During a period of eight years (1996 – 2004) 432 small and miniature breed dogs were screened for patellar luxation (PL). In order to achieve the diagnostic accuracy required for genetic screening to assist breeding programmes, examinations were based on the concept of a standardized examination protocol for patellar luxtion. Diagnostic criteria assessed by physical examination, inspection and palpation focussed on lameness, evaluation of patellar tracking in the standing and recumbent position, with special focus on patello-femoral instability, as well as on the deviation of the tibial tuberosity and any perceivable crepitation of the stifl e joint during manipulation. Evaluation of all fi ndings was made on the basis of PUTNAM´s (1968) classifi cation. Radiographic examinations were not performed.Patellar luxation (unilateral or bilateral, medial and/or lateral) was diagnosed in 61.6 % of the examined dogs, but permanent lameness was only present in 15.5 % (right stifl e) and 12.8 % (left stifl e), respectively. Intermittent lameness was observed in only 3.5 % (right stifl e) and 4.6 % (left stifl e), respectively. This means that almost 40 % of all dogs with patellar luxation are asymptomatic and their condition would not have been detected without diagnostic screening. The different diagnostic criteria showed signifi cant correlation between each other and with the fi nal fi ndings. In some parameters like “luxation in standing position“ and “luxation in recumbent position“, the correlation with fi nal fi ndings was particularly high. Thus, the examination protocol used in this study appears to be suitable for PL screenings in dogs.Investigation of the infl uence of parameters like body weight, age, gender and neutering on the presence of PL showed that, except for gender, all attributes were associated with the occurrence of PL. An increase in body weight of 1 kg decreased the odds of suffering from PL to the 0.8fold (p<0.05), while an increase of age of one year increased the odds to the 1.1 fold (p<0.051). For neutered dogs, the odds showed a 3.1 fold increase of being affected by PL (p<0.05). No signifi cant infl uence could be observed for the gender of the animals. In order to detect breed predispositions for patellar luxation, odds ratios were calculated for all breeds represented in the study by more than ten animals. The breeds involved were: Jack Russell Terrier, Pug, Papillon, Pekingese, Shih Tzu, Tibet Terrier, West Highland White Terrier, Poodle, Yorkshire Terrier, Maltese Terrier and Chihuahua. Only two breeds showed odds ratios that were signifi cantly different from 1: In the Jack Russell Terrier, the odds ratio was signifi cantly lower (0.31) (95 % confi dence intervals 0.14-0.67), while the odds ratio was signifi cantly increased (5.62) in Poodles (Miniature and Toy Poodles) (95 % confi dence intervals 1.93-16.41). This means that Jack Russell Terriers have a comparatively reduced risk of suffering from PL, while the chance to develop the condition seems increased in Poodles. These results are highly indicative of a genetic background but further investigation on the basis of familial anamneses and heritability studies is required to support this postulate. A standardized examination technique and offi cial validation of the PL screening tests represent an essential precondition for the acceptance of PL screening programmes by breeders.Based on the results of this study, it is strongly recommended to implement a uniform, internationally valid and highly accurate diagnostic screening programme for patellar luxation. At the moment, this screening protocol is used by veterinarians in Germany, Switzerland and Austria.