Pain and Laboratory Animals: Publication Practices for Better Data Reproducibility and Better Animal Welfare.
Pain and Laboratory Animals: Publication Practices for Better Data Reproducibility and Better Animal Welfare.
复制标题
DOI：
10.1371/journal.pone.0155001
复制
发表时间：
2016
期刊：
影响因子：
3.7
通讯作者：
Austin J
中科院分区：
文献类型：
--
作者：
Carbone L;Austin J
对实验室动物进行重大生存手术的科学家面临着双重福利和方法学挑战：如何选择最能控制动物痛苦的手术麻醉剂和术后镇痛剂，因为疼痛和控制疼痛的药物都会影响研究结果。发表动物程序的完整描述的科学家允许对数据进行批判性和系统性的审查，证明他们遵守动物福利规范，并指导其他科学家如何在该领域进行自己的研究。我们调查了一个相当勤奋的科学家在计划一个成功的实验时可能会发现什么关于动物疼痛管理的信息。为了探索不同领域的科学家如何描述他们对这种伦理和方法论问题的管理，我们对400篇科学文章进行了评分，这些文章包括将主要动物生存手术作为其实验方法的一部分，以确保麻醉和镇痛信息的完整性。这400篇文章（2011年之前接受发表的250篇，2014-15年接受发表的150篇，沿着他们引用的174篇文章）包括犬、灵长类动物、猪、小鼠、大鼠和其他啮齿动物的开胸术、开颅术、性腺切除术、器官移植、外周神经损伤、椎板切除术和骨科手术。我们对文章进行了发表完整性（PC）评分，这是任何提及麻醉剂或镇痛剂使用的内容;镇痛使用（Au），这是任何使用术后镇痛剂的内容;以及镇痛完整性（包括术中镇痛、延长术后镇痛和使用多模式镇痛的复合评分）。400篇文献中有338篇为PC。这338例患者中有98例为Au，其中一些患者提到了镇痛，而338例患者中有240例仅提到麻醉，但未提到术后镇痛。根据2013年的影响因子，期刊的口径对PC或Au没有影响。我们发现，期刊是否指示作者参考2010年在PC或AC上发表的ARRIVE出版指南对我们2014-15数据集中的150篇小鼠和大鼠文章没有影响。在302篇对镇痛药使用保持沉默的文章中，没有一篇明确声明镇痛药被停用，也没有一篇讨论疼痛管理或未经治疗的疼痛如何影响结果。我们的结论是，目前的科学文献不能被信任，以提供完整的细节使用动物麻醉剂和镇痛剂。我们报告说，出版指南更多地关注实验结果中偏倚的其他潜在来源，低估了疼痛和疼痛药物扭曲数据的可能性，因此在动物护理和使用委员会的管辖范围内，大多将疼痛管理视为纯粹的动物福利问题。与此同时，动物福利法规并不包括关于发表动物数据的指导，尽管发表是研究周期的一个组成部分，并且可能影响基于已发表工作的研究中的动物福利，让期刊和作者自愿决定发表动物使用的细节。我们建议期刊，科学家和动物福利监管机构应该修改目前的指南和法规，对疼痛的治疗和疼痛治疗的透明报告，以改善这种双重福利和数据质量的缺陷。
Scientists who perform major survival surgery on laboratory animals face a dual welfare and methodological challenge: how to choose surgical anesthetics and post-operative analgesics that will best control animal suffering, knowing that both pain and the drugs that manage pain can all affect research outcomes. Scientists who publish full descriptions of animal procedures allow critical and systematic reviews of data, demonstrate their adherence to animal welfare norms, and guide other scientists on how to conduct their own studies in the field. We investigated what information on animal pain management a reasonably diligent scientist might find in planning for a successful experiment. To explore how scientists in a range of fields describe their management of this ethical and methodological concern, we scored 400 scientific articles that included major animal survival surgeries as part of their experimental methods, for the completeness of information on anesthesia and analgesia. The 400 articles (250 accepted for publication pre-2011, and 150 in 2014–15, along with 174 articles they reference) included thoracotomies, craniotomies, gonadectomies, organ transplants, peripheral nerve injuries, spinal laminectomies and orthopedic procedures in dogs, primates, swine, mice, rats and other rodents. We scored articles for Publication Completeness (PC), which was any mention of use of anesthetics or analgesics; Analgesia Use (AU) which was any use of post-surgical analgesics, and Analgesia Completeness (a composite score comprising intra-operative analgesia, extended post-surgical analgesia, and use of multimodal analgesia). 338 of 400 articles were PC. 98 of these 338 were AU, with some mention of analgesia, while 240 of 338 mentioned anesthesia only but not post-surgical analgesia. Journals’ caliber, as measured by their 2013 Impact Factor, had no effect on PC or AU. We found no effect of whether a journal instructs authors to consult the ARRIVE publishing guidelines published in 2010 on PC or AC for the 150 mouse and rat articles in our 2014–15 dataset. None of the 302 articles that were silent about analgesic use included an explicit statement that analgesics were withheld, or a discussion of how pain management or untreated pain might affect results. We conclude that current scientific literature cannot be trusted to present full detail on use of animal anesthetics and analgesics. We report that publication guidelines focus more on other potential sources of bias in experimental results, under-appreciate the potential for pain and pain drugs to skew data, and thus mostly treat pain management as solely an animal welfare concern, in the jurisdiction of animal care and use committees. At the same time, animal welfare regulations do not include guidance on publishing animal data, even though publication is an integral part of the cycle of research and can affect the welfare of animals in studies building on published work, leaving it to journals and authors to voluntarily decide what details of animal use to publish. We suggest that journals, scientists and animal welfare regulators should revise current guidelines and regulations, on treatment of pain and on transparent reporting of treatment of pain, to improve this dual welfare and data-quality deficiency.