On the importance of long-term functional assessment after stroke to improve translation from bench to bedside

Regarding functional evaluations, the first point to consider is the body weight. Monitoring body weight changes after a stroke is of prime importance since postoperative weight loss may indicate feeding difficulties. Aside from all ethical considerations, such postoperative weight loss has been shown by some authors to be correlated with the extent of the lesion (extensive corticostriatal damage [1], or the involvement of the external carotid artery territory in the lesion [2]). An easy measurement such as this can advise as to the severity of the lesion. Beyond the lesion per se, feeding difficulties may also result from a reduced consciousness level or poor mobility due to anesthesia and/or surgery. For example, a surgical approach in which the temporal muscle is injured, such as the Tamura model [3] may induce severe mastication impairments, resulting thus in higher body weight loss. Nevertheless, a poor nutritional intake (Dennis, 2000) can be a bias, since it has been shown in patients [2] as in animals to have a negative effect on functional outcomes after stroke [2]. Body weight monitoring of patients has even been recommended as an index of functional outcome [4]. Thus, investigating weight changes in preclinical studies has to be recommended for all authors in the field, since it gives an independent and unambiguous assessment of animal welfare and safety. Animals should be weighed at least once before surgery and then regularly after. This parameter, accessible to everyone and not only to behaviorists since it does not require any specific skill, can also give, in some ways, information on how animals recover from surgery and can even be a prognostic index for functional outcome.

Concerning functional evaluations, few studies consider crucial long-term evaluation, even though it has been highly recommended during the Stroke Therapy Academic Industry Roundtables [5‐7]. As in clinical practice [8], animals display a large degree of spontaneous recovery within a short time after experimental cerebral ischemia [9‐12]. Even though demanding sensorimotor and cognitive tasks are powerful in revealing tiny deficits, long-term studies addressing such behaviors are unfortunately not very abundant. The characterization of long-term functional recovery is critical for evaluating the efficacy of potential therapeutic agents in an experimental stroke. Both acute (few days) and long-term (several weeks or months) evaluations have to be addressed in order to demonstrate a stable neuroprotection, and not only a slowing down of the lesion evolution [13, 14]. The issue of including behavioral assessments in animal stroke studies becomes even more critical with the recent interest in neurorestorative strategies, which requires a longer period of administration than classical treatments. Effectiveness of such strategies is more likely observable via changes in synapse number and dendritic structure, for example, than by changes in infarct volume [15, 16]. Since the ultimate goal of any stroke therapy is the restoration of functions that allow for a normal daily life of patients; an objective appraisal of the behavioral deficits should be done. Stroke-induced functional impairments can be divided into acute (pointing out effects of drugs on the rate of recovery - days or weeks) and long-term (pointing out the effects of drugs on the extent of recovery - several weeks to months). Ideally, a set of several different tests has to be performed to gather complementary information (see table 1: behavioral tests and time-points are given as an indication, and need to be adjusted according to the species/strain and the stroke model used). Several behavioral tests have been applied to ischemia research in regards to clinical criteria, from the simplest which measures global neurological status or motor reflexes (i.e. neurological score [17], limb placing test [18], cylinder test [19] - useful to assess an acute phase after stroke) to more complex tests assessing sensory and motor functions (i.e. adhesive removal [20] and rotarod or staircase [19, 21‐25]) - that are more relevant for the long-term phase. Similarly, cognitive tests such as those assessing memory functions are preferentially used for later time points because they require normalized motor functions [19, 22, 26‐28]. These behavioral tests have to be carefully chosen in accordance with the drugs tested and the nature of the targeted cerebral structures. Except from that of Winter and colleagues [27], very few publications deal with stoke-induced disturbances in emotional behavior. Because anatomical and functional brain regions are affected on a different timescale, and because treatments may also differentially affect those regions, it is our view and that of others [29] that direct (cortex, striatum) or indirect (thalamus) anatomical substrates hit during stroke, rather than global brain lesions, may be critical determinants of behavioral impairments and outcome.

Whereas the correlation between acute histological lesions and early behavioral impairment is well documented[24], less is known about the long-term evolution of this relationship. Correlation studies have to take into account the different brain structures, primarily or secondarily affected by stroke, in order to bring a better understanding of their involvement in behavioral impairments [23, 25, 30]. Additionally, the development of non-invasive methods (such as MRI) allowing longitudinal assessment of the evolution of the lesion may bring new insights to understanding the mechanisms underlying spontaneous functional recovery [31]. Longitudinal correlations must then be promoted since therapeutic agents targeting those mechanisms will free us from the current mandatory 3 to 6 hour therapeutic window.

The use of clinically relevant models taking into account associated factors and/or pathologies (i.e. aging[32], arterial hypertension, diabetes, ..) have, moreover, to be reinforced in the incoming studies. As such, the post-ischemic recovery may be different, depending on the presence of these aggravating factors.

The authors declare that they have no competing interests.