The majority of patients who undergo laser assisted 

 keratomileusis (LASIK) can expect to have some degree of transient postoperative dry eye. In fact, the association between refractive laser surgery and dry eye is so widely acknowledged it is not uncommon for a preoperative patient to ask his or her surgeon "How likely am I to get dry eye?"  

Despite the simplicity of this question, its answer is often less straightforward. Some asymptomatic patients with no history of dry eye disease develop very troublesome dry eye symptoms after LASIK while others experience no symptoms at all. Worse still, preoperatively distinguishing the former patient from the latter can prove difficult when traditional diagnostic tools are used. Patients with dry eye are also more prone to suboptimal visual results after refractive surgery, so preoperatively identifying the dry eye patient is extremely important. 

My colleagues and I performed a study - which was presented at the European Society of Cataract and Refractive Surgery (ESCRS) in September 2011. The study aimed to assess the efficacy of tear osmolarity as a predictor of visual outcomes following LASIK.  

In the study we used the TearLab Osmolarity Testing system (TearLab Corp., San Diego, California, USA) to measure the tear osmolarity of 256 eyes of 128 patients prior to LASIK, and again at one and three months after LASIK. We found that preoperative hyperosmolarity - defined as an osmolarity level above 308 mOsm/L - was predictive of longterm refractive results. Specifically, patients with elevated osmolarity had worse uncorrected visual acuities at three months post-LASIK than those with normal osmolarity.  

We also observed that patients with hyperosmolar tears did not have overtly abnormal corneal staining - indicating that traditional tests may not be sensitive enough to identify all cases of dry eye disease.  

A secondary aim of our study was to determine if the postoperative tear osmolarity values of LASIK patients differ with and without pretreatment with ocular lubricant. To assess this, we randomly allocated 60 patients to receive pretreatment with preservativefree Blink Tears (Abbott Medical Optics Inc., Santa Ana, California, USA) in addition to a standard postoperative treatment regime consisting of preservative-free Blink Tears, fluoroquinolone, corticosteroids and cyclosporine drops as appropriate. The remaining 62 patients received the same postoperative treatment but had no preoperative ocular lubricant.  

By the end of postoperative month one, we found that patients treated preoperatively with Blink Tears achieved normal postoperative tear osmolarity levels quicker than all other patients. Interestingly, we found that osmolarity was the only dry eye disease measure sensitive enough to track tear film changes over time. Indeed, patientreported symptoms and tear breakup time (TBUT) did not differ between the groups by postoperative month one. Corneal staining did differ between the two treatment groups, but only at a subclinical level.  

The identification and treatment of dry eye disease