AJVR, Vol 63, No. 7, July 2002 979 G oniometry is the measurement of angles, particu- larly those formed by joints. These angles may be measured in a standing position or in flexion or exten- sion. 1-3 Goniometry is performed with a measuring device, typically a transparent plastic goniometer. Goniometry is a simple, affordable, and noninvasive method to quantitate the range of motion of joints. The accuracy and reproducibility of goniometry in humans have been documented, 4-13 and various goniometric methods have been compared. 5,6,14-21 Goniometry is used extensively by orthopedic surgeons and physical therapists in human medicine to the quantify baseline limits of joint motion, to aid decisions on appropriate therapeutic interventions, and to document the effec- tiveness of these interventions. 9,11,20,22 Goniometry has been similarly used in canine orthopedics to assess treatment efficacy for problems involving the carpal, 2 elbow, 23-27 stifle, 28 and hip joints. 1,29,30 In dogs, however, goniometry has not been validated, and only scant information is available regarding goniometric meth- ods and reference values. 31-33 Ranges of motion in dogs have been measured in 10 mixed-breed dogs in 1 report and 15 mixed-breed dogs in another report. 31-33 The objective of the study reported here was to evaluate the reliability of goniometry by comparing measurements made with goniometry with measure- ments made with radiography, assessing inter- and intratester reliability, and evaluating the effects of seda- tion on the range of joint motion. We intended to establish reference ranges for joint range of motion in adult Labrador Retrievers free of orthopedic diseases. We hypothesized that goniometry was valid and reli- able and that the range of motion evaluated under sedation was not significantly different from the range of motion evaluated in awake dogs. We hypothesized that less variation between measurements was present when distal joints (carpal and tarsal joints) were eval- uated than when proximal joints (shoulder and hip joints) were evaluated. We also hypothesized that there is a steep learning curve associated with the goniomet- ric method, and subsequently, there would be less vari- ation in the measurements for the last 5 dogs evaluat- ed, compared with those of the first 5 dogs evaluated in this study. Materials and Methods Dogs—Sixteen Labrador Retrievers (6 males and 10 females) were included in the study. Median age was 3 years (range, 2 to 7 years) and median weight was 32 kg (range, 27 to 46 kg). The 16 dogs were randomly selected from our local breed club for inclusion in the study. Sample size was deter- mined prior to the onset of the study by conducting a statis- tical power analysis (type-1 error, 0.05; type-2 error, 0.8) to determine the minimal number of dogs necessary to conduct statistical comparisons between study groups. The criteria for inclusion in the study were that dogs were ≥ 18 months of age, registered by the American Kennel Club, had no direct blood relationship with other dogs in the study, had no lame- ness or history of orthopedic disease or trauma, had normal results of an orthopedic examination, and had no radi- ographic evidence of joint disease. All dogs were evaluated within a 1-week period. Study design—One forelimb and the ipsilateral hind limb were evaluated on each dog. The side of evaluation was randomly chosen by use of a draw for the first dog and alter- nated for each subsequent dog. Fourteen joint positions were evaluated 3 times each by each investigator, including flexion and extension of the carpal, elbow, shoulder, tarsal, stifle, and hip joints, as well as valgus and varus movements of the car- pus. The 3 investigators independently made goniometric measurements of each dog awake and under sedation, which yielded 4,032 measurements. To assess intratester variation, 5 sets of 3 goniometric measurements were performed for all joint positions on 1 awake dog by 1 investigator with an interval of 15 minutes or more between each set of measure- ments. Goniometry—The arms of a transparent plastic goniometer were aligned with anatomic landmarks on the Received Nov 13, 2001. Accepted Feb 7, 2002. From the Department of Clinical Sciences, College of Veterinary Medicine, North Carolina State University, Raleigh, NC 27606 (Hahulski, Marcellin-Little); and the Program in Physical Therapy, University of Tennessee, Chattanooga, TN 37403 (Levine). The authors thank Drs. Xiaofeng He and Jay F. Levine for statistical assistance. Address correspondence to Dr. Marcellin-Little. Reliability of goniometry in Labrador Retrievers Gayle Jaegger, DVM; Denis J. Marcellin-Little, DEDV; David Levine, PhD,PT Objective—To evaluate the reliability of goniometry by comparing goniometric measurements with radi- ographic measurements and evaluate the effects of sedation on range of joint motion. Animals—16 healthy adult Labrador Retrievers. Procedure—3 investigators blindly and independent- ly measured range of motion of the carpus, elbow, shoulder, tarsus, stifle, and hip joints of 16 Labrador Retrievers in triplicate before and after dogs were sedated. Radiographs of all joints in maximal flexion and extension were made during under sedation. Goniometric measurements were compared with radiographic measurements. The influence of seda- tion and the intra- and intertester variability were eval- uated; 95% confidence intervals for all ranges of motion were determined. Results—Results of goniometric and radiographic measurements were not significantly different. Results of measurements made by the 3 investiga- tors were not significantly different. Multiple mea- surements made by 1 investigator varied from 1 to 6° (median, 3°) depending on the joint. Sedation did not influence the range of motion of the evaluated joints. Conclusions and Clinical Relevance—Goniometry is a reliable and objective method for determining range of motion of joints in healthy Labrador Retrievers. (Am J Vet Res 2002;63:979–986)