The aim of this systematic review was to examine the effect of glucosamine and chondroitin sulfate treatments in the synovial knee joint tissues and specific biomarkers of the osteoarthritic preclinical studies. A total of 22 studies with 3 different types of nutraceuticals: glucosamine sulfate, glucosamine hydrochloride and chondroitin sulfate, administered alone or in different combinations, were meticulously analysed in order to elucidate their direct influence on the main structural and biochemical elements in the OA joints.

Taking into account the experimental animal model, in the studies considered for this review, the most commonly used species were rats and rabbits, whereas only one study employed mouse as animal models [29]. These findings were different from those observed by other OA preclinical research, where mouse models constituted the majority of the included studies [49,50]. In regards to OA induction is concerned, in agreement with other publications, surgically induced models were one of the most selected based on the rapid OA induction, repeatability and lower costs [23,51]. Regarding spontaneous models, a slower disease progression was observed and therefore it seems closer to what naturally occurs in primary osteoarthritic disease [52]. However this review included only one study of spontaneous model of OA in guinea pigs [32].

Generally, in this systematic review we found a large inconsistency among the experimental nutraceutical protocols. As an attempt to reduce the variability among studies, we excluded the articles in which animals received intra-articular therapy injections [53,54,55,56,57]. Even though according to the records screened, these local therapies demonstrated a positive chondroprotective effect and anti-inflammatory activity, we decided to evaluate other administrations routes, such as oral and intraperitoneal, in order to analyse the systemic and non-local effects of glucosamine and chondroitin sulfate.

With respect to the therapeutic regimen, there are also notable differences among studies, both in the frequency and in the dosage administered. In this regard, 2 out of 22 articles included in this review examined the efficacy of chondroitin sulfate [27] and glucosamine sulfate [41] at different doses, and both of them concluded that they seemed to reduce the cartilage changes and biomarker alterations in a dose-dependent manner.

Additionally, some articles analysed the potential chondroprotective effect of different nutraceutical combinations. Furthermore, the combination of nutraceutical and other therapies and drugs was also investigated. In this context beneficial effects were observed in the association of glucosamine and risedronate [31,36], glucosamine and fish collagen peptide [37] and glucosamine hydrochloride and chondroitin sulfate plus fursultiamine, where only the combined group with the addition of the vitamin showed a significant chondroprotective effect [39]. Furthermore, in one study using an OA experimental model in rats, enhanced responses were observed with the association of glucosamine sulfate, chondroitin sulfate and photobiomodulation [43].

Regarding the nutraceutical combinations between glucosamine and chondroitin therapies evaluated in this review, Silva et al. [48] observed that the association of glucosamine sulfate and chondroitin sulfate, rather than isolated glucosamine, significantly reduced the joint pain and prevented the cartilage histology alterations. Likewise, Terencio et al. [44] demonstrated the chondroprotective effects of the combination chondroitin sulfate-glucosamine, as well as reduced inflammatory mediator levels. By contrast, Roman-Blas et al. [35] did not find any beneficial effects in the combined therapy with chondroitin sulfate plus glucosamine sulfate or hydrochloride. Regarding this point, a study which used a chemically induced murine model, focused on examining the effect of glucosamine on cartilage degradation and bone resorption, comparing two different pharmacological forms, sulfate and hydrochloride; the results showed less histologic effectiveness in the sulfate form when both were administered under the same conditions [29]. It is important to point out that some authors suggested that glucosamine hydrochloride had poorer bioavailability and less beneficial effect in relieving clinical OA symptoms [21]. In our review, slightly higher chondroprotective effects were determined in the glucosamine sulfate studies comparing to the studies that included the glucosamine hydrochloride formulation (67% vs. 50%, respectively). Furthermore, fewer biochemical alterations were found in the glucosamine sulfate administration compared to the hydrochloride ones (80% vs. 67%). Nevertheless, in a previous review on the use of glucosamine in the management of human OA, the authors determined that, due to the heterogeneous effects observed in the available research studies, concluding which formulation could be more effective continues to be extremely difficult [58].

Another point of interest in the experimental design is the therapy timing initiation in relation to the OA induction. As previously described, the articles included in this review were grouped into three distinct protocols, pre-emptive, early and delayed administrations. Among these studies, the highest chondroprotective effects were determined in the pre-emptive therapies, followed by the delayed ones, which showed slightly positive higher values than the studies with early administrations. These findings were slightly different to those observed in a recently published systematic review about the effect of bisphosphonates therapies in OA preclinical studies [59], where an obvious time-dependent efficacy on cartilage status was determined, showing better chondroprotective effects in pre-emptive and early therapy initiations and greater cartilage damage in the delayed ones. In our opinion, the positive values observed in the delayed administrations could be associated with an inadequate selection of the period of time determined, given that it can be established as early as 14 days after de OA experimental induction and longer periods of time may be required.

In terms of duration, attention is drawn to the lack of evidence in long-term therapy, identifying only 1 out of 22 included publications, in which glucosamine hydrochloride and chondroitin sulfate were evaluated at 8, 12 and 18 months, showing reduced cartilage degeneration and biomarker alterations in both treated groups, the animal group treated with glucosamine hydrochloride and the group treated with chondroitin sulfate [32]. In this context, it is important to highlight that the histological and biochemical response after long-term nutraceutical administration is basically unpredictable. The initial positive response identified in some studies may not be sustained for long periods of time. However, the opposite is also possible, and a longer duration of treatment period may be necessary to observe a beneficial effect in the synovial joint. Therefore, additional preclinical studies in OA research evaluating the effect of dietary supplements in the long term are required [19,59].

Overall, in this systematic review, we observed a high variability among the experimental designs. Consequently, making an accurate assessment of how glucosamine and chondroitin sulfate affect the OA progression continues to be a challenge. In general terms, the evaluated nutraceuticals, alone or in combination, did not seem to prevent the subchondral bone changes, the synovial inflammation or the osteophyte formation, showing poor positive responses. Nevertheless, it is true that only a few of the publications included evaluations at those levels. Cartilage continues to be the primary focus in OA research and in this sense, positive chondroprotective effects were identified in approximately half of the publications, the studies of glucosamine sulfate and the combination of chondroitin sulfate plus glucosamine sulfate showing the most promising results. There is also increasing attention on the research of biochemical markers. As it could be observed in this study, they were the second more assessed parameter. In this context, a positive response was identified in more than half of the evaluations included in this review.

Regarding the risk of bias and the quality assessments of the articles included in this review, they were similar to the previous studies [49,59]. There are essential details about the experimental design which continue to be poorly reported in the studies, such as the sample size calculation, which was only reported in one of the manuscripts [33]. Inclusion and exclusion criteria were also badly or incompletely reported in most of the studies as well as blinding the experimental details. More specifically, half of the studies did not report the information and the other half only specified it in the outcome assessment stage, but not in the experimental and treatment administration stages. In addition, the information concerning the acclimatisation period of the animals, the housing and husbandry was also insufficient. In this sense, a previous research evaluated the adherence to the ARRIVE checklist in 236 papers between 2009 and 2015, and unexpectedly none of the evaluated manuscripts fully reported 100% of the items [60]. Consequently, the improvement of the research report in animal experimentation continues to be an essential task at present [25].

To conclude, OA management in companion animals continues to be a challenge in veterinary medicine. As we exposed in this review, glucosamine and chondroitin sulfate seems to provide chondroprotective effects and less inflammatory biochemical response in approximately half of the evaluations. However, these effects are inconsistent between the clinical and the preclinical studies. One explanation may be related to the great variety of histological scoring evaluations, the potential assessor’s subjectivity and the possibility of intra- and inter-observer variations [61]. Moreover, as these therapies have a slow onset of action, long-term administrations should be required to clarify their effectiveness. Additionally, a possible caregiver placebo effect may explain some of the beneficial responses observed in clinical trials with dogs [17,62]. For all these reasons, the use of glucosamine and chondroitin sulfate should be an individual veterinary/owner decision, reached by thoroughly evaluating each particular clinical case and its symptomatic response.