The orchidectomised rats in our study failed to gain weight normally until its body weight was significantly lower than the rest of the groups. Testosterone reduction due to orchiectomy may have caused the weight loss due to loss of muscle and bone mass . Testosterone and EL extract were able to prevent these reductions and therefore maintained the weight gain of the orchidectomized rats.
Orchidectomized rat is an accepted model for studying the skeletal effects of androgen deficiency in hypogonadal men . As expected, the testosterone levels of the orchidectomised rats in our study were significantly reduced while testosterone therapy significantly raised the testosterone levels. Similar results were obtained by other studies whereby orchiectomy had caused 80% reduction in serum testosterone in male rats , while testosterone replacement was able to raise the testosterone levels . Interestingly, EL extract supplementation to orchidectomized rats at 15 mg/kg for 6 weeks was able to elevate the serum testosterone level significantly. The only published report on the effects of EL on testosterone levels was by Zanoli et al. . They found that supplementation of EL extract to adult male Sprague–Dawley rats at 500 mg/kg for 6 days had significantly raised the testosterone levels compared to unsupplemented rats. The dose given to the intact rats by Zanoli et al.  was very much higher than ours, but the duration of treatment was shorter. A daily dose level of 270–350 mg/kg was found to be rather safe without any effects on the liver, kidneys, spleen, and testes .
The question now is how EL extract was able to raise the testosterone levels in the absence of the testes in our study. In rats, there are other extra testicular sites that synthesize androstenedione and testosterone such as the liver, kidneys and gastro-duodenal tract  but not the adrenal glands . Besides that, there was significant formation of 20α-dihydropregnenolone, the precursor of testosterone, in all rat tissues. The enzymes involved in testosterone synthesis such as 3β-hydroxysteroid dehydrogenase, 17β and 20α-hydroxysteroid dehydrogenases were also expressed extra-glandularly in rats . In humans, serum testosterone was detected above the sensitivity threshold even in patients who had undergone surgical bilateral orchiectomy. There is possibility that testosterone production is upregulated in those organs other than testis in surgically castrated patients with detectable serum testosterone measurement .
All these findings indicated that there will still be testosterone production even in the absence of testes. There is a possibility that EL extract was able to upregulate these extra testicular testosterone production in orchidectomised rats. More importantly, the testosterone-raising ability of EL extract may be responsible for its bone protective effects in orchidectomised rats.
Testosterone levels may vary between subjects and individuals. This was seen in mice of the same age and strain housed under identical conditions with the testosterone levels ranging from 1 ng/ml to 30 ng/ml. A two to five fold differences in plasma testosterone levels were recorded between samples collected from the same animal at different times . The wide variation in testosterone levels was also seen in human males, ranging between 250 to 850 ng/dl . The testosterone levels fluctuate throughout the day as it peaks in the morning and then gradually dropping throughout the day. We have attempted to reduce these variations by taking the blood samples at the same time at 9.00 am in the morning, when the testosterone levels were high. The testosterone levels also deteriorated with aging whereby its level dropped to half in a 60-year-old man compared to a young man .
Major changes in bone mass and micro-architecture can be observed four weeks post-orchiectomy and these alterations become more obvious after four months . The duration of treatment in our study was six weeks as in previous study by Shuid et al. . It should be sufficient for orchiectomy to produce significant bone changes. This was reflected by an elevated CTx level after orchiectomy, indicating increased rate of bone resorption. The elevation of CTx level was suppressed by EL extract supplementation and testosterone therapy. Therefore, both EL supplementations and testosterone therapy were able to prevent the increased bone resorption rate seen after orchiectomy. The results not only confirmed the findings by others that testosterone reduced bone resorption markers [9–11], but also confirmed an earlier study by Shuid et al.  that EL extract was capable of doing the same.
The bone resorptive changes were not accompanied by any significant changes in the biomechanical parameters. Previous studies have also failed to detect any significant change in the bone biomechanical properties of orchidectomised rats [42, 43]. The bone strength is determined by the bone mass and the intrinsic properties of the bone material [44, 45]. It is likely that there was no significant mechanical change in the organization of the bone matrix of the orchidectomised rat that may compromise the bone strength.
Since the bone marker changes in this study and another previous study by Shuid et al.,  were indicative that orchiectomy only affected the bone resorptive activity of osteoclast, we have studied the gene expressions of OPG, RANKL and MCSF. These factors play an important role in the osteoclast recruitment and activation. We found significant results only for the gene expression of OPG. Orchidectomy had down-regulated the OPG gene expression, which was elevated back to sham level with EL extract supplementation. However, testosterone therapy failed to emulate EL’s action. This suggested a novel regulation of OPG by EL, which may help us to understand the mechanism in protection against androgen-deficient bone loss. There were in vitro studies with mixed results on the effects of testosterone on OPG expression. Chen et al.  found that testosterone was able to amplify the OPG mRNA expression in a mouse bone-cell culture and in an osteoblastic cell line. In contrast, Hofbauer and Heufelder  demonstrated that androgens inhibited OPG production by mature osteoblast, marrow stromal cells and murine marrow stroma.
When the RANKL gene expression was measured in this study, there was no significant difference in the gene expression for all the groups. Proell et al.,  found that the RANKL expression was increased in the bone marrow of orchidectomised rats, while testosterone was able to return the RANKL gene expression to sham control levels. However two other in vitro studies did not find any association between RANKL and orchiectomy. Hofbauer and Heufelder  reported that the RANKL mRNA was not consistently detected in mature osteoblasts and marrow stromal cells and it was not regulated by androgens. Chen et al.  found that testosterone did not affect RANKL mRNA expression in MC3T3-E1 or mouse bone cells.
We did not find any significant change in the M-CSF gene expression after orchiectomy. There were also no significant differences in the M-CSF gene expression between all the groups. In vitro study on human endometrial stromal cells found that M-CSF production was dose-dependently enhanced by the addition of testosterone .
The present study has pointed out the potential of EL for treatment of male osteoporosis. Theoretically, the prevalence of male osteoporosis should be lower in South East Asian countries where EL is traditionally used as remedy for male wellbeing. There is no credible statistic on the incidence of male osteoporosis in Asian countries that used EL like Malaysia and Indonesia. By comparing the hip fracture incidence in men, the incidence is 88 and 114 per 100 000 in Malaysia and Thailand respectively . The incidence is 362 and 311 per 100 000 in Norway and Denmark respectively . Therefore the incidence is much lower in countries that used EL. However, the correlation with EL use is difficult to determine as the percentage of men taking EL is not known. Assuming that a significant number of men were taking EL in South East Asia compared to Scandinavia, it would explain why the prevalence of hip fracture in men in South East Asia was only one fourth to one third compared to Scandinavia.