The august krogh principle: seeking unity in diversity: reply.

Abstract

Reply With attention and appreciation, we have read Dr. Geoffrey P. Dobson’s comments articulated in his most recent Letter to the Editor titled, “The August Krogh Principle: Seeking Unity in Diversity,” (1) which was provoked by our recent “Abandon the Mouse Research Ship? Not Just Yet!” review (2). First, we thank the author for his interest in our article. Second, we consider his discussion regarding the presence and potentially vital influence of torpor on various immunoinflammatory end points in the mouse models of severe trauma as a valuable extension of our “cons” argumentation included in the review. Indeed, neither torpor in the mouse nor the similar “hibernation-like” state present in other small mammals was discussed in our review—this (and other) omission(s) were intentional as the primary focus of our review was to provide a detailed commentary on the PNAS article (3). Consequently, the more general section on the utility of the mouse modeling in critical illnesses could not include all pertinent shortcomings. A famous statement by the Persian physician and philosopher, Avicenna (Ibn Sina), written in The Canon of Medicine, “...for testing on a lion or a horse (or a mouse) might not prove anything about its (i.e., experimentation) effects on man,” stresses that the doubts pertaining to validity of experimentation in various animal species (including the mouse) have echoed throughout the ages and have spanned the continents. Thus, the cautionary note from the Land Down Under (1) pertaining to the unique protective mechanisms (i.e., reversible hibernation-like state of acute hypometabolism) is very appropriate. The response to hypoxic insults present in the mouse but absent in larger mammals and humans (4) deserves the attention of fellow scientists who use different types of mouse-based shock models in their research. Yet again, the above discussion points to the critical importance (and complexity) of a proper match between a defined scientific question posited and the choice of model to be used for its verification. Depending on the investigative context, the same (mouse) model can be both misleading and well suited—even within the same research field. For example, while the concept of a hydrogen sulfide–induced “suspended animation” has failed to properly translate from the mouse to large animal species partly due to their metabolic idiosyncrasies (4,5), the opposite was true regarding some key features of organ pathophysiology during shock. Specifically, resuscitated, hyperdynamic polymicrobial septic shock in mice with well-maintained core temperature is characterized by impaired gluconeogenesis (6). More than 3 decades ago, Wilmore et al. (7) reported a failure to increase hepatic glucose release despite increased liver blood flow in patients with “complicated bacteremia.” Compared with patients, mice also show a strikingly similar pattern of septic cardiac (dys)function: in a landmark study, Parker et al. (8) reported that survivors of sepsis presented with a transitory (and reversible) reduction of left ventricular ejection fraction coinciding with increased end-diastolic ventricular volume, the latter being referred to as an adaptive mechanism to maintain stroke volume. In contrast, in nonsurvivors, systolic function and end-diastolic volume remained unchanged, suggesting a “stiffer” myocardium with disturbed diastolic relaxation. Despite the typical fall in the core temperature, the same pattern was present in mice challenged with cecal ligation and puncture (9): survival in “dilators” was twice as frequent as in “nondilators.” (10) Indeed, “all small animal models are not equal,” and this modeling inequality is further exacerbated by the wide-ranging palette of research questions we, the investigators, put forth for testing. As pointed out by Dr. Dobson (1) in his commentary and stressed in our review (2), it is vibrantly clear that mouse models are not the ultimate (and foolproof) station in preclinical critical care research. To the contrary, they constitute only the beginning of the path that may lead to the most desired gratification—a reproducible, lifesaving effect confirmed in a phase 3 randomized clinical trial. Nonetheless, the potential aid rendered to the design of the future clinical trials by aptly designed and executed mouse-based experiments cannot be overvalued, but this is a topic for yet another commentary. Marcin F. Osuchowski, DVM, PhD Ludwig Boltzmann Institute for Experimental and Clinical Traumatology in the AUVA Research Center Vienna, Austria Peter Radermacher, MD Institut für Anästhesiologische Pathophysiologie und Verfahrensentwicklung, Universitätsklinikum Ulm, Germany Daniel G. Remick, MD Boston University School of Medicine Boston, Massachusetts Irshad Chaudry, PhD University of Alabama at Birmingham Birmingham, Alabama