Reporting experimental studies on animals - The problems with translating of outcomes to clinical benefits. Methodological and statistical considerations: The example of pulmonary hypertension.

The variability inherent in animal models and the methods used to define drug response can lead to highly divergent results when evaluating new drug candidates. Several guidelines exist for high-quality and comprehensive reporting of experiments with animals. The present survey makes a quantitative demonstration of whether compliance with good preclinical practice guidelines can affect the results and reduce risk of over- or underestimation of treatment benefit. This meta-analysis was performed on more than 400 animal studies concerning pulmonary hypertension, where a wide range of potential therapeutic agents have been recently positively assessed in preclinical experiments. The experiments were reviewed according to the selected methodological and statistical items being recommended to report in papers. A quantitative evaluation was performed of their impact on effect size, which defined the efficacy of particular drug candidates. In addition, the paper also examines whether the quality score calculated for the papers included in the analysis changed over the previous 25 years. In conclusion, some information in experimental studies is often missed or incomplete, which complicates the correct evaluation and comprehension of obtained results. This in turn could subsequently increase the potential hazards involved in translating positive experimental outcomes to possible clinical benefits in patients.