Artificial tears potpourri: a literature review

Literature search

This study involved a review of the literature analyzing artificial tear treatments of DTS. The resources utilized were the electronic databases Medline (PubMed; http://www.ncbi.nlm.nih.gov), National Institutes of Health (NIH) clinical trials (http://clinicaltrials.gov), Google Scholar, and the Cochrane Library. Keywords used in the search included dry eye, dysfunctional tear syndrome, Actimist, Advance Eye Relief, Akorn, Akwa, Blink, Clarymist, Clear Eyes, Freshkote, GenTeal, Hylogel, Isopto, Just Tears, Lacril, Lacrisert, Liposic, Lubrifresh, Murine Tears, Natural Balance, Nature’s Tears, Nutratear, Oasis Tears, Paralube, Refresh, Rohto Hydra, Systane, Soothe, Tearisol, Tears Again, Tears Natural, Thera Tears, Ultra Tears, Visco Tears, and Vizulize Dry Eye Mist. Both published articles and data from clinical trials (including multicenter, blinded studies as well as open label, industry-funded studies) were included in this study. Studies were excluded if the brand and specific type of artificial tear could not be identified. Data from unpublished clinical trials and articles from artificial tears that are no longer being produced were excluded. No other exclusion criteria were applied.

The etiology of dry eye syndrome in the studies reviewed included the following conditions: environmental (humidity, pollution, etc), situational (reading/prolonged focus), contact lens wear, LASIK surgery, autoimmune syndromes, nutritional deficiencies, vitamin A, Stevens–Johnson syndrome, and irritant exposure. The patient subgroups were not excluded but were also not specifically identified in many of the studies during analysis of the treatment of DTS, and thus we cannot readily distinguish between these patient populations. In future studies, this is an issue which will need to be clearly addressed.

Data compilation

Data from all studies were compiled into one group. Each study had its own means of data collection and definition for successful treatment with artificial tears. Methods for collecting the subjective data were mainly through questionnaires including McMonnies Dry Eye Symptom survey,²⁴ Ocular Surface Disease Index (ODSI) questionnaire,¹⁸ impact of dry eye on everyday life (IDEEL) questionnaire,²⁵ Visual Analogue Scale, Salisbury Eye Evaluation (SEE) questionnaire,²⁶ Ocular Discomfort Severity questionnaire, direct brand comparison with drop preference selection, quality of life through the Measure Yourself Medical Outcome Profile-2 (MYMOP-2),²⁷ and Dry Eye Disease Comfort Assessment Score. Several studies also utilized custom questionnaires.^28–31 Objective data took the form of ocular surface staining using various dyes, usually fluorescein and rose bengal. Additionally, comparative data was collected through analysis of a variety of other parameters including Schirmer’s test, tear break-up time, post-therapy corneal topography, tear meniscus volume, mucinous layer analysis with rose bengal staining, tear osmolarity, lipid layer thickness (LLT), improvement of visual acuity, conjunctival hyperemia, Ocular Protection Index (OPI) (examining the tear break-up time divided by the interblink interval), number of eyelid parallel conjunctival folds, blinking time analysis through the OPI 2.0 system, Global Staining Score, the TearLab Osmolarity System, and tear film normalization test (measurement of lines of improvement in visual acuity after administration of artificial tears).^{28,29,32–36}

Commercially available tear film substitutes

To provide maximum utility for the patient and clinician, commercially available artificial tears were identified and categorized based on active ingredient (Tables S1–S5). Other over-the-counter treatments for DTS including gels, ointments, and sprays/mists were included in the compilation (Table S4). Active ingredients and preservatives were verified via package inserts for each product. Artificial tears were divided into groups based on active ingredients including hydroxypropyl methylcellulose (HPMC), carboxy methylcellulose (CMC), polyvinyl alcohol, homeopathic remedies, and the liquid polyols. A final group was made for other delivery methods including gel/ointments, spray/mist over-the-counter treatments, and the prescribed ophthalmic insert, Lacrisert^® (Valeant, Bridgewater, NJ, USA). Lacrisert was included in the study because its composition and active ingredient mimic those of some artificial tears.

Refresh versus Systane

The first study compared Systane^® (Alcon Laboratories, Inc., Fort Worth, TX, USA) to Refresh Tears^® (Allergan, Irvine, CA, USA) in 87 patients with dry eyes over a 6-week period.³⁷ Both investigators and patients were blinded to the artificial tear being used. The study evaluated the conjunctival/corneal staining and a custom symptom questionnaire at days 7, 14, 28, and 42. At days 14 and 28, patients using Systane^® showed significantly improved conjunctival staining compared to patients using Refresh Tears^®. At days 14 and 42, patients using Systane also had significantly decreased temporal corneal staining compared to patients using Refresh. Furthermore, subjective symptomatic improvement was significantly increased in patients using Systane when compared to those using Refresh.

Another study comparing Systane versus Refresh products utilized a three-way cross-over study design comparing Systane^®, Refresh Tears^®, and Refresh Endura^® (now called Refresh Optive^®; Allergan).³⁸ Including only patients with a history of dry eye signs or symptoms, 50 patients were evaluated using tear film breakup time (TBUT) and the OPI over three separate clinical visits. TBUT measurements were taken at 5, 10, 15, 20, 30, 45, and 60 minutes after tear application. Systane^® significantly increased TBUT compared to both Refresh Tears^® and Refresh Endura^® at 5, 10, 15, 20, and 60 minutes after artificial tear application.

A third study analyzing Refresh and Systane products compared the effects of Refresh Liquigel^® and Systane^® on corneal staining and symptomatic improvement in 60 patients.³⁰ A reduced sum score of corneal staining and a reduction of corneal staining from baseline were only observed in the Refresh Liquigel^® group (P=0.008 and P=0.019, respectively). Patient’s tear preference and comfort were also analyzed in this study; however, patients were not blinded to the artificial tear assigned. Using one eye for each artificial tear, patient preference was recorded 5 minutes after application with 36% of patients preferring Refresh and 24% preferring Systane. Limited value may be drawn from the subjective component of this study since unilateral symptoms may have been present in the study participants.

The largest of the studies comparing artificial tears was completed in Germany.³⁹ The study was a multicenter, observational study involving patients from 835 ophthalmologists. Data from 5,277 patients who required a change in their artificial tear formulation or were naive to artificial tear treatment were analyzed after 2–4 weeks of treatment with Refresh Optive^®. Patients had previously been using either Systane^®, Hylo-Comod^®, or Lacophtal^® (Ursapharm Arzneimittel GmbH, Saarbrücken, Germany). Nearly 85% of patients reported improvement in ocular comfort with Refresh Optive^®, and nearly 75% experienced an improvement in their symptoms after changing artificial tear treatment regimens. TBUT also significantly increased in patients using Refresh Optive^® from a mean of 7.7 seconds to 10.0 seconds (P<0.001).

In an industry-sponsored, Alcon, head-to-head clinical trial, visual acuity was measured after application of Systane^® Ultra and Refresh Optive^®.36 The study population included 48 patients with a history of dry eye. Each patient underwent visual acuity evaluation while completing a computer task at 15, 45, and 90 minutes postapplication. The amount of time the patients maintained their best-corrected visual acuity was the measured end point. At 90 minutes, the mean time best-corrected visual acuity was maintained for 9.17 seconds with Systane versus 6.84 seconds with Refresh. These data were not statistically analyzed.

Alcon sponsored another clinical trial comparing Systane^® Ultra and Refresh Optive^® with an emphasis on TBUT, corneal staining, and conjunctival staining after 0, 7, 14, 28, and 42 days of artificial tear use.⁴⁰ A total of 109 patients with a diagnosis of dry eye were enrolled in the clinical trial with 53 in the Systane^® Ultra group and 56 in the Refresh Optive^® group. Mean TBUT after 42 days was 4.5 seconds with Systane^® Ultra and 4.2 seconds with Refresh Optive^®. The study also compared corneal and conjunctival staining using a 0–10 scale, with 0 equaling no staining present. At day 14 and 42, Systane^® Ultra-treated eyes had a mean corneal staining score of 2.9 on both visits while Refresh Optive^®-treated eyes had a mean score of 4.5 and 4.2 for the respective visits.

Refresh versus Refresh

Allergan funded a project studying the efficacy of Refresh Tears^®, Refresh Ultra^®, and Refresh Optive^® in relieving the signs and symptoms of dry eye.⁴³ A total of 37 participants completed the study; 18 with a history of dry eye and 19 controls without a history of dry eye. In this three-way cross-over study, each patient used each artificial tear formulation for 2 weeks while TBUT, tear evaporation, osmolarity, tear structure, and patient symptoms were evaluated. All patients with a history of dry eye experienced improvement in both their signs and symptoms of dry eye with all formulations (P<0.05). In the control group, Refresh Optive^® and Refresh Ultra^® treatment resulted in a decreased rate of evaporation, significantly greater than Refresh Tears^® on cross-group comparison. Refresh Tears^® was the only formulation observed to decrease tear osmolarity in the control group (P<0.05). No other statistical difference appeared on cross-group analysis.

A combination of three separate, independent studies aimed to find the differences in corneal staining and OPI in patients treated with either Refresh Tears^® or Refresh Liquigel^®, an artificial tear product with a thicker consistency than most drops.²⁹ A total of 607 patients were evaluated after 1 month of using either Refresh Tears^® or Refresh Liquigel^®. With regard to change in OPI, Refresh Liquigel^®-treated patients showed greater improvements than those treated with Refresh Tears^® (P<0.05). Additionally, after 1 week and 1 month of use, patients treated with Refresh Liquigel^® showed significantly reduced corneal staining compared to those treated with Refresh Tears^® (P<0.001 and P=0.011, respectively). Although study subjects did report increased blurring after application with Refresh Liquigel^®, both artificial tears had statistically equivalent acceptability. Over a 1-month period, patients also reported using the Refresh Liquigel^® less frequently compared to Refresh Tears^® (P=0.05).

To assess the effects on blurring, distortion, and contrast sensitivity, Allergan funded a study to quantify changes in visual acuity after administration of Refresh Liquigel^® and Refresh Celluvisc^®, two products of thicker consistency than most artificial tears.⁴² In 20 normal subjects without a history of dry eye disease, artificial tears were applied at different time points followed by testing of contrast sensitivity via a computer controlled stimulus. Both artificial tears significantly reduced contrast sensitivity immediately after application (P<0.001); however, Refresh Celluvisc^® did so to a greater degree (P<0.001).

Further research compared the effect of Refresh Liquigel^® and Refresh Tears^® on TBUT in 39 patients.⁴² For each patient, either Refresh Liquigel^® or Refresh Tears^® was initially applied, and TBUT was measured at 5, 10, 15, 20, 30, 45, and 60 minutes. This was repeated 1 week later with the other artificial tear formulation. Order of administration was randomly assigned. At 5 minutes, both drops increased TBUT; however, only Refresh Liquigel^® significantly increased TBUT beyond 5 minutes. This effect was notable for up to 20 minutes postapplication.

Blink versus Refresh versus Systane

In a study of 60 participants, Blink^® Intensive (Abbott Laboratories, Abbott Park, IL, USA), Systane^®, and Refresh Celluvisc^® were compared at baseline and after 30 days of product use.⁴⁴ Tear osmolarity measurement, Schirmer tear test, TBUT, fluorescein staining, corneal wavefront aberrometry, and visual acuity were all measured. All three treatment groups demonstrated improvement in all measured endpoints. However, cross–group comparison found Blink to better reduce the tear osmolarity compared to both Systane^® and Refresh Celluvisc^® (P<0.001). No other statistical differences were found between the groups.

One unpublished, Abbot-funded NIH clinical trial with 80 patients compared Blink^® Tears to Systane^® Ultra with regard to TBUT and visual acuity 1 month after treatment.⁴⁵ Blink^® Tears proved superior to Systane^® Ultra in both TBUT (P=0.003) and improvement in visual acuity (P<0.001).⁴⁶

Blink^® Tears was also compared to Refresh Optive^® in a trial of 51 patients with a history of dry eye.³¹ The primary goal of this Allergan-sponsored study was to improve subjective symptoms on the Dry Eye Disease Comfort Assessment questionnaire score over a 16-day period. No statistical analysis was performed on these data, but changes between the two groups appeared similar with a mean decrease on the questionnaire score of 1.41 with Refresh Optive^® and 1.47 with Blink^® Tears.

In an Alcon-sponsored clinical trial, Systane^® Ultra, Refresh Optive^®, GenTeal Moderate^® (Novartis Pharmaceuticals, East Hanover, NJ, USA), and Blink^® Tears were compared with regard to postapplication comfort.⁴⁶ Drop comfort grading was measured using a 0–9 scale, with 0 representing the highest level of comfort. The design was a randomized, double-masked, cross-over study with 20 patient participants. After drop administration, the comfort scores were 0.7±1.26 for Systane^® Ultra, 1.05±1.10 for Refresh Optive^®, 1.84±2.19 for Blink^® Tears, and 1.1±1.21 for GenTeal^®. No statistical analysis was performed.

Soothe versus Refresh or Systane

Soothe^® (Bausch & Lomb Incorporated, Bridgewater, NJ, USA) is a lipid-based artificial tear. A pair of separate studies aimed to elucidate its effects on the LLT within the tear film. The first study evaluated the effectiveness of Soothe^® versus Refresh Optive^®. Enrollment included 41 patients with a LLT under 70 nm in both eyes and baseline visual acuity greater than 20/70. Each patient received a drop of Soothe^® in one eye and Refresh Optive^® in the contralateral eye. A custom-designed lipid layer interferometer quantified the LLT at 1, 5, and 15 minutes after application. Both artificial tears increased the LLT from a baseline of 61.5±1.6 nm (P<0.001), with a mean LLT of 83.2±3.6 nm for Refresh Optive^® and 121.5±3.8 nm for Soothe^® (P<0.001).⁴⁷

A second study utilized the same inclusion criteria and study design but compared Soothe^® to Systane^®.48 A total of 40 patients were included in this study. Results showed a mean LLT of 124.4±4.9 nm for Soothe^® and a mean LLT of 71.3±2.6 nm for Systane^®. This represents an increase from baseline of 107% for Soothe and only 16% for Systane. Both studies were industry sponsored by Ocular Research (Boston, MA, USA) and Alimera Sciences (Alpharetta, GA, USA), which have partial ownership of Soothe.

Other comparative studies

Another randomized controlled study analyzing lipid-based tear substitutes looked at their effect on TBUT, Schirmer’s test, tear meniscus, and subjective symptoms.⁴⁹ This crossover study compared Tears Again^® (OcuSoft, Rosenberg, TX, USA) to Liposic^® (Bausch & Lomb) in 74 patients with a history of dry eye over two separate 6-week periods. In both the initial treatment and after cross-over, Tears Again^® improved all the aforementioned subjective and objective endpoints (P<0.05); 62.5% of patients preferred Tears Again^®, 25% preferred Liposic^®, and 12.5% found the preparations to be equal.

GenTeal Tears^® with preservative, GenAqua^®, and preservative-free Tears Naturale^® (Alcon Laboratories, Inc.) were compared in an open-label, two-treatment, two-period study.⁵⁰ After 4 weeks of treatment, patients were evaluated with TBUT, Schirmer’s test, and corneal staining as well as via a symptom questionnaire. A total of 37 patients completed the study. Both TBUT and Schirmer testing improved in the GenTeal group but not in the Tears Naturale^® group (P=0.27). Both artificial tears were rated as excellent for tolerability and convenience. Subjective symptoms were not different between the two treatments.

Clinical improvement

Table 2 summarizes the clinical improvement of dry eye subjects based on objective and subjective criteria used on each study included in this review. Due to the heterogeneity of criteria used on these studies, a percent of improvement was calculated for each of the artificial tears evaluated. The percent of improvement was calculated based on the amount of subjects with symptomatic and clinical improvement with respect to the total subjects treated.

Recommendations

Our recommendations for suggested drop brands in each of the respective categories is based upon both the number of studies completed on these brands as well as the trials that compared their efficacy against other brand name artificial tears (see Tables 1 and 2). Due to the heterogeneous data available and lack of standardization between studies, the recommendations made are not intended to be definite and should be individualized based on disease severity and patient’s expectations. This is a step approach, initiating therapy with the most studied artificial tears. Treatment recommendations (summarized in Figure 1) are as follows:

Frequency/duration

Based on the duration and frequency of artificial tear use reported in all the studies referenced in Tables 1 and 2, a mean of 3.47 doses/day over a period 60.1 days was established. Thus, our recommendation would be to use the artificial tears three to four times per day over a period of 2 months before transitioning to the next step. If artificial tear use extends beyond four to six times per day, then a preservative-free formulation should be used.⁶²

Preserved versus preservative-free

As seen in Tables S1–S5, many different tear replacement formulations include preservatives, but not all preservatives have the same effects. The most commonly used preservative was benzalkonium chloride (BAK). BAK has more antimicrobial activity than any other preservative in both animal and human subjects. However, BAK has been shown to damage the corneal epithelium and disrupt the tear film immediately after administration, which directly contradicts the goal of artificial tears.^63,64 Some of the newer preservative compounds appear to have a better safety profile than BAK but do not entirely prevent corneal epithelium damage.⁶⁵ These preservatives include Purite^® (Bio-Cide International Inc., Norman, OK, USA), Polyquad^® (Alcon Laboratories, Inc.), GenAqua^® (Novartis Ophthalmics, East Hanover, NJ, USA), OcuPure^® (Abbott Laboratories Inc.), Dissipate^® (OCuSOFT, Rosenberg, TX, USA).^50,63,66 Since most of the corneal changes occur when the preservatives reach high concentrations, if one chooses to use a tear replacement with preservatives, daily use should be limited to four to six times.⁶² Due to the risk of contamination, if a preservative-free artificial teardrop is chosen, single dose vials are recommended over multidose administration bottles.⁶⁷ Increased cost represents another downside to preservative-free artificial tears compared to those with preservatives. All things considered, a choice between preservative-free and preserved artificial tears should be discussed on an individual basis between physician and patient.

Limitations

The US is currently the epicenter of clinical research on DTS as it has been previously documented to conduct 70.8% of the registered clinical trials around the world which focused on dry eye.²³ The data presented in this paper may not be applicable in a global setting, as the American diet, and cultural and daily activities may play a role in the development of DTS and its corresponding treatment. With growing research in other countries, such as Japan, Australia, UK, and the Netherlands, treatment of DTS with artificial tears will progressively be refined on a global scale.⁶⁸ Furthermore, the research that has been presented in this article is based on a nonregulated health product, and the US permits distribution of artificial tears without any data revealing positive efficacy.^68,37 Any research investments made into this field may be affected by financial interests of the investigators and call into question the integrity of the outcomes, which may have influenced our recommendations. A recent meta-analysis of the comprehensive research on dry eye disease revealed that the pharmaceutical industry sponsored 78% of 185 clinical trials on this topic in the US.⁶⁸ For this reason, care should be taken when analyzing and applying data from industry-funded studies. Independent, unbiased research must increase on a worldwide level in order to more objectively and accurately elucidate the management of DTS.

In addition, this review article addressed a myriad of subjective and objective data with a multitude of data collection methods and varying result parameters, including dosing and length of treatment phase. For these reasons, further statistical analyses of improvements were not completed in this paper, but our study does suffer from lack of statistical backing. Furthermore, many of the individual papers which contributed to our information pool validated their data prior to their release. In future clinical trials, standardization of measurement instruments as well as dosage and time of treatment could enrich the generalizability of the outcomes and lead to a more consistent review of resultant data.

Finally, our study provides recommendations based on comparative studies previously published in the literature. Most of the artificial tears available in the market, as seen in Tables 2 and S1–S5, were not included in the recommendations since they have not been included in comparative studies. We recognize that the recommendations made are still a good guide for clinicians and patients when initiating DTS therapy based on the data available.