Arlt S, Dicty V, Heuwieser W
Clinic for Animal Reproduction, Faculty of Veterinary Medicine, Freie Universität Berlin, Berlin, Germany.
Reprod Domest Anim. 2010 Dec;45(6):1052-8. doi: 10.1111/j.1439-0531.2009.01492.x.
The objective of this study was to evaluate the quality of published literature on reproduction in dogs. A systematic search in online databases revealed 287 papers that met the inclusion criteria. For evaluation a questionnaire comprising 40 criteria regarding materials and methodology, study design, statistics, presentation and information content, applicability and conclusions was developed. In a pre-test including seven independent scientists the applicability and explanatory power of the questionnaire and its results were validated. Out of 287 publications evaluated, 90 (31.4%) were classified as clinical trials. The remaining 197 publications were case reports or contained information based on personal experience. Not a single meta-analysis was found. Sixty (66.7%) of the 90 clinical trials included a control group. Randomization was conducted in 23 and blinding in eight articles respectively. In total five articles were determined as randomized, controlled and blinded clinical trials. Information content of the publications was variable concerning details on included animals, type or dosage of used remedies or conducted interventions. For example, in 99.7% of the articles, the exact number of animals was given, but in 79.8%, housing and feeding of the animals were not described. Statistical procedures of clinical trials were determined adequate in 55.6%. However, the data of 67.9% of the articles were evaluated to be not sufficient to draw valid conclusions. This study revealed evidence of deficits in the field of canine reproduction. The demand for more high quality clinical research is obvious. Requisite for the further implementation of the evidence-based veterinary medicine is an improvement of the quantity and the quality of well-designed, conducted and reported clinical trials. The practitioner should always assess the quality of information before implementing results into practice to provide best available care for the animals.
本研究的目的是评估已发表的关于犬类繁殖文献的质量。在在线数据库中进行的系统检索发现了287篇符合纳入标准的论文。为了进行评估，制定了一份包含40项关于材料与方法、研究设计、统计学、呈现方式和信息内容、适用性及结论的标准的问卷。在一项包括7位独立科学家的预测试中，对问卷及其结果的适用性和解释力进行了验证。在评估的287篇出版物中，90篇（31.4%）被归类为临床试验。其余197篇出版物为病例报告或包含基于个人经验的信息。未发现一篇荟萃分析。90项临床试验中有60项（66.7%）包括对照组。分别有23篇文章进行了随机分组，8篇文章采用了盲法。总共有5篇文章被确定为随机、对照和盲法临床试验。出版物的信息内容在纳入动物的详细信息、所用药物的类型或剂量或所进行的干预方面存在差异。例如，99.7%的文章给出了动物的确切数量，但79.8%的文章未描述动物的饲养和喂养情况。55.6%的临床试验的统计程序被判定为充分。然而，67.9%的文章的数据被评估为不足以得出有效结论。本研究揭示了犬类繁殖领域存在缺陷的证据。显然需要更多高质量的临床研究。进一步实施循证兽医学的必要条件是改进设计良好、实施得当且报告规范的临床试验的数量和质量。从业者在将研究结果应用于实践之前，应始终评估信息的质量，以便为动物提供最佳的现有护理。
