Although recent findings from animal models and human studies have shown a promising effect of capsinoids on body weight reduction [9, 10], the effect of continuous administration of capsiate for 8 weeks on body weight in our population was negligible. Anyway, some unexpected results emerged concerning BMD and body composition.
First, we observed a significant increase in BMD of the spine after 8 weeks of treatment with capsiate. This result appears in contrast with a recent observational study conducted on a large population of over 500,000 Chinese subjects showing an increased fracture risk in subjects with higher consumption of spicy food (OR up to 1.18 for subjects reporting daily consumption of spicy food) . As suggested by the authors, the action of capsaicin on bone metabolism could be dependent from the activation of the TRPV1 receptor, the same present in the digestive tract and that mediates the effects of capsaicin on induction of thermogenesis by brown adipose tissue (BAT) and mobilization of lipid reserves through the orthosympatetic nervous system [12, 13]. This receptor is actually a non-selective cation channel and it is expressed on most of the unmyelinated nerve fibers and on some small myelinated sensitive nerve fibers [8, 14], acting as a nociceptor [14–16]. CGRP (calcitonin-gene related peptide), on the other hand, is a neuropeptide widely distributed in the central and peripheral nervous system  whose release from sensory nerve terminals appears to be stimulated by the local administration of capsaicin  and inhibited by TRPV1 antagonists . In vitro and in vivo experimental studies have demonstrated the inhibitory effect of CGRP on osteoclastic activity and, consequently, on bone resorption , perhaps even modulating the action of calcitonin . Furthermore, the CGRP receptor appears to be expressed also on osteoblasts, of which it regulates the differentiation from medullary progenitors . In the literature, there is not concordance on the effect of capsaicin on bone metabolism, as several experimental studies have provided conflicting results. Rossi et al., for example, reported that genetic or pharmacological (by administration of palvanil) inactivation of TRPV1 receptor appears to be protective against osteoclastic activity, reducing bone resorption in mice . Similarly, Offley et al. administered capsaicin or placebo to 16 rats that subsequently underwent DXA examination and bone histomorphometry. After 4 weeks of treatment with capsaicin, the authors reported a decrease in BMD on the metaphysis of the tibia and femur. At histology, the number of osteoclasts and the resorption surface increased in the proximal tract of the tibia, while the osteoblastic activity and the volume of trabecular bone decreased. Fifty-seven percent destruction of the unmyelinated fibers was also reported, with reduced levels of CGRP in the sciatic nerve and in the proximal tract of the tibia . In another study, rats were treated with capsaicin in a single dose, then after 14 days of treatment underwent tooth extraction and were killed after further 4 days. The study presented both a control group and a placebo group. Differently from the results of the above-mentioned studies, in the group of animals treated with capsaicin there was a 40% and 54% reduction in bone resorption surface compared to placebo and controls, respectively, and an absolute 26% and 34% reduction of osteoclasts and cells in active reabsorption . Similarly, Hill et al., in another experimental study conducted on animals (47 rats in 3 groups: guanethidine, capsaicin, control) reported that sympathetic denervation at birth by capsaicin causes a 21% reduction in alveolar bone resorption after tooth extraction in adult animals . A partial explanation for the different effects of capsaicin in animal models has been provided by Holzer. Indeed, he states that most studies on capsaicin have been conducted on small rodents (mice, rats, guinea pigs) and it is important to note that the effects of low doses of capsaicin differ quantitatively and qualitatively from higher doses. At low doses (in the pg/kg range), capsaicin exerts a powerful excitatory effect on peripheral nerve endings (apparently only on type C unmyelinated fibers). The initial excitement, however, is soon followed by drug desensitization and nerve conduction block. In addition, the systemic administration of high doses of capsaicin (in the mg/kg range) has a toxic effects on sensory neurons, with the extent of the damage that depends on dosage, route of administration, species and age of the animal .
Concerning the effects of capsaicin in humans, there are no studies in the current literature evaluating bone metabolism. Interestingly, the preliminary findings of our study suggest that the administration of capsaicin may determine an improvement in bone mineralization, especially at spine level, which represents the skeletal segment richest in trabecular bone. Since extracellular calcium levels act on preadipocytes inducing adipogenesis , it could be hypothesized that adipose tissue deposition could be contrasted by biological mechanisms aimed at reducing circulating calcium levels, that typically underlie an antiresorptive mechanism on bone. Another possible explanation could be related to the ability of capsaicin to induce brown fat differentiation, as suggested by other authors . The presence of brown fat, indeed, has been associated with better BMD values in women [27, 28], suggesting its possible direct involvement in bone metabolism. On this purpose, Nirengi et al. used near-infrared spectroscopy to estimate the activity of brown fat in the supraclavicular region of 20 subjects (20 years old, 10 males and 10 females, normal weight) treated with 9 mg of capsinoids/day for 8 weeks, but neither body composition (except for visceral fat) nor bone mass assessed by DXA showed significant changes after treatment . In another study by Yoneshiro et al., 18 men underwent positron emission tomography (PET) examination (in a standardized cold condition) to distinguish BAT-positive from BAT-negative subjects, then they were all treated with a single dose of capsinoids (9 mg) or placebo, subjected to calorimetry and, after 1-3 weeks, the experiment was repeated with crossover mode. Since the highest energy expenditure response occurred in BAT-positive subjects after administration of capsinoids, the authors concluded that capsinoids, acting at the intestinal level through TRPV1, can activate the already present BAT and they also hypothesized that the capsinoids may induce the recruitment of new BAT in subjects who do not have it . In our sample, patients treated with capsiate showed a significant change in composition of the supraclavicular region with an increase in the percentage of adipose tissue and a reduction in the percentage of lean mass. Considering that supraclavicular region is an important site for BAT in adults , these data are relatively disappointing, since in the group treated with capsiate we would have expected an opposite variation, as an expression of a conversion of white adipose tissue into BAT, with an increase in lean mass and a decrease of fat mass. In fact, given the higher density of BAT, we would have expected an increase in the amount of tissue detected as lean, corresponding to the new BAT, at the expense of a reduction of the pre-existing white adipose tissue. Such a change would seem to be indirectly suggested by the radiological characteristics of BAT on computerized tomography (CT): BAT, compared to white adipose tissue, is characterized by higher density (assessed as Hounsfield Units) , even it in the literature data concerning the characteristics of BAT detected by DXA are lacking. In any case, this finding appears to be worthy of further study.
Epidemiological studies suggest that a reduction of 1 standard deviation (SD) in BMD is associated with an increase in the relative fracture risk of approximately 1.5-2  and a change of 1 DS corresponds approximatively to a 10% change in the BMD . It goes without saying that an increase in BMD of 4.7% represent a clinically relevant finding. In addition, such an increase was found after only 8 weeks of treatment, while in clinical practice densitometric checks are performed at 18-24 months intervals.
Despite these promising results, we are aware that there are important limitations in our study. First, the sample size is extremely low. This may have limited our ability to detect statistically significant changes in weight or body composition. In addition, the short treatment period, and the short time between DXA assessments may have reduced the magnitude of the effect induced by capsiate intake on BMD, leading us to underestimate it.