Comparing the efficacy of sodium hyaluronate eye drops and carboxymethylcellulose eye drops in treating mild to moderate dry eye disease

Abstract

Purpose:

Aim of this study was to compare the efficacy of sodium hyaluronate (SH) and carboxymethyl cellulose (CMC) eye drops in treating mild to moderate dry eye disease in terms of relief of symptoms, mean change in tear film breakup time, Schirmer’s test, and impression cytology of conjunctiva from baseline.

Methods:

An observational study was carried out during a 2-year period in our tertiary referral hospital. The study consisted of 60 patients randomly allotted to two groups to receive SH and CMC eye drops for an 8-week period. Ocular surface disease index, tear film breakup time, and Schirmer’s test were performed at baseline visit and 4 and 8 weeks of treatment, and impression cytology of conjunctiva was performed at baseline and 8 weeks.

Results:

Significant improvement in patient symptoms, tear film breakup time, and Schirmer’s test from baseline was seen in both SH and CMC groups at 8 weeks posttreatment, whereas impression cytology of conjunctiva in both groups did not show significant improvement at 8 weeks of treatment. Data analysis using unpaired t-test showed comparable results.

Conclusion:

Both CMC and SH demonstrated equal efficacy in treating mild to moderate dry eye disease.

Dry eye is a common chronic condition that causes ocular discomfort, pain, and limited vision and impacts the quality of life.[1] The prevalence ranges from 5% to 50%[2] worldwide and is slightly higher in the Indian scenario at 18.4%–54.3%.[3] Dry eye is a disorder of the ocular surface with a multifactorial etiology characterized by lack of tear film homeostasis resulting in tear film instability and hyperosmolarity, which further results in ocular surface inflammation and damage. Subjective symptoms include ocular discomfort, grittiness, burning, and visual disturbances. The objective symptoms are ocular surface inflammation along with tear film instability.[4] Artificial tears form the main treatment, with the main tear formulations being carboxymethyl cellulose (CMC) and sodium hyaluronate (SH).[5] CMC is an anionic polymer of glucopyranose subunits which helps in water retention and has moistening properties.[6,7]

Hyaluronic acid is a glycosaminoglycan found in vitreous and tears and has the ability to tether water molecules and avoid tear film desiccation. Several authors have compared CMC with hyaluronate based on quantifiable parameters, which include tear film breakup time (TBUT), Schirmer’s test, corneal and conjunctival staining, and dry eye symptom score. In the available literature, some authors have demonstrated comparable efficacy between the two, while others claim one to be more efficient than the other. This study aims at determining which is more efficient in our Indian scenario and describes its implications in forming an effective treatment protocol.

Methods

Clearance from the ethical and scientific committee was obtained before sample collection.

This observational study was carried out during a 2-year period from October 2019 to June 2021 in our tertiary referral hospital and comprised 60 patients randomly allotted to receive 0.1% SH or 0.5% CMC eye drops three to four times a day for an 8-week period. TBUT, Schirmer’s test without anesthesia, and dry eye symptom score found using the ocular surface disease index (OSDI) were assessed at baseline and 4 and 8 weeks of treatment, and impression cytology (IC) was performed at baseline and 8 weeks of treatment. Patients with symptomatic dry eye disease with an OSDI score >13 were included in the study. Patients were categorized into evaporative (computer vision syndrome), aqueous deficient (elderly, diabetes mellitus (DM)), and mixed based on TBUT and Schirmer values: TBUT <10 s into evaporative group and Schirmer value <10 mm into aqueous deficient group. The exclusion criteria were patients who had undergone refractive surgery, patients with active ocular infection, contact lens wearers, and those who have used any ocular medication within 14 days of study enrollment. TBUT was performed using a fluorescein-impregnated strip, and the dye distributed by blinking was then observed under cobalt blue filter. The time interval between the last blink and the development of first dry spot was taken. Normal value being more than 10 s, anything less was considered as dry eye. For Schirmer’s test, Whatman filter paper was kept in the lateral one-third of lower eyelid after folding 5 mm from one end and the length of the moistened portion was measured after 5 min; this was done without using topical anesthesia. Value less than 10 mm was taken as abnormal. IC of conjunctiva was performed by applying cellulose acetate filter paper on the temporal bulbar conjunctiva under topical anesthesia. The specimen was then transferred to a coplin jar containing 95% ethyl alcohol and then stained with hematoxylin and eosin stain. Then, grading was done as per Nelson’s classification.

Data analysis

Analysis was done using IBM Statistical Package for the Social Sciences (SPSS) 25. Age, TBUT, Schirmer’s test, and dry eye symptom score (OSDI) between the two groups were compared using unpaired t-test. Gender, dry eye severity grading (DRSG), and etiological classification (EC) distribution between the two groups were compared using Chi-square test. IC comparison was done using Mann–Whitney U test. Two-way repeated measures analysis of variance (ANOVA) was used to assess the effect of time and group on the outcome variables. A P value of <0.05 was considered significant for all the analyses.

Results

The study enrolled 60 patients with thirty each in SH and CMC groups, which included 27 males and 33 females in total (CMC group had 15 males and females, while the SH group had 12 males and 18 females). The participants had a mean age of 42.63 years in the CMC group and 45.67 years in the SH group [Table 1]. A TBUT value of less than 10 s was taken as evaporative dry eye. Schirmer’s test without anesthesia, which reflects the basal and reflex tear secretion, with values less than 10 mm was considered as aqueous deficient dry eye. Altered TBUT and Schirmer’s values were considered as combined dry eye disease. In keeping with this, among our 60 patients, 25 had evaporative dry eye disease, 13 were aqueous deficient, and 22 had combined disease. The CMC group had 15 evaporative, seven aqueous deficient, and eight with combined dry eye disease patients. The SH group included 10 evaporative, six aqueous deficient, and 14 combined dry eye disease patients.

Table 1.

Comparison of age, TBUT, SH, and OSDSI between the two groups (group 1: CMC, group 2: SH)

Variable	Groups	n	Mean	Std. deviation	t statistic (P)
Age	CMC	30	42.63	18.03	−0.666 (0.508)
	SH	30	45.67	17.21
OSDSI-BL	CMC	30	15.93	4.884	−0.985 (0.329)
	SH	30	17.10	4.270
OSDSI 4W	CMC	30	13.07	4.323	0.498 (0.620)
	SH	30	12.50	4.486
OSDSI 8W	CMC	30	12.33	4.205	1.341 (0.185)
	SH	30	10.83	4.457
TBUT-BL	CMC	30	7.583	2.5260	0.743 (0.460)
	SH	30	7.133	2.1493
TBUT 4W	CMC	30	9.000	2.8977	0.236 (0.815)
	SH	30	8.833	2.5742
TBUT 8W	CMC	30	9.467	2.7384	−1.951 (0.056)
	SH	30	11.033	3.4415
Schirmer score-BL	CMC	30	11.800	6.8224	−0.521 (0.604)
	SH	30	12.867	8.9036
Schirmer score 4W	CMC	30	12.533	7.1931	−1.041 (0.302)
	SH	30	14.750	9.1856
Schirmer score 8W	CMC	30	13.533	7.8267	−1.112 (0.271)
	SH	30	15.967	9.0762

4W=4 weeks, 8W=8 weeks, BL=baseline, CMC=carboxymethyl cellulose, SH=sodium hyaluronate, TBUT=tear film breakup time

Comparison of OSDI values from baseline to 4 and 8 weeks between the two groups (group 1 = CMC, group 2 = SH) showed a significant effect of both time (F[1.51, 57] = 174.01, P = 0.000, h² = 0.75) and the interaction of group and time (F[1.51, 57] = 12.03, P = 0.000, h² = 0.17) on the ocular surface disease score index (OSDSI) scores, with a greater decrease in the SH group from baseline to 4 and 8 weeks [Table 2]. Comparison of TBUT values from baseline to 4–8 weeks between the two groups showed a significant effect of both time (F[1.73, 57] = 68.48, P = 0.000, η2 = 0.54) and the interaction of group and time (F[1.73, 57] = 9.74, P = 0.000, h² = 0.14) on the TBUT scores, with a greater increase in the SH group from 4 to 8 weeks [Table 3]. Comparing Schirmer’s score from baseline to 4–8 weeks between the two groups showed a significant effect of only time (F[1.82, 57] = 38.35, P = 0.000, h² = 0.39), but not the interaction of group and time (F[1.82, 57] = 3.53, P = 0.06, h² = 0.06) on the Schirmer scores [Table 4]. Comparison of IC score from baseline to 8 weeks between the two groups showed no significant effect of either time (F^{[1, 58]} = 3.15, P = 0.08, h² = 0.05) or the interaction of group and time (F^{[1, 58]} = 1.90, P = 0.17, h² = 0.03) on the IC scores [Table 5].

Table 2.

Two-way repeated measures ANOVA for comparing the change in OSDSI values from baseline-4W-8W between the two groups

	Group (n=30 each)	Mean	Std. deviation	F and P for time	F and P for time×group interaction
OSDSI-BL	CMC	15.93	4.884	174.01 (0.000)	12.03 (0.000)
	SH	17.10	4.270
OSDSI-4W	CMC	13.07	4.323
	SH	12.50	4.486
OSDSI-8W	CMC	12.33	4.205
	SH	10.83	4.457

4W=4 weeks, 8W=8 weeks, ANOVA=analysis of variance, BL=baseline, CMC=carboxymethyl cellulose, SH=sodium hyaluronate

Table 3.

Two-way repeated measures ANOVA for comparing the change in TBUT values from baseline-4W-8W between the two groups

	Group (n=30 each)	Mean	Std. deviation	F and P for time	F and P for time×group interaction
TBUT-BL	CMC	7.583	2.5260	68.48 (0.000)	9.74 (0.000)
	SH	7.133	2.1493
TBUT-4W	CMC	9.000	2.8977
	SH	8.833	2.5742
TBUT-8W	CMC	9.467	2.7384
	SH	11.033	3.4415

4W=4 weeks, 8W=8 weeks, ANOVA=analysis of variance, BL=baseline, CMC=carboxymethyl cellulose, SH=sodium hyaluronate, TBUT=tear film breakup time

Table 4.

Two-way repeated measures ANOVA for comparing the change in Schirmer scores from baseline-4W-8W between the two groups

	Group (n=30 each)	Mean	Std. deviation	F and P for time	F and P for time × group interaction
Schirmer-BL	CMC	11.800	6.8224	29.65 (0.000)	3.09 (0.06)
	SH	12.867	8.9036
Schirmer-4W	CMC	12.533	7.1931
	SH	14.750	9.1856
Schirmer-8W	CMC	13.533	7.8267
	SH	15.967	9.0762

4W=4 weeks, 8W=8 weeks, ANOVA=analysis of variance, BL=baseline, CMC=carboxymethyl cellulose, SH=sodium hyaluronate

Table 5.

Two-way repeated measures ANOVA for comparing the change in IC scores from baseline-4W-8W between the two groups

	Group (n=30 each)	Mean	Std. deviation	F and P for time	F and P for time × group interaction
IC-BL	CMC	1.70	0.915	3.15 (0.08)	1.90 (0.17)
	SH	2.00	0.830
IC8W	CMC	1.67	0.994
	SH	1.73	0.868

4W=4 weeks, 8W=8 weeks, ANOVA=analysis of variance, BL=baseline, CMC=carboxymethyl cellulose, IC=impression cytology, SH=sodium hyaluronate

Discussion

The OSDI score at baseline was 15.93 for the CMC group and 17.10 for the SH group. This score reduced to 13.07 in the CMC group and 12.50 in the SH group at 4 weeks of treatment and further reduced to 12.33 in the CMC group and 10.83 in the SH group at 8 weeks of initiation of treatment [Table 2]. The OSDI score showed significant decrease in both groups from baseline to 8 weeks, with intergroup comparison showing a greater decrease in the SH group from baseline to 4 weeks and to 8 weeks [Table 2]. This was consistent with the studies by Brignole et al.[8] and Groß et al.,[9] which demonstrated the superiority of SH over non-hyaluronate preparations in alleviating dry eye symptoms. A total of 15 studies have compared symptoms as a parameter to compare the effectiveness of SH over non-hyaluronate preparations (majority of the non-hyaluronate formulations were cellulose based).[10] Among these, 10 studies showed SH to be superior, with six of them being statistically significant. The study by Brignole et al.[8] compared 0.18% hyaluronic acid (sodium hyaluronate) with 1% CMC in moderate dry eye disease and showed the SH group to be better at the 8-week period. Similar results were also obtained by Dorothea et al.[9] with 0.1% hyaluronate and 0.5% CMC. The study by Baudouin et al.[11] showed osmoprotective CMC to be superior to SH, which was attributed to the presence of osmoprotective additives like erythritol, l-carnitine, and glycerine, which are internalized by cells, giving them a longer duration on the ocular surface beyond the typical artificial tear preparations. The TBUT at baseline was 7.83 s and that of SH was 7.133 s [Table 3]. The mean value improved to 9 s in the CMC group and 8.8 s in the SH group at 4 weeks. These values further improved to 9.46 s in the CMC group and 11 s in the SH group at 8 weeks. There was significant equal improvement in both the groups at 4 weeks. From 4 to 8 weeks, the SH group demonstrated slightly higher increase in TBUT values than the CMC group. This finding is contradictory to the meta-analysis study by Song et al.,[12] which analyzed five studies where CMC and hyaluronate were compared in dry eye disease. The comparison was done based on TBUT. The results of this meta-analysis were in favor of CMC which showed a better trend in TBUT values, but the results obtained were not statistically significant. One more limitation of this meta-analysis study was that it included studies with a smaller sample size. Another meta-analysis by Yang et al.[13] comprising 17 studies (a total of 707 subjects in the hyaluronate group and 693 in the non-hyaluronate group) observed similar improvement in TBUT values between the two groups. Though literature stresses on the anionic nature of CMC and its effectiveness on tear retention time, the contradictory result in our study can be because of the greater epithelial repair action and greater viscosity of SH, as noted by Brignole et al.[8] They also theorized that not just the active ingredient, but other added elements can also influence the outcome. In our case, the commonly used SH preparation contained glycerine as a buffered vehicle, which could have affected the results due to its positive reinforcing effects in restoring cellular water.The mean baseline Schirmer value in the CMC group was 11.8 mm, while that in the SH group was 12.8 mm [Table 4]. At 4 weeks of treatment, this improved to 12.53 mm in the CMC group and 14.7 mm in the SH group. These values subsequently improved to 13.5 mm in the CMC group and 15.9 mm in the SH group at 8 weeks of treatment [Table 4]. There was significant improvement in Schirmer values from baseline to 8 weeks in both groups. But it should be noted that, significant increase in values from baseline to 4 weeks was seen only in the SH group and not in the CMC group. This is in accordance with a meta-analysis by Bryan Chin Hou Ang et al.[10] that included 18 studies, of which 11 studies used Schirmer1 (without anesthesia) as a parameter to compare SH and non-hyaluronate preparations in dry eye syndrome. Schirmer’s test without anesthesia is performed to assess both reflex and basal tear secretion. This test has a number of limitations such as low replicability, lack of standardized evaluation of wetted length, and uncertainty regarding the quantity of fluid absorbed by the strip with its wetted length. The study by Tsubota K et al. observed that Schirmer’s test lacks sensitivity in mild to moderate dry eye disease and is reliable only in severe dry eye.[14] They also stated that Schirmer’s values can even improve with placebo preparations as they increase the tear volume, reduce the osmotic pressure, and also cause dilution of proinflammatory markers. The IC in both groups did not show any significant improvement in Nelson’s grade from baseline to 8 weeks [Table 5]. Five studies had been previously done with conjunctival IC, of which two studies utilized flow cytometry and three utilized morphology of goblet cells. The two studies using flow cytometry by Sanchez et al.[15] and Brignole et al.[8] observed underexpression of human leukocyte antigen D related ( HLA-DR) in both hyaluronic and non-hyaluronic groups. The study by Brignole et al.[8] showed greater decrease in cluster of differentiation (CD) 44 in the hyaluronic group compared to the CMC group. This effect is due to the immunomodulator function of SH, which selectively targets the CD44 and the receptor for hyaluronan-mediated motility (RHAMM). The three studies based on morphology of goblet cells and epithelial cells were those of Aragona et al.,[16] Iester et al.,[17] and Nelson.[18] The study by Aragona et al.[16] with SH in 44 patients demonstrated significant reduction in the ocular surface damage over a 3-month period.[16] The one by Iester et al.[17] comparing hyaluronate with hydroxymethyl cellulose in 135 patients showed significant reduction in cytology score after completion of 90 days of treatment. While the study by Nelson[18] showed no improvement in cytology after 8 weeks of treatment with SH. A study by RB Grene et al.[19] on CMC in dry eye patients based on IC of conjunctiva mapping technique over a 2-month period showed significant improvement in ocular surface histologic findings after completion of 2 months of treatment.[19] Our study corroborates with that of Nelson et al.[18] at the end of 8 weeks, indicating that significant change in this parameter would take more than 8 weeks due to longer cellular turnover. Our study showed that both CMC and SH significantly improved dry eye symptomatology and TBUT, with slightly better performance observed for SH. This is in keeping with the existing literature. Whereas in case of Schirmer’s test, rapid improvement in scores was seen in the SH group, but the results were comparable at 8 weeks. This is also in keeping with the existing literature. No significant difference was obtained in conjunctival IC, which can be due to the longer cellular turnover time needed for the histologic changes to manifest. No adverse effects were reported in both groups, and patients were well compliant to use topical medications. One of the drawbacks of our study was its short duration of 8 weeks, which was inadequate time to assess the changes in conjunctival cytology. The other limitation was that the commercially available SH used in the study had in addition to 0.1% SH, glycerin as an aqueous buffer, which might have modified the results. Another drawback in using Schirmer’s test and TBUT was the lack of standardization with low reproducibility.

Conclusion

SH and CMC are equally effective in treating mild to moderate dry eye disease, with greater improvement observed in terms of subjective symptoms for SH. Both SH and CMC improved the objective parameters for dry eye disease over the 8-week treatment course, with better outcome obtained for SH.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.