin were compared using two drug delivery methods in dogs. An implantable drug pump, used to continuously infuse clindamycin directly into the cortical bone, was compared with clindamycin administered i.v. Dosage for the direct continuous infusion was 4 mg/kg/day, and 44 mg/kg/day for the i.v. bolus regimen. Serum concentrations of clindamycin were significantly higher during i.v. bolus administration when compared with those achieved during pump infusion (p less than 0.05). However, tibial cortical bone concentrations were significantly higher during pump infusion than were those achieved by i.v. bolus. When examining serum and bone clindamycin concentrations over 21 days of direct local infusion, there was no significant difference in concentrations between sampling days within each tissue (p greater than 0.05). Furthermore, there were significantly greater concentrations of clindamycin in the cortical bone than in the serum at each sampling period (p less than 0.05). Results indicate that delivery of clindamycin to canine bone by implantable drug pump achieve significantly higher bone concentrations than i.v. bolus administration of the drug at higher dosages. Direct infusion also can sustain high concentrations in cortical bone without increasing systemic concentrations of clindamycin.

Zidovudine serum, cerebrospinal fluid, and brain concentrations following chronic administration of a new zidovudine formulation via an implantable pump in dogs [corrected].

Zidovudine (AZT), prepared as an alkaline solution, was administered iv and intraarterially (ia) by continuous infusion via an implantable pump in dogs. The AZT serum and cerebrospinal fluid (CSF) concentrations were measured over a 28-day treatment period by HPLC. Terminal brain AZT concentrations were also measured. Control (vehicle only) animals were also studied. All animals were evaluated for pathological changes associated with the AZT and vehicle infusions in catheterized vessels and other organs. In the iv AZT treatment group, serum AZT concentrations were relatively constant, with individual coefficients of variations (%CV) of 20% or less. Mean CSF:serum and brain:serum AZT concentration ratios were 0.149 and 0.212, respectively. In the ia AZT treatment group, serum AZT concentrations were more variable than in the iv group, with %CV ranging from 22 to 79%. The fluctuations in serum concentrations were attributed to temporary blockages of the outflow catheter. Mean CSF:serum and brain:serum AZT concentration ratios were 0.126 and 0.249, respectively. Pathological changes, similar in both control and treatment groups, included endothelial denudation and myointimal proliferation at the infusion sites. The conclusions of the study are (1) steady-state greater than 1 microM AZT serum concentrations can be maintained chronically by use of an implantable pump containing a basic pH AZT solution; (2) ia delivery of AZT did not increase central nervous system uptake compared with iv administration; and (3) morbidity associated with the infused solutions does not seem to be a limitation for this mode of therapy.

Distribution of clindamycin in cortical bone during direct local infusion of the canine tibia.

The distribution of clindamycin in tibial cortical bone, administered via direct local infusion with an implantable pump, is described. Clindamycin concentrations in cortical bone were measured after 3, 7, and 21 days of intraosseous infusion. The tibia were divided into four quadrants relative to the outflow infusion catheter site located in the medial aspect of the mid-diaphysis. A gradient of 5-30 mm from the infusion site was documented in all four quadrants (proximal lateral, proximal medial, distal lateral, and distal medial). At all sampling times, clindamycin concentrations in all quadrants exceeded the minimum inhibitory concentrations for gram-positive aerobic bacteria, including Staphylococcus aureus and S. epidermidis, and both the gram-positive and gram-negative anaerobes, including Peptostreptococcus species and Bacteroides species. The data suggest that gravitational forces affect the diffusion of the clindamycin because concentrations in both distal quadrants were greater than in corresponding proximal quadrants.

Long-term temporal evaluation of ground reaction forces during development of experimentally induced osteoarthritis in dogs.

OBJECTIVE: To describe changes in vertical ground reaction forces (GRF) over 48 months in dogs with osteoarthritis (OA) of the stifle joint induced by transection of a cranial cruciate ligament (CCL). ANIMALS: 12 clinically normal adult dogs. PROCEDURE: Vertical GRF (eg, peak force and impulse) were determined prior to and 1, 2, 3, 6, 10, and 12 months after transection of the right CCL. In 7 dogs, data were also collected 24, 32, 38, 42, and 48 months after transection. RESULTS: Vertical peak force and impulse were significantly decreased in the right hind limb at all times after transection, compared with baseline values. From 10 through 48 months after transection, vertical GRF remained essentially static. Ground reaction forces in the unoperated (left) hind limb also changed significantly during the study. Left vertical impulse significantly increased 3 months after transection, whereas at 24, 38, 42, and 48 months after transection, left vertical peak force was significantly decreased, compared with the baseline value. Mean intradog coefficients of variation (CV) for peak vertical force and impulse ranged from 738 and 9.32, respectively, 1 month after transection to 1.96 and 2.76, respectively, at 42 months. CONCLUSIONS AND CLINICAL RELEVANCE: Vertical GRF in the affected hind limb equilibrated approximately 10 months after CCL transection. Prior to this, force transmission across the affected stifle joint changed significantly over time. Intradog CV were small, indicating that GRF may be an appropriate outcome measurement for evaluation of OA development induced by CCL transection in dogs.

Cortical bone concentrations of enrofloxacin in dogs.

Cortical bone concentrations of enrofloxacin were determined over time in dogs after SC administration of the drug. Nineteen healthy adult dogs were anesthetized and were given 2.5 or 5.0 mg of enrofloxacin/kg of body weight, SC. Serial serum and bone samples were obtained for determination of enrofloxacin concentrations at intervals until 8 hours after drug administration. Cortical bone samples were procured by surgical disarticulation of successive second phalanges. Additional cortical bone samples were taken from long bones in 4 dogs. Mean +/- SD peak serum enrofloxacin concentration was 0.54 +/- 0.10 micrograms/ml for the 2.5 mg/kg dosage and 0.97 +/- 0.34 micrograms/ml for the 5.0-mg/kg dosage. Serum concentration was significantly higher than bone concentration for each dosage. Mean peak bone concentrations reached 29% of peak serum values: 0.15 +/- 0.09 micrograms/g and 0.29 +/- 0.09 micrograms/g for 2.5-mg/kg and 5.0-mg/kg dosages, respectively. Serum concentration for the 5.0-mg/kg dosage was significantly greater than that for the 2.5-mg/kg dosage for all times, whereas bone concentrations for the 5.0-mg/kg dosage were significantly higher at all times after 180 minutes. For the duration of the study, cortical bone concentrations of enrofloxacin at either dosage exceeded the minimum inhibitory concentration (MIC) for the Enterobacteriaceae, but reliably exceeded the MIC for Staphylococcus sp only at the 5.0-mg/kg dosage.(ABSTRACT TRUNCATED AT 250 WORDS)

Pharmacokinetics of clindamycin phosphate in dogs after single intravenous and intramuscular administrations.

Clindamycin phosphate was administered to dogs at dosage of 11 mg/kg of body weight via IV and IM routes. The disposition curve for IV administration was best represented as a 2-compartment open model. Mean elimination half life was 194.6 +/- 24.5 minutes for IV administration and 234.8 +/- 27.3 minutes for IM administration. Bioavailability after IM administration was 87%. Dosage of 11 mg/kg, IV, given every 8 hours, provided serum concentration of clindamycin that exceeded the minimal inhibitory concentration for all Staphylococcus spp, as well as most pathogenic anaerobes, throughout the dosing interval. Intramuscular administration induced signs of pain and cannot be recommended.

Enrofloxacin concentrations in dogs with normal prostate and dogs with chronic bacterial prostatitis.

Chronic Escherichia coli-associated prostatitis was induced in 16 dogs; 9 noninfected dogs served as controls; and all dogs were vasectomized. Two to 3 weeks after instillation of bacteria directly into the prostate, the urine or prostatic fluid or both from 13 of 16 dogs were culture positive. Enrofloxacin, a fluoroquinolone antimicrobial agent, was administered orally to all dogs during the third or fourth week after surgery, at a dosage of approximately 5 mg/kg of body weight, every 12 hours for 7 days. Serum and prostatic fluid concentrations of enrofloxacin were concurrently measured in all dogs on days 2, 4, and 6 at 2 hours after dosing. Serum and prostatic tissue concentrations of enrofloxacin were concurrently measured in all dogs on day 7, at 1.5, 3, 4.5, and 6 hours after dosing. When values for these samples were compared, using a two-factor ANOVA, significant differences were not found. Use of this dosing regimen of enrofloxacin resulted in prostatic fluid and prostatic tissue concentrations exceeding the minimum inhibitory concentration of most pathogens that cause bacterial prostatitis. In addition, prostatic fluid-to-serum and prostatic tissue-to-serum concentration ratios were greater than 1.0.

Anatomic predisposition to perianal fistulae formation in the German shepherd dog.

Gross and microanatomic features which may predispose the German Shepherd Dog to perianal fistulae formation were studied in 2 groups of clinically healthy dogs: a predisposed group (German Shepherd Dogs) and a control group comprising breeds not ordinarily affected by perianal fistulae. The dimensions of the anal crypts (depth, base width, and length), measured and compared statistically between samples, identified no significant variation between groups (P greater than 0.05). Major tissue components of the anal canal were measured microscopically and were similarly evaluated: epithelial height in each zone, thickness of the lamina propria in each zone, thickness of the internal and external anal sphincter muscles, and density of the circumanal, sebaceous, and apocrine sweat glands. The only significant finding was an increase in density of apocrine sweat glands in the zona cutanea in the pre-disposed dog group. In a semiquantitative analysis of the inflammatory responses frequently seen in the anal glands, more mature fibroplasia was seen in the German Shepherd Dogs, indicating that inflammation was more longstanding in this group.

Force plate analysis of the walking gait in healthy dogs.

Ground reaction forces, impulses, and their relationships to morphometric measurements were evaluated for walking gait in 17 healthy dogs. A force plate was used to record forces at 1-ms intervals. Vertical, craniocaudal, and mediolateral forces were measured and normalized by body weight. Impulses, defined as the total force applied over time, were calculated in vertical and craniocaudal directions. Craniocaudal impulses were further divided into braking and propulsion phases. Braking impulses were significantly greater in the forelimbs (P less than or equal to 0.001), whereas propulsion impulses were generally greater in the hind limbs. Impulses and peak forces were then compared with morphometric measurements (body weight, humeral and femoral lengths, and paw length). All relationships were linear, with correlation coefficients significant (P less than or equal to 0.001). As the size of the dog increased, braking, propulsion and vertical impulses increased. Conversely, as morphometric measurements increased, peak vertical forces decreased. Thus, larger dogs had a lower peak force on each limb, but had a higher total impulse applied during stance phase. As stance phase time increased, peak vertical forces decreased. The results indicated that healthy dogs had significant correlations between ground reaction forces, impulses, and morphometric measurements.

Kinetic gait analysis assessment of meloxicam efficacy in a sodium urate-induced synovitis model in dogs.

OBJECTIVE: To examine the ability of meloxicam, a cyclooxygenase inhibitor, to mediate the effects of sodium urate-induced acute stifle synovitis in dogs. ANIMALS: 12 clinically normal adult hound-type dogs. PROCEDURE: A blinded, randomized, controlled single crossover design study was performed to determine the efficacy of meloxicam, using 2 dosage groups. In 2 experimental phases, dogs, according to group, received meloxicam (0.1 or 0.5 mg/kg of body weight) or matched volume of meloxicam vehicle, with a washout period of 21 to 28 days between phases. Blood samples for hematologic and biochemical analysis, as well as synovial fluid or cytologic analysis, were collected immediately before and approximately 24 hours after articular challenge of dogs under propofol anesthesia. Ground reaction forces (GRF) and subjective clinical scores were determined before and at 4, 8, 12, and 24 hours after articular challenge. Vertical force data included peak force, impulse, limb loading, and unloading rates. Craniocaudal data were divided into braking and propulsion phases and consisted of peak force and associated impulses. RESULTS: Except for propulsion impulse at 24 hours, all GRF variables were significantly greater at all post-synovitis induction times in the group receiving the high meloxicam dose. Significant differences in all GRF variables were seen at various times between the low-dose meloxicam group and the corresponding control group, and between the low- and high-dose meloxicam groups. Similar significance was seen in the subjective clinical evaluations. Strong correlations existed between the subjective and objective data. CONCLUSIONS: Meloxicam was effective in attenuating the effects of sodium urate-induced acute synovitis in dogs. Kinetic gait data provided an objective measurement of lameness in an experimentally induced arthritis model and quantified lameness improvements in response to medication with a nonsteroidal anti-inflammatory drug.

Vertical loading rates in clinically normal dogs at a trot.

Present data describe the rates of vertical loading and unloading generated by clinically normal dogs in a trotting gait. Forward velocity was found to influence maximal rates of limb loading and unloading in forelimbs and hind limbs. The rates increased as the velocity of the dog/handler increased. The position of maximal limb loading during the stance phase was independent of velocity in the forelimbs, but in the hind limbs, as velocity increased, the position of maximal unloading occurred earlier in the stance phase. Within velocity groups, the forelimbs had greater rates of vertical loading and unloading than did hind limbs. The position at which maximal loading occurred was earlier in the forelimbs than in the hind limbs. There was a difference in the position of maximal unloading between forelimbs and hind limbs, with the forelimbs unloading earlier in the stance phase. Difference between paired forelimbs or paired hind limbs was not found for any measurement within any group. Calculation of loading and unloading rates provides another method of examining functional limb loading in dogs. This method of analysis can be adapted to any animal gaited across a force platform in which single limb strides can be recorded. Calculations can also be done in any axis of measurement. Data indicated loading and unloading rates to be consistent and easily determined, Use of data generated from rates of limb loading can be classified into 2 areas: documentation of acceptance of load by the limb, and indirect measurement of functional stresses placed on bones of the appendicular skeleton.

Comparison of stance time and velocity as control variables in force plate analysis of dogs.

OBJECTIVE: To investigate effects of the use of stance time or velocity as control variables on ground reaction forces in lame dogs. ANIMALS: 12 dogs with pelvic osteotomies. PROCEDURE: Data for ground reaction forces were obtained preoperatively and at 2, 4, 6, 8, 12, 16, 20, 24, and 28 weeks postoperatively, using velocity and stance time as control variables. Ground reaction forces obtained were compared between the 2 methods of data collection, as were velocities and stance times of the trials. RESULTS: Significant differences in ground reaction forces were not found between the use of velocity or stance time as a control variable at any time. Also, significant differences in stance times or velocities were not found between the 2 methods of data collection. Greatest variation in stance time and velocity was found during periods of greatest lameness. CONCLUSIONS AND CLINICAL RELEVANCE: Use of stance time as a control variable in force plate analysis does not lead to significantly different results from use of velocity as a control variable, indicating that either method may be used in force plate analysis of dogs.

Force plate analyses before and after stabilization of canine stifles for cruciate injury.

Ground reaction forces were measured from the hind limbs of 9 dogs before and after stabilization of unilateral cranial cruciate ligament rupture. Before surgery, peak vertical force, associated impulses, and weight distribution were significantly less (multivariate analysis P less than 0.02) in the affected limb, compared with the clinically normal limb. Craniocaudal peak forces and impulses, divided into braking and propulsion, also were significantly less in the affected limb. At a minimum of 7 months after retinacular imbrication, all vertical and craniocaudal measurements in the affected limb were increased significantly. Significant changes were not found in the normal limb. Furthermore, at the postoperative evaluation, there was no significant difference in any measurement between the affected and normal hind limbs. The results indicated restoration of function in the cruciate-deficient limb when compared with the clinically normal hind limb at a walking gait during the study time period.

Evaluation of limb symmetry indices, using ground reaction forces in healthy dogs.

Limb symmetry was evaluated by measuring ground reaction forces in 2 groups of normal-gaited dogs at a trot. Data were collected from 2 groups of 21 dogs trotted at dog/handler velocities of 1.25 to 1.55 m/s and 1.85 to 2.05 m/s, respectively. Of these dogs, 9 participated in both groups to allow comparison of data at both velocities. Additionally, 16 of the dogs in group 1 were measured in 2 directions of movement to determine whether directional dependence was present. Collected data were then applied to 3 described symmetry indices. Each index was easy to calculate, but all had limitations. A major limitation was variation in magnitude of ground reaction forces measured between the different axes and the effect of this variation on precision of the derived indices. Vertical ground forces provided the most consistent symmetry indices, in part because of their large magnitude. The indices indicated that no dog had perfect right-to-left symmetry during a trotting gait. Statistical differences were not found in any of the measurements of directional dependence. Likewise, comparing symmetry data in dogs trotted at both velocities indicated no significant differences in any axis. However, further analysis of the data revealed the actual amount that a variance attributable to right-left limb variation was negligible. Most of the variance was attributable to trial variation. Thus, the aforementioned indices, which use nonconsecutive footfall methods to evaluate limb symmetry, actually measure principally trial variation and not limb-to-limb variation.

Serum and tissue fluid norfloxacin concentrations after oral administration of the drug to healthy dogs.

Norfloxacin, a 4-quinolone antibiotic, was administered orally to 4 healthy dogs at dosages of 11 and 22 mg/kg of body weight, every 12 hours for 4 days, with a 4-week interval between dosing regimens. Serum and tissue cage fluid (TCF) norfloxacin concentrations were measured at 0, 0.5, 1, 1.5, 2, 3, 4, 5, 6, 8, 10, and 12 hours after the first and seventh dose of each dosing regimen. When administered at a dosage of 11 mg/kg, the mean peak serum concentration (Cmax) was 1.0 microgram/ml at 1 hour, the time of mean peak concentration (Tmax) after the first dose. After the seventh dose, the Cmax was 1.4 micrograms/ml at Tmax of 1.5 hours. The Tmax for the TCF concentration was 5 hours, with Cmax of 0.3 microgram/ml and 0.7 microgram/ml after the first and seventh dose, respectively. When administered at a dosage of 22 mg/kg, the serum Tmax was 2 hours after the first dose, with Cmax of 2.8 micrograms/ml. After the seventh dose, the serum Tmax was 1.5 hours, with Cmax of 2.8 micrograms/ml. The Tmax for the TCF concentration was 5 hours after the first and seventh doses, with Cmax of 1.2 micrograms/ml and 1.6 micrograms/ml, respectively. After the seventh dose, the serum elimination half-life was 6.3 hours for a dosage of 11 mg/kg and was 6.7 hours for a dosage of 22 mg/kg. For serum concentration, the area under the curve from 0 to 12 hours (AUC0----12) was 8.77 micrograms.h/ml and 18.27 micrograms.h/ml for dosages of 11 mg/kg and 22 mg/kg, respectively.(ABSTRACT TRUNCATED AT 250 WORDS)