There are several possible explanations to our finding of robust tear secretion in the absence of the HG and main LG. First, we specifically chose male rabbits to study in order to minimize possible influence from sex hormones on the conjunctival function. Sex hormone alteration, especially androgen deficiency, adversely affects ocular surface homeostasis.
3 Whereas previously, Chen et al.
25 used Japanese albino rabbits of both sexes in their studies, and Li et al.
41 used female New Zealand white rabbits only. Female rabbits, under the influence of hormones, possibly have a less robust capacity to compensate for the surgically induced dry eye condition. Second, bilateral surgery could also contribute to the discrepancy seen in our study. We believe bilateral surgery is important because DES is an immune mediated ocular surface disorder involving both eyes.
32,42 In addition, the regulation of lacrimal secretion involves the central nervous control: afferent trigeminal sensory fibers from the ocular surface (cornea and conjunctiva), superior salivary nucleus in the pons and efferent fibers through the pterygopalatine ganglion.
1 The neural pathway control, such as blink reflex to noxious ocular surface stimuli, involves bilateral somatosensory cortex.
43 As a result, the response to the unilateral, surgically induced dry eye would ultimately lead to a compensatory over secretion on the unoperated side. In our study, comparisons were made not between eyes of the same animal, but against the same eyes BE and AE at different time points. Third, there have been reports supporting that the main LG is not indispensable under certain conditions. Unilateral removal of the main LG did not result in keartoconjunctivitis sicca (KCS) in squirrel monkeys, although tear secretion decreased significantly.
44 In patients who underwent palpebral dacryoadenectomy for the treatment of epiphora, up to 86% did not develop dry eyes and up to 50% continued to have epiphora.
45,46 In this procedure only the palpebral lobe of the main LG is surgically removed through a conjunctival incision. However, because the tear ducts from the orbital lobe of the LG traverse the palpebral lobe, removal of the palpebral lobe essentially blocks orbital lobe secretion to the superior lateral conjunctival fornix. Fourth, even in some rabbit dry eye models created with similar surgeries as ours, significant reduction in tear secretion was not seen in all studies. Gilbard et al.
47 cauterized bilateral lacrimal excretory ducts in their model, in addition to excising the NM and HG. Although tear osmolarity increased and goblet cell density decreased significantly, tear secretion was unchanged. This was thought by others to be due possibly to incomplete cauterization of the excretory ducts,
25 but it is very unlikely that incomplete cauterization took place in all animals. In a study by Odaka et al.,
48 not all rabbits had evidence of dry eyes (inadequate tear volume, decreased goblet cell density, and KCS) 4 weeks after resection of the NM, HG, and main LG. Moreover, laboratory studies also supported that rabbit conjunctival epithelium has the capacity to be the primary source of TF. It was measured in one study that in rabbits, the basal conjunctival fluid secretory rate is 0.79 μL/min, whereas the basal tear production rate is 0.72 μL/min.
23 Another study demonstrated that the conjunctival fluid secretion was 175% greater than tear turnover.
24 Taken together, it is plausible to speculate that, at least in rabbits, alternative tear secretion can be augmented in compensation for the deficiency of the main LG.
6,46,49,21 The conjunctiva and the accessory LGs underneath it are most likely responsible for such compensatory tear production.