To compare the efficacies and safety profiles of 3% diquafosol and 0.1% sodium hyaluronate in patients with dry eye after cataract surgery.
Randomized controlled trial.
setting : Soonchunhyang University Hospital, Seoul, South Korea. study population : In all, 130 eyes of 86 dry eye patients who had undergone cataract surgery between January 2014 and January 2015 were enrolled and randomly divided into a diquafosol group and a sodium hyaluronate group. intervention : The diquafosol group used diquafosol 6 times a day and the hyaluronate group used sodium hyaluronate 6 times a day after cataract surgery. main outcome measures : Evaluations of efficacy were conducted based on an Ocular Surface Disease Index questionnaire, tear breakup time (TBUT), Schirmer I test, corneal fluorescein and conjunctival lissamine green staining scores, serial measurement of ocular higher-order aberrations (HOAs), corneal HOAs, and uncorrected distance visual acuity test. Safety evaluations were based on anterior chamber inflammation and discontinuation of the eye drops.
Objective signs and subjective symptoms were aggravated at 1 week postoperatively and began to recover significantly 4 weeks after surgery. The diquafosol group showed significantly superior TBUT ( P < .001), corneal fluorescein ( P = .045), and conjunctival staining ( P = .001) compared to the sodium hyaluronate group throughout the study period. TBUT ( P < .001) and the change in HOAs ( P = .018) recovered significantly more quickly in the diquafosol group. The safety evaluations showed no intergroup differences.
Eye drops of 3% diquafosol may be an effective and safe treatment for the management of cataract surgery–induced dry eye aggravation in patients with preexisting dry eye.
Dry eye disease is a multifactorial disease of the tears and ocular surface that results in symptoms of discomfort, visual disturbance, and tear film instability with potential damage to the ocular surface. Its putative pathogenic mechanisms include hyperosmolarity of the tear film and inflammation of the ocular surface and lacrimal gland. The incidence of dry eye increases significantly with age.
Cataract is also a disease caused by aging, and cataract surgery can cause dry eye disease. In fact, several studies have indicated worsening of subjective symptoms and keratoconjunctival epithelial disorders of patients after phacoemulsification. The mechanism of dry eye after phacoemulsification is diverse. Transection of the corneal nerves by clear corneal incisions, damage to the corneal epithelial cells, exposure to microscopic light, vigorous intraoperative irrigation of the tear film, elevation of inflammatory factors in the tear film owing to ocular surface irrigation, intraoperative use of topical anesthesia, and postoperative use of topical eye drops can cause dry eye. Ocular symptoms such as soreness, pain, a burning sensation, foreign body sensation, and poor vision can result from dry eye. Severe dry eye affects the patient’s ocular and general health, well-being, and quality of life. Therefore, management of dry eye is necessary to improve patient satisfaction with treatment and quality of life.
Diquafosol, an activator of the P2Y2 receptor, promotes mucin secretion from the conjunctival tissue as well as tear secretion. This can improve corneal and conjunctival epithelial disorders. Diquafosol has been shown to be superior over sodium hyaluronate in terms of improving rose bengal staining scores in dry eye patients, and the 2 compounds exhibit similar efficacy in terms of improving fluorescein staining scores. On the other hand, as P2Y2 receptor agonists mediate the recruitment of leukocytes to the site of tissue damage and promote phagocytic clearance of apoptotic cells or bacteria by macrophages and neutrophils, diquafosol could possibly induce inflammation after cataract surgery.
Some studies have shown that higher-order aberrations (HOAs) after blinking tend to increase in dry eyes, and both ocular and corneal HOAs are significantly greater in dry eyes than in normal eyes. Serial measurement of HOAs may also be useful for evaluating optical quality in patients with dry eye.
In this study, we evaluated the efficacy and safety of 3% diquafosol ophthalmic solution compared to 0.1% sodium hyaluronate ophthalmic solution in patients with dry eye after phacoemulsification.
Study Group and Protocol
This was a prospective randomized controlled study. All patients were given a full explanation of the study, and written informed consent was obtained from all participants. The study protocol adhered to the tenets of the Declaration of Helsinki. This study was approved by the Institutional Review Board and Ethics Committee, Soonchunhyang University Seoul Hospital (SCHUH 2013-12-012), and registered at www.clinicaltrials.gov (identification no. NCT02608489 ). Patients were randomly allocated into the diquafosol group (D group) or the hyaluronate group (H group) using a simple unrestricted randomization method by the controller. The D group used 3% diquafosol tetrasodium ophthalmic solution (Diquas; Santen Pharmaceutical Co, Ltd, Osaka, Japan) 6 times a day and the H group used 0.1% sodium hyaluronate ophthalmic solution (Hyalein; Santen Pharmaceutical Co, Ltd) 6 times a day. Both groups instilled each eye drop from postoperative day 1 to postoperative week 12.
Consecutive patients aged 20–90 years with bilateral or unilateral cataract undergoing uncomplicated phacoemulsification and intraocular lens (IOL) implantation at Soonchunhyang University Hospital between January 1, 2014 and January 31, 2015 were enrolled. Patients with mild to moderate dry eye (level 1 or 2) based on the dry eye severity scale adopted by the Dry Eye Workshop (DEWS) report and who used only artificial tears occasionally were included. The exclusion criteria were presence of any complications after cataract surgery such as cystoid macular edema, patients using any topical eye drops on a regular basis, treatment history of dry eye beyond artificial tears, any ocular surgery within the prior 6 months, contact lens wear, serious ocular surface disease (eg, Sjögren syndrome, ocular pemphigoid, conjunctival scarring, chemical injury), lacrimal or eyelid disease (eg, anterior blepharitis or moderate to severe posterior obstructive blepharitis ), use of concomitant medications that could cause dry eye (eg, antihistamines, antidepressants, decongestants, anticholinergic drugs), and allergy to any of the study medications.
Objective and Subjective Clinical Assessments of Dry Eye
All patients underwent ophthalmic examinations preoperatively and postoperatively in the following order: uncorrected distance visual acuity (UDVA) test, an Ocular Surface Disease Index (OSDI) questionnaire, changes in HOAs after blinking, Schirmer I test without anesthesia, tear breakup time (TBUT), corneal fluorescein staining, and lissamine green (LG) conjunctival staining. Anterior chamber cells were evaluated by slit-lamp biomicroscopy to assess intraocular inflammation. Routine postoperative examinations were scheduled for 1 week, 4 weeks, and 12 weeks after surgery. The OSDI questionnaire consists of 12 questions that evaluate subjective symptoms related to dry eye and vision. We used total OSDI score to evaluate subjective symptoms of dry eye disease. TBUT was assessed by instilling a drop of 2% sterile fluorescein into the conjunctival sac and recording the interval between the last complete blink and the first appearance of a dry spot or disruption of the tear film. For the Schirmer I test without anesthesia, Schirmer paper strips were placed into the temporal one-third of the lower conjunctival sac for 5 minutes and the wetness on the strips was measured. Ocular surface damage was assessed by the National Eye Institute (NEI) workshop grading system, and corneal fluorescein staining and conjunctival LG staining were evaluated. The maximum staining score was 15 for the cornea and 18 for the conjunctiva. Corneal HOAs and serial measurement of ocular total HOAs were evaluated using a KR-1W wavefront analyzer (Topcon Medical System, Inc, Tokyo, Japan). Serial measurement of total ocular HOAs was measured every second for 10 seconds after complete blinking in continuous measurement mode. The difference between the fifth and first HOA was used to evaluate the tear film instability. The difference between the tenth and first HOA was inaccurate because many patients could not hold blinking for 10 seconds. Corneal trefoil, coma, tetrafoil, and second astigmatism were included in the corneal HOAs. Anterior chamber inflammation was examined with a slit lamp clinically and divided into 6 grades using the Standardization of Uveitis Nomenclature (SUN) working group grading scheme.
Efficacy and Safety Evaluation
To evaluate the efficacy of the 2 different eye drops, we compared each measurement between the 2 groups throughout the study period and at each follow-up, and also compared the patterns of changes in each group to identify which group recovered earlier. Previous studies have demonstrated that inappropriate activation of the P2Y2 receptor is associated with neutrophil-induced hyperinflammation and tissue damage during sepsis, chronic lung disease, and hepatitis in animal models. In addition, patients treated with diquafosol may suffer adverse drug reactions (23.7%), such as eye irritation, discharge, conjunctival injection, pain, pruritus, and other discomfort. Therefore, we evaluated the safety of the 2 eye drops, including anterior chamber inflammation and discontinuation of the eye drops owing to drug-related discomfort. Patients were taught to distinguish between the pain or discomfort due to the surgery itself and that due to the eye drops. The latter was defined as having started after eye drop instillation and lasting several minutes, occurring every time during instillation at any time up to 12 weeks into the follow-up period. Patients that had any adverse events or wanted to stop using the eye drops were excluded from the study, but these patients were included in the safety evaluation.
A 2.8 mm clear corneal incision was made at the location of the steep corneal astigmatism axis. A standard phacoemulsification technique was used with topical anesthesia with 2% lidocaine. A foldable IOL was implanted into the capsular bag. There was no suture at the corneal incision site. Patients received moxifloxacin (Vigamox; Alcon, Fort Worth, Texas, USA) and rimexolone (Vexol; Alcon) 4 times a day after surgery for 4 weeks.
For statistical analysis, UDVA was converted from Snellen into logMAR values. Baseline data were compared between the groups using the Mann-Whitney U test, Fisher exact test, and the linear mixed model after adjusting for intereye correlations, age, and sex. In addition, another mixed model was used to compare measurement data between the 2 groups by considering the correlation between both eyes of each patient at each follow-up. To obtain an overall comparison between treatment responses throughout the study period considering the 2 levels of correlation in subjects and follow-up, a bivariate generalized linear mixed (GLM) model with asymmetric random effects was used. In the last step, a bivariate GLM model was also used to identify the patterns of changes in measured data throughout the study period. The intergroup differences in adverse events were analyzed using Fisher exact test. SPSS software (version 21; SPSS, Inc, Chicago, Illinois, USA) was used for all statistical analyses, and P < .05 was taken to indicate statistical significance ( Supplemental Figure ).
A total of 212 eyes of 150 patients were assessed for eligibility and 130 eyes of 86 patients were enrolled. Of them, 34 eyes of 22 patients were lost to follow-up and 2 eyes of 1 patient discontinued intervention because of eye irritation. Ultimately, data from 94 eyes of 63 patients were included in the analyses. The demographic data are shown in the Table . There were no statistically significant differences between groups in the baseline data.
|Demographics and Baseline Data||Diquafosol Group |
(30 Patients, 45 Eyes)
|Hyaluronate Group |
(33 Patients, 49 Eyes)
|Mean age ± SD (y)||65.53 ± 11.15||65.37 ± 10.02||.550 a|
|OSDI ± SD||22.23 ± 14.65||23.64 ± 16.62||.623 c|
|TBUT ± SD (s)||4.88 ± 2.52||4.54 ± 1.85||.371 c|
|Schirmer test ± SD (mm)||3.52 ± 4.13||3.67 ± 2.88||.921 c|
|Fluorescein staining score ± SD||1.62 ± 1.77||1.77 ± 1.73||.530 c|
|Lissamine green staining score ± SD||1.55 ± 1.19||1.94 ± 1.56||.213 c|
|Total HOA changes ± SD (1–5 s)||0.042 ± 0.041||0.041 ± 0.048||.953 c|
|Uncorrected visual acuity (logMAR)||0.80 ± 0.52||0.77 ± 0.45||.787 c|
|Anterior chamber cells||0||0|
Regarding the OSDI score, the bivariate GLM model indicated that there were no statistically significant overall differences between the 2 groups throughout the study period ( P = .221). In addition, the patterns of changes in OSDI score also did not differ significantly between the 2 groups ( P = .453). However, the D group (9.76 ± 7.95) had a significantly lower OSDI score than the H group (16.92 ± 13.11) at postoperative week 12 ( P = .015). OSDI score tended to be significantly increased at postoperative week 1, and then started to recover toward baseline levels from 4 weeks after surgery in both groups. OSDI scores at postoperative week 12 were significantly lower than the preoperative scores in both groups (both P < .001) ( Figure 1 ).
Overall TBUT was significantly longer in the D group than in the H group throughout the study period ( P < .001). The values in the D group were 3.50 ± 1.35, 5.64 ± 1.89, and 6.69 ± 2.23 seconds at postoperative weeks 1, 4, and 12, respectively, and those in the H group were 3.00 ± 1.16, 3.96 ± 1.46, and 4.38 ± 1.92 seconds, respectively, significantly longer at postoperative weeks 4 and 12 in the D group than in the H group (both P < .001). The patterns of changes in TBUT were decreased at 1 week after surgery and then recovered toward baseline levels from 4 weeks after surgery in both groups, but recovered earlier in the D group than in the H group ( P < .001). TBUT at postoperative week 12 was significantly longer than baseline in the D group ( P < .001), while it returned to the baseline level in the H group ( P = .343) ( Figure 2 , Left).
Schirmer tests indicated no significant overall differences throughout the study period or in the patterns of changes between the 2 groups. However, the D group (4.10 ± 2.87) was significantly superior to the H group (2.52 ± 2.08) at 12 weeks postoperatively ( P = .025). The patterns of changes were similar to OSDI and TBUT, showing aggravation at postoperative week 1 and recovery from postoperative week 4 in both groups. There were no statistically significant differences between preoperative and postoperative week 12 data in the D group ( P = .505). Conversely, Schirmer test had not yet returned to the baseline level in the H group ( P = .016) ( Figure 2 , Right).
The overall corneal fluorescein staining score was significantly lower in the D group than in the H group throughout the study period ( P = .045). However, there were no significant differences at each follow-up or in the patterns of changes over time between groups. The scores at postoperative week 12 were significantly improved compared to the preoperative scores in both groups ( P < .001 for the D group, P = .025 for the H group) ( Figure 3 , Left).