We appreciate the response of Tong et al. to our recently published article.
1 First of all, we agree that the positions of * and † in Table 1 have to be swapped. The meibomian gland dysfunction (MGD) score criterion is mandatory for the DRYlip group, as stated in Material and Methods.
Regarding the classification of control (CTRL) patients, the patients included in the CTRL group were classified on the basis of clinical standard parameters, such as the basic secretory test (BST) and tear breakup time (TBUT), and the results of both tests had to be within normal limits. Further, these patients had neither subjective symptoms of dry eye, such as burning or itching, nor any other abnormalities. Thus, they must be classified as healthy. Excluding patients with a meibomian gland dysfunction (MGD) score >18 from this group and classifying these patients as having dry eye, exclusively on the basis of the MGD-score, would mean ignoring all other parameters and symptoms and classifying a proband as having dry eye, even if he does not show any symptoms or any of the discomfort related to the disease.
However, in response to the classification criteria suggested by Tong et al., the data provided for the comparison of CTRL and DRylip subjects may be biased, leading to a smaller group difference, as potentially occurs when applying their criteria. But, in comparison to group differences demonstrated for the comparison of CTRL subjects and patients with an exclusive aqueous deficiency, differences between CTRL and DRYlip are much smaller, even if CTRL patients with an MGD-score >18 are classified as DRYlip patients. Thus, one of the main outcomes of our study is that the increase in proinflammatory cytokines is much more prominent in patients with an exclusive aqueous deficiency—an outcome that is noteworthy and is absolutely unaffected by the parameters selected for classification of CTRL subjects.
Further, Tong et al. stated that MGD may have an effect on the protein composition of the tear film, as they published in June 2011 in the
British Journal of Ophthalmology.
2 This finding is a very interesting one, but we do not see it as a contradiction of the results of our study. Tong et al. analyzed several proteins in tears of MGD patients and found them to correlate with MGD severity, but their MGD group also included subjects with a strong deficit in the aqueous phase (mean BST of all study subjects: 5 mm; range, 1–17 mm). Thus, in part, these patients fit more the criteria for the DRYaqlip group of our study, in which subjects exhibited a distinct alteration of protein profiles.
However, the present discussion about the clinical phenotyping of dry-eye patients and controls demonstrates the importance of proper phenotyping and the urgent need for commonly accepted guidelines for classification and subclassification of dry-eye patients.