BASIC INVESTIGATION
Characterization of Dry Eye Disease in Diabetic Patients Versus Nondiabetic Patients Kenneth A. Beckman, MD
Purpose: The purpose of this investigation was to assess the presence and relative severity of dry eye evaluated by a panel of diagnostic methods in diabetic patients as compared with that in nondiabetic patients.
Methods: Patients ($40 years of age) scheduled for a routine eye examination at the clinical site were recruited for the study. Exclusion criteria included current use of topical medication for glaucoma or prior ocular surgery if the patient was within the postoperative recovery period (3 months). Study endpoints included tear film break-up time, tear film osmolarity, corneal fluorescein and conjunctival lissamine green staining, Schirmer strip testing, Ocular Surface Disease Index questionnaire, and Dry Eye International Task Force severity ranking. Results: Sixty-three patients were enrolled in this study, 38 in the diabetic group and 25 in the nondiabetic group. All enrolled patients scored at least a 1 on the International Task Force ranking scale. A significantly higher mean tear film osmolarity was observed in the nondiabetic patient group, 312 versus 303 mOsm/L (P = 0.02). The mean conjunctival staining scores were significantly higher in the diabetic patient group, 2.72 versus 2.11 (P = 0.034). No statistically significant differences were observed between patient groups for corneal staining, tear film break-up time, Schirmer strip, or Ocular Surface Disease Index scores.
Conclusions: The overall presence and severity of dry eye was found to be similar in the diabetic and nondiabetic patient groups. However, significant differences were observed between groups with regard to individual diagnostic assessments (lissamine green staining and tear film osmolarity). Key Words: dry eye, diabetes, prevalence, osmolarity (Cornea 2014;33:851–854)
Received for publication January 7, 2014; revision received March 27, 2014; accepted April 22, 2014. Published online ahead of print June 9, 2014. From the Comprehensive Eye Care of Central Ohio, Westerville, OH; and The Ohio State University Department of Ophthalmology, Columbus, OH. K. A. Beckman, Consultant to Allergan, Bausch & Lomb Incorporated, and TearLab. Supported by Allergan via an independent and unrestricted research grant. Allergan had the opportunity to review the final version of the paper to address any factual inaccuracies or request the redaction of information deemed to be proprietary or confidential and ensure that study support was disclosed. Reprints: Kenneth A. Beckman, MD, Comprehensive Eye Care of Central Ohio, 450 Alkyre Run Drive, #100, Westerville, OH 43082 (e-mail: [email protected]). Copyright © 2014 by Lippincott Williams & Wilkins
Cornea Volume 33, Number 8, August 2014
K
eratoconjunctivitis sicca, or dry eye, is a common condition of the ocular surface characterized by an insufficient or abnormal tear film affecting the conjunctiva and cornea. Patients with dry eye often experience discomfort and/or irritation of the ocular surface, photophobia, and blurred or fluctuating vision. Although dry eye may result from a number of contributing etiologies, the condition is generally considered to have an inflammatory component and results in hyperosmolarity in the tear film. Several classification systems have been developed in an effort to categorize subtypes of dry eye and rank the severity of the disease.1–3 Estimates of the prevalence of dry eye in the US population ranges from approximately 6 to 43 million people, with a wide range of potential risk factors that may contribute to the development of the condition, including age, gender, diet, diabetes mellitus, and environmental conditions.4 Diabetes belongs to a group of diseases that result in high levels of blood glucose. Approximately 25 million people in the US population have diabetes.5 Ocular disorders are common in diabetic patients, particularly microvascular complications such as retinopathy. Based on the potential ocular comorbidities present in diabetic patients, ophthalmologists, in general, and particularly cornea specialists have an opportunity to screen and detect early manifestations of ocular surface disease secondary to diabetes. A wide array of diagnostic methods is available for the assessment of dry eye. A thorough medical and ocular history, patient questionnaires, vital staining to assess the integrity of the ocular surface, and Schirmer testing to evaluate the tear film are among the options that are often used to aid in the diagnosis of dry eye.3 More recently, tear film osmolarity has been included in the battery of testing to evaluate the relative health of the ocular surface.6 A recent clinical study conducted in Turkey reported the association of a high tear film osmolarity in diabetic patients as compared with that in nondiabetic patients.7 The purpose of this investigation was to assess the presence and relative severity of dry eye in diabetic patients as compared with that in nondiabetic patients at a cornea/ocular surface specialist clinical site. In addition to a comparative evaluation of tear film osmolarity, as conducted in a prior clinical study,7 a panel of additional diagnostic methods, including fluorescein/lissamine green staining, tear film break-up time (TBUT), Schirmer testing, and patient questionnaires, was used to assess the dry eye status of patients enrolled in the study. www.corneajrnl.com |
851
Cornea Volume 33, Number 8, August 2014
Beckman
MATERIALS AND METHODS
RESULTS
This study was conducted at single-center, cornea/ ocular surface specialist clinical site. All the visits associated with the study occurred in a clinical setting, and all the study procedures were conducted by certified investigators and technicians. All study documents, including the protocol, informed consent form, and recruiting materials were approved by the Mount Carmel Institutional Review Board (Mount Carmel IRB, Columbus, OH, IRB # 1110004-1). This study was conducted in accordance with current Good Clinical Practice guidelines and the Declaration of Helsinki. The study was registered at ClinicalTrials.gov before the initiation under the registry number NCT01450787. Patients who participated in this study were recruited from a pool of patients who were scheduled for a routine eye examination at the clinical site. All patients were required to be at least 40 years of age. Patients were considered ineligible for participation in the study if at the time of screening they were currently using topical medication for glaucoma. Additionally, patients were excluded from the study if they had undergone an ocular surgery and were currently within the postoperative recovery period (3 months). All patients were required to read and sign an IRBapproved informed consent form before any study-related procedures were conducted. A single study visit was required for each patient to complete the study. The screening and patient evaluation procedures were conducted at the single visit. The following assessments were performed during the study visit: TBUT, tear film osmolarity (TearLab Osmolarity System, San Diego, CA), corneal fluorescein (0–5 grading scale), and conjunctival lissamine green (0–5 grading scale) staining using the Oxford Grading Scheme,8 and Schirmer strip testing.9 All patients participating in the study also completed the Ocular Surface Disease Index (OSDI) questionnaire at the study visit.10 Subsequent to the study visit assessments, a Dry Eye severity scale ranking was assigned to each patient based on the grading scale (0–4 scale) from the Dry Eye International Task Force (ITF) (2007).2 All patients were queried with regard to their current use of artificial tears and/or the use of any dry eye medications and/or dietary supplements. For patients in the diabetes group, the number of years since a diagnosis of diabetes was confirmed was recorded. Any adverse events that occurred during the study visit were also recorded. Descriptive statistics were prepared for the patient demographic information (diabetic vs. nondiabetic, gender), and the clinical assessments (mean 6 SD or mean 6 standard error) were evaluated in the study: TBUT, tear film osmolarity, corneal staining, conjunctival staining, Schirmer strip scores, OSDI scores, and ITF ranking. Subgroup analyses were conducted for dry eye diagnostic parameters within and across patient groups (by gender and according to a response regarding the current use of at least occasional dry eye therapy artificial tear, dry eye medications, or dietary supplements) to evaluate potential trends for the patient population. Statistical testing was conducted via a Student t test. A P value #0.05 was the threshold for determining statistical significance.
Sixty-three patients were enrolled in this study, 38 in the diabetic group and 25 in the nondiabetic group. All enrolled patients met the inclusion criteria after screening and completed the assessments required at the single study visit. No ocular or nonocular adverse events were recorded during the study. The demographic information collected from the patients is listed in Table 1. The mean duration of diabetes (years since diagnosis) for patients in this study was 11.1 6 9.0 (SD) years (range 1–41 years). A higher percentage of nondiabetic patients reported at least the occasional use of dry eye therapy when queried compared with what the diabetic patients reported (Table 1). Seventy-two percent of nondiabetic patients (18 out of 25) reported the consistent use of artificial tears, dry eye medications, or dietary supplements, as compared with 50% of diabetic patients (19 out of 38). The mean tear film osmolarity (all eyes) and the mean osmolarity difference between a patient’s eyes are presented by patient group in Table 2 (all subjects). The mean tear film osmolarity of nondiabetic patients was significantly higher than the osmolarity observed in the diabetic patient group (P = 0.02). Additionally, the mean difference in the osmolarity between eyes was significantly higher in the nondiabetic group as compared with that in the diabetic group (P = 0.017). A similar profile was observed when calculating mean osmolarity by patient group based only on the eye with the higher tear film osmolarity score for each patient; using this method, mean osmolarity scores were 322.00 versus 309.22 mOsm/L for the nondiabetic and diabetic patient groups, respectively. Mean corneal fluorescein staining was generally low across the study population (all subjects); no statistically significant differences were observed between patient groups. Mean conjunctival staining with lissamine green was more pronounced in both patient groups. Significantly higher mean conjunctival staining scores were observed in the diabetic patient group as compared with in the nondiabetic group (P = 0.034). The mean TBUT was similar for both diabetic and nondiabetic patient groups, and was
852
| www.corneajrnl.com
TABLE 1. Demographics Listed by Patient Group
Gender, n (%) Male Female Subjects reporting at least occasional use of dry eye therapy*, n (%) Male Female Subjects reporting NO use of dry eye therapy*, n (%) Male Female
Diabetic Patients (n = 38)
Nondiabetic Patients (n = 25)
All Patients (n = 63)
16 (42.1) 22 (57.9)
4 (16.0) 21 (84.0)
20 (31.7) 43 (68.3)
5 (13.1) 14 (36.8)
1 (4.0) 17 (68.0)
6 (9.5) 31 (49.2)
11 (28.9) 8 (21.1)
3 (12.0) 4 (16.0)
14 (22.2) 12 (19.0)
*Artificial tears, dietary supplements, and/or dry eye medications.
Ó 2014 Lippincott Williams & Wilkins
Cornea Volume 33, Number 8, August 2014
Characterization of Dry Eye in Diabetic Patients
TABLE 2. Dry Eye Assessments Listed by Patient Group (All Subjects) Diabetic Patients (n = 38) Osmolarity Osmolarity difference Fluorescein corneal staining Lissamine green conjunctival staining TBUT time, s Schirmer test score, mm OSDI test score ITF score
Nondiabetic Patients (n = 25)
TABLE 3. Dry Eye Assessments Listed by Patient Group (Subjects Reporting at Least Occasional Use of Dry Eye Therapy) Diabetic Patients (n = 19)
P
303.4 11.61 0.31 2.72
(2.03) (1.71) (0.08) (0.16)
312.1 19.88 0.31 2.11
(3.05) (2.89) (0.13) (0.19)
0.020 0.017 0.948 0.034
3.38 18.36 19.58 1.32
(0.17) (1.06) (2.62) (0.09)
3.66 15.88 27.50 1.24
(0.26) (1.34) (4.60) (0.12)
0.374 0.149 0.140 0.607
Osmolarity Osmolarity difference Fluorescein corneal staining Lissamine green conjunctival staining TBUT time, s Schirmer test score, mm OSDI test score ITF score
Nondiabetic Patients (n = 18)
P
301.6 11.1 0.24 3.0
(2.6) (2.0) (0.1) (0.2)
315.1 19.2 0.42 2.2
(3.8) (3.7) (0.2) (0.2)
0.004 0.06 0.38 0.008
2.76 16.8 26.3 1.3
(0.2) (1.1) (3.9) (0.1)
3.6 17.7 30.8 1.3
(0.3) (1.6) (5.8) (0.2)
0.017 0.64 0.528 0.736
Data shown as mean 6 standard error of the mean. Entries in bold represent statistically significant findings.
Data shown as mean 6 standard error of the mean. Entries in bold represent statistically significant findings.
below the normal range ($10 seconds).6 Mean Schirmer strip scores were in the normal range ($10 mm)6 for both patient groups. The mean OSDI score was higher in nondiabetic patients as compared with that in diabetic patients. The mean ITF rank was similar for both patient groups. The majority of the patients in the diabetic group were classified as having “mild” dry eye, or a “1” for an ITF rank (71.1%); 26.3% of the diabetic patients were classified as having “moderate” dry eye, or “2” ranking, and 2.6% were classified as having “severe” dry eye, or a “3” ranking. In the nondiabetic patient group, 84.0% of the patients were classified as having mild (1) dry eye, 8.0% as having moderate (2) dry eye, and 8.0% as having severe (3) dry eye. No statistically significant differences were observed between the patient groups for TBUT, Schirmer strip, OSDI test scores, or ITF rank for the primary analysis of the full subject population. Secondary analyses were conducted for the subgroups of the study population based on diabetic versus nondiabetic subjects and the response to the query regarding the current use of dry eye therapy. Significant differences were observed between nondiabetic and diabetic subjects reporting at least the occasional use of dry eye therapy; mean tear film osmolarity was significantly higher in nondiabetic patients as compared with that in diabetic subjects (P = 0.004), whereas mean conjunctival staining scores were significantly higher in diabetic as compared with that in nondiabetic subjects (P = 0.008), and a significantly more rapid mean TBUT was observed for diabetic subjects as compared with that in nondiabetic subjects (P = 0.017) (Table 3). Within the subgroup of subjects that reported no current use of dry eye therapy, significant differences were also observed between diabetic and nondiabetic subjects. Mean corneal fluorescein staining scores and mean ITF ranking scores were significantly higher in diabetic subjects as compared with that in nondiabetic subjects (P = 0.003 for both tests between groups), whereas mean Schirmer test scores were significantly lower for nondiabetic subjects as compared with that in diabetic subjects (P = 0.001) (Table 4). Subgroup analyses by gender (only) revealed several trends within the study population. Female patients had
a higher mean OSDI score across the study population without regard to diabetic classification. The mean female patient OSDI score was 26.98, whereas the mean OSDI score was 13.57 for male patients. In comparing male diabetic patients to male nondiabetic patients, there were no significant differences for any categories. In contrast, nondiabetic females had a higher mean tear film osmolarity than female diabetic patients had (313.73 and 303.38 mOsm/L, respectively; P = 0.017). Additionally, female diabetic patients had a higher mean conjunctival staining score than nondiabetic patients had (2.90 and 2.11, respectively; P = 0.009).
Ó 2014 Lippincott Williams & Wilkins
DISCUSSION The purpose of this single-center clinical study was to assess the presence and relative severity of dry eye in diabetic patients as compared with that in nondiabetic patients at the time of a routine comprehensive examination. The panel of diagnostic indicators of dry eye used in this study allowed for an assessment of the disease severity level as per the ITF ranking criteria. Although the overall severity of dry eye was
TABLE 4. Dry Eye Assessments Listed by Patient Group (Subjects Reporting NO Use of Dry Eye Therapy) Diabetic Patients (n = 19) Osmolarity Osmolarity difference Fluorescein corneal staining Lissamine green conjunctival staining TBUT time, s Schirmer test score, mm OSDI test score ITF score
Nondiabetic Patients (n = 7)
P
305.5 12.2 0.39 2.4
(3.2) (2.9) (0.1) (0.2)
303.7 21.8 0.0 2.3
(4.5) (4.9) (0.0) (0.4)
0.749 0.09 0.003 0.68
4.0 20.0 26.3 1.4
(0.3) (1.8) (3.9) (0.1)
3.8 11.1 30.8 1.0
(0.6) (1.8) (5.8) (0.0)
0.736 0.001 0.528 0.003
Data shown as mean 6 standard error of the mean. Entries in bold represent statistically significant findings.
www.corneajrnl.com |
853
Beckman
found to be similar in the diabetic and nondiabetic patient groups according to the ITF scale (Table 2), there were significant differences observed between the 2 patient groups based on individual diagnostic criteria. The observation that the mean severity of lissamine green staining of the conjunctiva was found to be higher in the diabetic patient population (Table 2), and higher mean corneal fluorescein staining in diabetic patients reporting no use of dry eye therapy (in conjunction with a higher mean ITF ranking) may be in accordance with a general perception of ocular surface damage in diabetic patients. However, the finding that the mean tear film osmolarity was significantly higher in the nondiabetic patient group was unexpected based on previously reported findings.7 Corneal and conjunctival staining has been used for many years by clinicians to assess the integrity of the ocular surface. Osmolarity assessments also provide valuable insight into the quality of the tear film and can assist in a full diagnosis of a patient that may have dry eye. However, as illustrated by the results of this study, tear film osmolarity may not necessarily be expected to be higher in diabetic patients as compared with that in nondiabetic patients. Although abnormal tear film osmolarity values were observed in both patient groups, the mean osmolarity score was significantly higher in the nondiabetic patient group. The overall prevalence of dry eye and other diseases affecting the ocular surface is predictably high in the patient population visiting a cornea specialist. Although the severity of dry eye may vary widely between patients, the presence of ocular surface tissue staining (corneal or conjunctival) and rapid TBUT are particularly common diagnostic indicators for dry eye and noted in virtually all patients in this study. The ITF severity scores ranged from 1 to 3 in both patient groups in this study, with the majority of patients classified as having mild (ITF 1) dry eye as a result of the low incidence of corneal staining. Tear film osmolarity readings .308 mOsm/L and/or measurements between eyes that were $8 mOsm/L, values typically considered outside the normal range,6 were also a common finding in this study. Patients with diabetes frequently report symptoms of dry eye, such as foreign body sensation and burning, and upon evaluation are found to have an unstable tear film. In some patients, however, a reduction in corneal sensation is also associated with diabetes.11 One possibility for the lower percentage of diabetic patients reporting the use of artificial tears, dry eye medications, or dietary supplements in this study may be because of a reduced perception of dry eye symptoms. The secondary analysis of the study population indicated that the clinical assessment profile of diabetic and nondiabetic subjects that reported at least the occasional use of dry eye therapy is similar to that in the full study population (higher mean tear film osmolarity in nondiabetic subjects, and higher mean conjunctival staining in diabetic subjects), suggesting that although artificial tears/lubricants may be providing palliative relief to the patient’s dry eye symptoms, the basic clinical profile remains unchanged. This is an important consideration for the general clinical assessment of the dry eye status of diabetic and nondiabetic patients. The conduct of this study at a single clinical site and a relatively small sample size are limitations of the study.
854
| www.corneajrnl.com
Cornea Volume 33, Number 8, August 2014
A larger sample size at multiple clinical sites would be beneficial in the overall assessment of the differences in the characteristics of dry eye presented by diabetic and nondiabetic patients. However, one challenge that is frequently encountered in large, multicenter studies concerns the differences in assessment practices among clinical investigators. In that respect, the evaluation of all subjects reported in this study by a single investigator provides for uniformity across the data set. Despite the frequent association of diabetes and dry eye, the consequence of this risk factor for the ocular surface may be overlooked in clinical practice. As indicated by the differing clinical assessment profiles observed between diabetic and nondiabetic patients, it is important to evaluate diabetic patients for dry eye and other potential conditions affecting the ocular surface using a range of diagnostic techniques, particularly tear film osmolarity and vital staining. Early detection and appropriate treatment can play a vital role in the management of dry eye and other ocular inflammatory diseases.
ACKNOWLEDGMENTS The author would like to express his appreciation to Dr Ayelet Ben-David for preparing the statistical analysis. Writing and editorial assistance was provided to the authors by Kurt Brubaker of Bridge Over Brook Inc (Shady Cove, OR), and funded by Allergan Inc (Irvine, CA) at the request of the investigator. The author met the ICMJE authorship criteria. Neither honoraria nor payments were made for authorship. REFERENCES 1. Lemp MA. Report of the National Eye Institute/Industry Workshop on Clinical Trials in Dry Eyes. CLAO J. 1995;21:221–232. 2. Behrens A, Doyle JJ, Stern L, et al. Dysfunctional tear syndrome: a Delphi approach to treatment recommendations. Cornea. 2006;25:900–907. 3. Methodologies to diagnose and monitor dry eye disease: report of the Diagnostic Methodology Subcommittee of the International Dry Eye Workshop (2007). Ocul Surf. 2007;5:108–152. 4. Stern ME, Schaumburg CS, Pflugfelder SC. Dry eye as a mucosal autoimmune disease. Int Rev Immunol. 2013;32:19–41. 5. National Diabetes Fact Sheet: national estimates and general information on diabetes and prediabetes in the United States. Centers for Disease Control and Prevention. Atlanta, GA: US Department of Health and Human Services; 2011. 6. Tomlinson A, Bron AJ, Korb DR, et al. The International Workshop on meibomian gland dysfunction: report of the diagnosis subcommittee. Invest Ophthalmol Vis Sci. 2011;52:2006–2049. 7. Sagdık HM, Ugurbas SH, Can M, et al. Tear film osmolarity in patients with diabetes mellitus. Ophthalmic Res. 2013;50:1–5. 8. Bron A, Evans VE, Smith JA. Grading of corneal and conjunctival staining in the context of other dry eye tests. Cornea. 2003;22:640–650. 9. Guillon JP. Current clinical techniques to study the tear film and tear secretions. In: Korb D, Craig J, Doughty M, et al, eds. The Tear Film. London, United Kingdom: Buttersworth-Heinemann; 2002:51–83. 10. Schiffman RM, Christianson MD, Jacobsen G, et al. Reliability and validity of the Ocular Surface Disease Index. Arch Ophthalmol. 2000; 118:615–621. 11. Alves Mde C, Carvalheira JB, Módulo CM, et al. Tear film and ocular surface changes in diabetes mellitus. Arq Bras Oftalmol. 2008; 71:96–103.
Ó 2014 Lippincott Williams & Wilkins
Characterization of Dry Eye Disease in Diabetic Patients Versus Nondiabetic Patients Kenneth A. Beckman, MD
Purpose: The purpose of this investigation was to assess the presence and relative severity of dry eye evaluated by a panel of diagnostic methods in diabetic patients as compared with that in nondiabetic patients.
Methods: Patients ($40 years of age) scheduled for a routine eye examination at the clinical site were recruited for the study. Exclusion criteria included current use of topical medication for glaucoma or prior ocular surgery if the patient was within the postoperative recovery period (3 months). Study endpoints included tear film break-up time, tear film osmolarity, corneal fluorescein and conjunctival lissamine green staining, Schirmer strip testing, Ocular Surface Disease Index questionnaire, and Dry Eye International Task Force severity ranking. Results: Sixty-three patients were enrolled in this study, 38 in the diabetic group and 25 in the nondiabetic group. All enrolled patients scored at least a 1 on the International Task Force ranking scale. A significantly higher mean tear film osmolarity was observed in the nondiabetic patient group, 312 versus 303 mOsm/L (P = 0.02). The mean conjunctival staining scores were significantly higher in the diabetic patient group, 2.72 versus 2.11 (P = 0.034). No statistically significant differences were observed between patient groups for corneal staining, tear film break-up time, Schirmer strip, or Ocular Surface Disease Index scores.
Conclusions: The overall presence and severity of dry eye was found to be similar in the diabetic and nondiabetic patient groups. However, significant differences were observed between groups with regard to individual diagnostic assessments (lissamine green staining and tear film osmolarity). Key Words: dry eye, diabetes, prevalence, osmolarity (Cornea 2014;33:851–854)
Received for publication January 7, 2014; revision received March 27, 2014; accepted April 22, 2014. Published online ahead of print June 9, 2014. From the Comprehensive Eye Care of Central Ohio, Westerville, OH; and The Ohio State University Department of Ophthalmology, Columbus, OH. K. A. Beckman, Consultant to Allergan, Bausch & Lomb Incorporated, and TearLab. Supported by Allergan via an independent and unrestricted research grant. Allergan had the opportunity to review the final version of the paper to address any factual inaccuracies or request the redaction of information deemed to be proprietary or confidential and ensure that study support was disclosed. Reprints: Kenneth A. Beckman, MD, Comprehensive Eye Care of Central Ohio, 450 Alkyre Run Drive, #100, Westerville, OH 43082 (e-mail: [email protected]). Copyright © 2014 by Lippincott Williams & Wilkins
Cornea Volume 33, Number 8, August 2014
K
eratoconjunctivitis sicca, or dry eye, is a common condition of the ocular surface characterized by an insufficient or abnormal tear film affecting the conjunctiva and cornea. Patients with dry eye often experience discomfort and/or irritation of the ocular surface, photophobia, and blurred or fluctuating vision. Although dry eye may result from a number of contributing etiologies, the condition is generally considered to have an inflammatory component and results in hyperosmolarity in the tear film. Several classification systems have been developed in an effort to categorize subtypes of dry eye and rank the severity of the disease.1–3 Estimates of the prevalence of dry eye in the US population ranges from approximately 6 to 43 million people, with a wide range of potential risk factors that may contribute to the development of the condition, including age, gender, diet, diabetes mellitus, and environmental conditions.4 Diabetes belongs to a group of diseases that result in high levels of blood glucose. Approximately 25 million people in the US population have diabetes.5 Ocular disorders are common in diabetic patients, particularly microvascular complications such as retinopathy. Based on the potential ocular comorbidities present in diabetic patients, ophthalmologists, in general, and particularly cornea specialists have an opportunity to screen and detect early manifestations of ocular surface disease secondary to diabetes. A wide array of diagnostic methods is available for the assessment of dry eye. A thorough medical and ocular history, patient questionnaires, vital staining to assess the integrity of the ocular surface, and Schirmer testing to evaluate the tear film are among the options that are often used to aid in the diagnosis of dry eye.3 More recently, tear film osmolarity has been included in the battery of testing to evaluate the relative health of the ocular surface.6 A recent clinical study conducted in Turkey reported the association of a high tear film osmolarity in diabetic patients as compared with that in nondiabetic patients.7 The purpose of this investigation was to assess the presence and relative severity of dry eye in diabetic patients as compared with that in nondiabetic patients at a cornea/ocular surface specialist clinical site. In addition to a comparative evaluation of tear film osmolarity, as conducted in a prior clinical study,7 a panel of additional diagnostic methods, including fluorescein/lissamine green staining, tear film break-up time (TBUT), Schirmer testing, and patient questionnaires, was used to assess the dry eye status of patients enrolled in the study. www.corneajrnl.com |
851
Cornea Volume 33, Number 8, August 2014
Beckman
MATERIALS AND METHODS
RESULTS
This study was conducted at single-center, cornea/ ocular surface specialist clinical site. All the visits associated with the study occurred in a clinical setting, and all the study procedures were conducted by certified investigators and technicians. All study documents, including the protocol, informed consent form, and recruiting materials were approved by the Mount Carmel Institutional Review Board (Mount Carmel IRB, Columbus, OH, IRB # 1110004-1). This study was conducted in accordance with current Good Clinical Practice guidelines and the Declaration of Helsinki. The study was registered at ClinicalTrials.gov before the initiation under the registry number NCT01450787. Patients who participated in this study were recruited from a pool of patients who were scheduled for a routine eye examination at the clinical site. All patients were required to be at least 40 years of age. Patients were considered ineligible for participation in the study if at the time of screening they were currently using topical medication for glaucoma. Additionally, patients were excluded from the study if they had undergone an ocular surgery and were currently within the postoperative recovery period (3 months). All patients were required to read and sign an IRBapproved informed consent form before any study-related procedures were conducted. A single study visit was required for each patient to complete the study. The screening and patient evaluation procedures were conducted at the single visit. The following assessments were performed during the study visit: TBUT, tear film osmolarity (TearLab Osmolarity System, San Diego, CA), corneal fluorescein (0–5 grading scale), and conjunctival lissamine green (0–5 grading scale) staining using the Oxford Grading Scheme,8 and Schirmer strip testing.9 All patients participating in the study also completed the Ocular Surface Disease Index (OSDI) questionnaire at the study visit.10 Subsequent to the study visit assessments, a Dry Eye severity scale ranking was assigned to each patient based on the grading scale (0–4 scale) from the Dry Eye International Task Force (ITF) (2007).2 All patients were queried with regard to their current use of artificial tears and/or the use of any dry eye medications and/or dietary supplements. For patients in the diabetes group, the number of years since a diagnosis of diabetes was confirmed was recorded. Any adverse events that occurred during the study visit were also recorded. Descriptive statistics were prepared for the patient demographic information (diabetic vs. nondiabetic, gender), and the clinical assessments (mean 6 SD or mean 6 standard error) were evaluated in the study: TBUT, tear film osmolarity, corneal staining, conjunctival staining, Schirmer strip scores, OSDI scores, and ITF ranking. Subgroup analyses were conducted for dry eye diagnostic parameters within and across patient groups (by gender and according to a response regarding the current use of at least occasional dry eye therapy artificial tear, dry eye medications, or dietary supplements) to evaluate potential trends for the patient population. Statistical testing was conducted via a Student t test. A P value #0.05 was the threshold for determining statistical significance.
Sixty-three patients were enrolled in this study, 38 in the diabetic group and 25 in the nondiabetic group. All enrolled patients met the inclusion criteria after screening and completed the assessments required at the single study visit. No ocular or nonocular adverse events were recorded during the study. The demographic information collected from the patients is listed in Table 1. The mean duration of diabetes (years since diagnosis) for patients in this study was 11.1 6 9.0 (SD) years (range 1–41 years). A higher percentage of nondiabetic patients reported at least the occasional use of dry eye therapy when queried compared with what the diabetic patients reported (Table 1). Seventy-two percent of nondiabetic patients (18 out of 25) reported the consistent use of artificial tears, dry eye medications, or dietary supplements, as compared with 50% of diabetic patients (19 out of 38). The mean tear film osmolarity (all eyes) and the mean osmolarity difference between a patient’s eyes are presented by patient group in Table 2 (all subjects). The mean tear film osmolarity of nondiabetic patients was significantly higher than the osmolarity observed in the diabetic patient group (P = 0.02). Additionally, the mean difference in the osmolarity between eyes was significantly higher in the nondiabetic group as compared with that in the diabetic group (P = 0.017). A similar profile was observed when calculating mean osmolarity by patient group based only on the eye with the higher tear film osmolarity score for each patient; using this method, mean osmolarity scores were 322.00 versus 309.22 mOsm/L for the nondiabetic and diabetic patient groups, respectively. Mean corneal fluorescein staining was generally low across the study population (all subjects); no statistically significant differences were observed between patient groups. Mean conjunctival staining with lissamine green was more pronounced in both patient groups. Significantly higher mean conjunctival staining scores were observed in the diabetic patient group as compared with in the nondiabetic group (P = 0.034). The mean TBUT was similar for both diabetic and nondiabetic patient groups, and was
852
| www.corneajrnl.com
TABLE 1. Demographics Listed by Patient Group
Gender, n (%) Male Female Subjects reporting at least occasional use of dry eye therapy*, n (%) Male Female Subjects reporting NO use of dry eye therapy*, n (%) Male Female
Diabetic Patients (n = 38)
Nondiabetic Patients (n = 25)
All Patients (n = 63)
16 (42.1) 22 (57.9)
4 (16.0) 21 (84.0)
20 (31.7) 43 (68.3)
5 (13.1) 14 (36.8)
1 (4.0) 17 (68.0)
6 (9.5) 31 (49.2)
11 (28.9) 8 (21.1)
3 (12.0) 4 (16.0)
14 (22.2) 12 (19.0)
*Artificial tears, dietary supplements, and/or dry eye medications.
Ó 2014 Lippincott Williams & Wilkins
Cornea Volume 33, Number 8, August 2014
Characterization of Dry Eye in Diabetic Patients
TABLE 2. Dry Eye Assessments Listed by Patient Group (All Subjects) Diabetic Patients (n = 38) Osmolarity Osmolarity difference Fluorescein corneal staining Lissamine green conjunctival staining TBUT time, s Schirmer test score, mm OSDI test score ITF score
Nondiabetic Patients (n = 25)
TABLE 3. Dry Eye Assessments Listed by Patient Group (Subjects Reporting at Least Occasional Use of Dry Eye Therapy) Diabetic Patients (n = 19)
P
303.4 11.61 0.31 2.72
(2.03) (1.71) (0.08) (0.16)
312.1 19.88 0.31 2.11
(3.05) (2.89) (0.13) (0.19)
0.020 0.017 0.948 0.034
3.38 18.36 19.58 1.32
(0.17) (1.06) (2.62) (0.09)
3.66 15.88 27.50 1.24
(0.26) (1.34) (4.60) (0.12)
0.374 0.149 0.140 0.607
Osmolarity Osmolarity difference Fluorescein corneal staining Lissamine green conjunctival staining TBUT time, s Schirmer test score, mm OSDI test score ITF score
Nondiabetic Patients (n = 18)
P
301.6 11.1 0.24 3.0
(2.6) (2.0) (0.1) (0.2)
315.1 19.2 0.42 2.2
(3.8) (3.7) (0.2) (0.2)
0.004 0.06 0.38 0.008
2.76 16.8 26.3 1.3
(0.2) (1.1) (3.9) (0.1)
3.6 17.7 30.8 1.3
(0.3) (1.6) (5.8) (0.2)
0.017 0.64 0.528 0.736
Data shown as mean 6 standard error of the mean. Entries in bold represent statistically significant findings.
Data shown as mean 6 standard error of the mean. Entries in bold represent statistically significant findings.
below the normal range ($10 seconds).6 Mean Schirmer strip scores were in the normal range ($10 mm)6 for both patient groups. The mean OSDI score was higher in nondiabetic patients as compared with that in diabetic patients. The mean ITF rank was similar for both patient groups. The majority of the patients in the diabetic group were classified as having “mild” dry eye, or a “1” for an ITF rank (71.1%); 26.3% of the diabetic patients were classified as having “moderate” dry eye, or “2” ranking, and 2.6% were classified as having “severe” dry eye, or a “3” ranking. In the nondiabetic patient group, 84.0% of the patients were classified as having mild (1) dry eye, 8.0% as having moderate (2) dry eye, and 8.0% as having severe (3) dry eye. No statistically significant differences were observed between the patient groups for TBUT, Schirmer strip, OSDI test scores, or ITF rank for the primary analysis of the full subject population. Secondary analyses were conducted for the subgroups of the study population based on diabetic versus nondiabetic subjects and the response to the query regarding the current use of dry eye therapy. Significant differences were observed between nondiabetic and diabetic subjects reporting at least the occasional use of dry eye therapy; mean tear film osmolarity was significantly higher in nondiabetic patients as compared with that in diabetic subjects (P = 0.004), whereas mean conjunctival staining scores were significantly higher in diabetic as compared with that in nondiabetic subjects (P = 0.008), and a significantly more rapid mean TBUT was observed for diabetic subjects as compared with that in nondiabetic subjects (P = 0.017) (Table 3). Within the subgroup of subjects that reported no current use of dry eye therapy, significant differences were also observed between diabetic and nondiabetic subjects. Mean corneal fluorescein staining scores and mean ITF ranking scores were significantly higher in diabetic subjects as compared with that in nondiabetic subjects (P = 0.003 for both tests between groups), whereas mean Schirmer test scores were significantly lower for nondiabetic subjects as compared with that in diabetic subjects (P = 0.001) (Table 4). Subgroup analyses by gender (only) revealed several trends within the study population. Female patients had
a higher mean OSDI score across the study population without regard to diabetic classification. The mean female patient OSDI score was 26.98, whereas the mean OSDI score was 13.57 for male patients. In comparing male diabetic patients to male nondiabetic patients, there were no significant differences for any categories. In contrast, nondiabetic females had a higher mean tear film osmolarity than female diabetic patients had (313.73 and 303.38 mOsm/L, respectively; P = 0.017). Additionally, female diabetic patients had a higher mean conjunctival staining score than nondiabetic patients had (2.90 and 2.11, respectively; P = 0.009).
Ó 2014 Lippincott Williams & Wilkins
DISCUSSION The purpose of this single-center clinical study was to assess the presence and relative severity of dry eye in diabetic patients as compared with that in nondiabetic patients at the time of a routine comprehensive examination. The panel of diagnostic indicators of dry eye used in this study allowed for an assessment of the disease severity level as per the ITF ranking criteria. Although the overall severity of dry eye was
TABLE 4. Dry Eye Assessments Listed by Patient Group (Subjects Reporting NO Use of Dry Eye Therapy) Diabetic Patients (n = 19) Osmolarity Osmolarity difference Fluorescein corneal staining Lissamine green conjunctival staining TBUT time, s Schirmer test score, mm OSDI test score ITF score
Nondiabetic Patients (n = 7)
P
305.5 12.2 0.39 2.4
(3.2) (2.9) (0.1) (0.2)
303.7 21.8 0.0 2.3
(4.5) (4.9) (0.0) (0.4)
0.749 0.09 0.003 0.68
4.0 20.0 26.3 1.4
(0.3) (1.8) (3.9) (0.1)
3.8 11.1 30.8 1.0
(0.6) (1.8) (5.8) (0.0)
0.736 0.001 0.528 0.003
Data shown as mean 6 standard error of the mean. Entries in bold represent statistically significant findings.
www.corneajrnl.com |
853
Beckman
found to be similar in the diabetic and nondiabetic patient groups according to the ITF scale (Table 2), there were significant differences observed between the 2 patient groups based on individual diagnostic criteria. The observation that the mean severity of lissamine green staining of the conjunctiva was found to be higher in the diabetic patient population (Table 2), and higher mean corneal fluorescein staining in diabetic patients reporting no use of dry eye therapy (in conjunction with a higher mean ITF ranking) may be in accordance with a general perception of ocular surface damage in diabetic patients. However, the finding that the mean tear film osmolarity was significantly higher in the nondiabetic patient group was unexpected based on previously reported findings.7 Corneal and conjunctival staining has been used for many years by clinicians to assess the integrity of the ocular surface. Osmolarity assessments also provide valuable insight into the quality of the tear film and can assist in a full diagnosis of a patient that may have dry eye. However, as illustrated by the results of this study, tear film osmolarity may not necessarily be expected to be higher in diabetic patients as compared with that in nondiabetic patients. Although abnormal tear film osmolarity values were observed in both patient groups, the mean osmolarity score was significantly higher in the nondiabetic patient group. The overall prevalence of dry eye and other diseases affecting the ocular surface is predictably high in the patient population visiting a cornea specialist. Although the severity of dry eye may vary widely between patients, the presence of ocular surface tissue staining (corneal or conjunctival) and rapid TBUT are particularly common diagnostic indicators for dry eye and noted in virtually all patients in this study. The ITF severity scores ranged from 1 to 3 in both patient groups in this study, with the majority of patients classified as having mild (ITF 1) dry eye as a result of the low incidence of corneal staining. Tear film osmolarity readings .308 mOsm/L and/or measurements between eyes that were $8 mOsm/L, values typically considered outside the normal range,6 were also a common finding in this study. Patients with diabetes frequently report symptoms of dry eye, such as foreign body sensation and burning, and upon evaluation are found to have an unstable tear film. In some patients, however, a reduction in corneal sensation is also associated with diabetes.11 One possibility for the lower percentage of diabetic patients reporting the use of artificial tears, dry eye medications, or dietary supplements in this study may be because of a reduced perception of dry eye symptoms. The secondary analysis of the study population indicated that the clinical assessment profile of diabetic and nondiabetic subjects that reported at least the occasional use of dry eye therapy is similar to that in the full study population (higher mean tear film osmolarity in nondiabetic subjects, and higher mean conjunctival staining in diabetic subjects), suggesting that although artificial tears/lubricants may be providing palliative relief to the patient’s dry eye symptoms, the basic clinical profile remains unchanged. This is an important consideration for the general clinical assessment of the dry eye status of diabetic and nondiabetic patients. The conduct of this study at a single clinical site and a relatively small sample size are limitations of the study.
854
| www.corneajrnl.com
Cornea Volume 33, Number 8, August 2014
A larger sample size at multiple clinical sites would be beneficial in the overall assessment of the differences in the characteristics of dry eye presented by diabetic and nondiabetic patients. However, one challenge that is frequently encountered in large, multicenter studies concerns the differences in assessment practices among clinical investigators. In that respect, the evaluation of all subjects reported in this study by a single investigator provides for uniformity across the data set. Despite the frequent association of diabetes and dry eye, the consequence of this risk factor for the ocular surface may be overlooked in clinical practice. As indicated by the differing clinical assessment profiles observed between diabetic and nondiabetic patients, it is important to evaluate diabetic patients for dry eye and other potential conditions affecting the ocular surface using a range of diagnostic techniques, particularly tear film osmolarity and vital staining. Early detection and appropriate treatment can play a vital role in the management of dry eye and other ocular inflammatory diseases.
ACKNOWLEDGMENTS The author would like to express his appreciation to Dr Ayelet Ben-David for preparing the statistical analysis. Writing and editorial assistance was provided to the authors by Kurt Brubaker of Bridge Over Brook Inc (Shady Cove, OR), and funded by Allergan Inc (Irvine, CA) at the request of the investigator. The author met the ICMJE authorship criteria. Neither honoraria nor payments were made for authorship. REFERENCES 1. Lemp MA. Report of the National Eye Institute/Industry Workshop on Clinical Trials in Dry Eyes. CLAO J. 1995;21:221–232. 2. Behrens A, Doyle JJ, Stern L, et al. Dysfunctional tear syndrome: a Delphi approach to treatment recommendations. Cornea. 2006;25:900–907. 3. Methodologies to diagnose and monitor dry eye disease: report of the Diagnostic Methodology Subcommittee of the International Dry Eye Workshop (2007). Ocul Surf. 2007;5:108–152. 4. Stern ME, Schaumburg CS, Pflugfelder SC. Dry eye as a mucosal autoimmune disease. Int Rev Immunol. 2013;32:19–41. 5. National Diabetes Fact Sheet: national estimates and general information on diabetes and prediabetes in the United States. Centers for Disease Control and Prevention. Atlanta, GA: US Department of Health and Human Services; 2011. 6. Tomlinson A, Bron AJ, Korb DR, et al. The International Workshop on meibomian gland dysfunction: report of the diagnosis subcommittee. Invest Ophthalmol Vis Sci. 2011;52:2006–2049. 7. Sagdık HM, Ugurbas SH, Can M, et al. Tear film osmolarity in patients with diabetes mellitus. Ophthalmic Res. 2013;50:1–5. 8. Bron A, Evans VE, Smith JA. Grading of corneal and conjunctival staining in the context of other dry eye tests. Cornea. 2003;22:640–650. 9. Guillon JP. Current clinical techniques to study the tear film and tear secretions. In: Korb D, Craig J, Doughty M, et al, eds. The Tear Film. London, United Kingdom: Buttersworth-Heinemann; 2002:51–83. 10. Schiffman RM, Christianson MD, Jacobsen G, et al. Reliability and validity of the Ocular Surface Disease Index. Arch Ophthalmol. 2000; 118:615–621. 11. Alves Mde C, Carvalheira JB, Módulo CM, et al. Tear film and ocular surface changes in diabetes mellitus. Arq Bras Oftalmol. 2008; 71:96–103.
Ó 2014 Lippincott Williams & Wilkins
