Glycemic variability and other risk factors for diabetic retinopathy: A pilot case-control study

Amruta S Kulkarni; Karakkattu V Kavitha; Nikita S Sarkar; Vedavati B Purandare; Savita Bhat; Shalbha Tiwari; Ambika G Unnikrishnan

doi:10.4103/cdrp.cdrp_3_21

ORIGINAL ARTICLE

Year : 2022 | Volume : 1 | Issue : 1 | Page : 13-17

Glycemic variability and other risk factors for diabetic retinopathy: A pilot case-control study

Amruta S Kulkarni¹, Karakkattu V Kavitha², Nikita S Sarkar³, Vedavati B Purandare³, Savita Bhat¹, Shalbha Tiwari⁴, Ambika G Unnikrishnan³
¹ Department of Ophthalmology, Chellaram Diabetes Institute, Pune, Maharashtra, India
² Department of Podiatry, Chellaram Diabetes Institute, Pune, Maharashtra, India
³ Department of Diabetes and Endocrinology, Chellaram Diabetes Institute, Pune, Maharashtra, India
⁴ Department of Research, Chellaram Diabetes Institute, Pune, Maharashtra, India

Date of Submission	14-Oct-2021
Date of Decision	23-Nov-2021
Date of Acceptance	23-Nov-2021
Date of Web Publication	07-Jan-2022

Correspondence Address:
Ambika G Unnikrishnan
Chellaram Diabetes Institute, Lalani Quantum, Pune-Bangalore Highway, Bavdhan, Pune -411 021, Maharashtra
India

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/cdrp.cdrp_3_21

Abstract

Background: Hyperglycemia is a known risk factor for diabetic retinopathy (DR) but the association between glycemic variability and DR is unclear. We aim to evaluate the glycemic variability in DR through retrospective continuous glucose monitoring (CGM) and assess its effect on the clinical profile of participants with or without DR. Material and Methods: Retrospective observational hospital-based case-control study. We collected anthropometric and clinical data of 74 people with type 2 diabetes from our ophthalmology database whose retrospective CGM data were available. Among them, 37 had DR (cases) and 37 did not have DR (controls). The data were analyzed using SPSS version 27. Results: Duration of diabetes and glycosylated hemoglobin (HbA1c) was significantly higher and the estimated glomerular filtration rate (eGFR) was significantly lower in the cases compared to the controls. CGM markers, like time-above-range, average glucose, glucose management indicator, were higher while time-in-range was lower in the cases compared to the controls (P = ns). Time-below-range targets in people >65 years were met in a lower proportion (p < 0.05) of people in the cases (50%) compared to the controls (92%). Conclusion: Duration of diabetes, low eGFR, and high HbA1c showed significant association with retinopathy in type 2 diabetes. Although markers of glycemic variability did not show a statistically significant difference in cases compared to controls, all indices of glycemic variability were numerically higher in people with DR. Hypoglycemia in elderly participants with DR and its implications on achieving targets requires more research.

Keywords: Continuous glucose monitoring, diabetes, glycemic markers, retinopathy, time in range

How to cite this article:
Kulkarni AS, Kavitha KV, Sarkar NS, Purandare VB, Bhat S, Tiwari S, Unnikrishnan AG. Glycemic variability and other risk factors for diabetic retinopathy: A pilot case-control study. Chron Diabetes Res Pract 2022;1:13-7

How to cite this URL:
Kulkarni AS, Kavitha KV, Sarkar NS, Purandare VB, Bhat S, Tiwari S, Unnikrishnan AG. Glycemic variability and other risk factors for diabetic retinopathy: A pilot case-control study. Chron Diabetes Res Pract [serial online] 2022 [cited 2023 Mar 29];1:13-7. Available from: https://cdrpj.org//text.asp?2022/1/1/13/335256

Introduction

Diabetic retinopathy (DR) is the fifth-most common cause of moderate-to-severe vision loss, globally.^[1],[2] Conventional risk factors for DR include uncontrolled blood glucose levels, a longer duration of diabetes, higher blood pressure, dyslipidemia, and the presence of albuminuria.^[2] Among the above risk factors, high glucose levels and duration of diabetes are considered to be the strongest risk factors for DR.^[3] Therefore, there has been a strong focus on intensively controlling the blood glucose levels, such as glycosylated hemoglobin (HbA1c), fasting and postmeal glucose levels in preventing the onset and progression of retinopathy. However, reaching optimal levels of the glycemic parameters, like HbA1c, while necessary, may not reflect a sufficient degree of control to completely prevent the onset and progression of DR.^[4]

Glycemic variability has long been invoked as a possible risk factor for DR.^[5],[6] The term glycemic variability (GV) refers to fluctuations in blood glucose oscillations that occur throughout the day, or at the same time on different days. The term, time-in-range (TIR), considered to be a marker of glycemic variability, is of recent interest, and refers to the time spent in the target blood glucose levels, without hypoglycemia or hyperglycemia. A lower TIR suggests a higher HbA1c as well as more glycemic variability. In a large study from China, using a 72-h continuous glucose monitoring (CGM) device in type 2 diabetes, it was shown that a low TIR correlated with a higher prevalence of DR.^[7]

Recently, fourteen-day retrospective CGM has become a more commonly preferred option for assessment of hyperglycemia and TIR, as it is calibration-free and measures glycemia for a longer duration. In this pilot study, we retrospectively evaluated markers of glycemic variability via a retrospective CGM in people with type 2 diabetes with and without retinopathy, in a case-control study design.

Material and Methods

We included 37 adults with type 2 diabetes and retinopathy (DR) who were visiting the ophthalmology unit of our institution and had been advised the CGM device by the physician as cases. As a control group, we selected 37 consecutive adults with type 2 diabetes and no retinopathy (no-DR), who had also been advised CGM by their physician. The technology used for assessment was the FreeStyle Libre-pro system.^[8] This is a retrospective CGM system and measures the interstitial glucose every 15 min.

The FreeStyle Libre device is a disc-like sensor placed on the outer and posterior aspect of the upper arm. The device uses wired enzyme technology. The device is factory calibrated and does not need finger-stick glucose calibration during use. The sensor is stable for up to 14 days. The study was conducted in a masked manner, as the FreeStyle Libre Pro System used did not give patients real-time access to their glucose data. Hence, the glucose values were less likely to be affected by deliberate changes in diet, exercise, or medication adherence by the patients which may occur during real-time monitoring. The data stored in the sensor are downloaded by “flashing” a reader over the sensor at a later date, a technique called “intermittent scanning.”

The diagnosis of type 2 diabetes was made based on American Diabetes Association criteria.^[9] Retinopathy was diagnosed via fundus photography by a retina specialist, and stages were graded according to the modified Airlie House classification used in the Early Treatment DR Study (ETDRS).^{[10],[11],[12]} The extent and location of specific retinal lesions were using 7 stereoscopic pairs of photographs for each eye (referred to as ETDRS 7 standard fields). The ETDRS criteria strongly correlate with the risk of progression of retinopathy in diabetes.^{[10],[11],[12]}

Markers of glycemic variability as well as TIR (time spent between 70 and 180 mg/dl) were calculated based on the methods listed in the references.^{[13],[14],[15]} A recent consensus was referred to for choosing TIR, time-above-range (TAR), and time-below-range (TBR) cutoffs.^[15] A time in range of >70% was considered adequate for participants at or <65 years, while a time in range of >50% was considered adequate for participants >65 years; for TBR, the respective cutoffs of <4% and <1%, respectively, were considered optimal.^[15] It is well known that the percentage coefficient of variation (CV) is an indicator of glycemic variability. In general, the value of CV ≥36 is considered to reflect significant glycemic variability.^[16]

Statistical analysis

Descriptive statistics are given as means and standard deviations which were done using Microsoft excel 2010. The Chi-square value and P value were determined using SPSS version 27.0. (IBM SPSS, Chicago, USA). Markers of glycemic variability such as continuous overall net glycemic action, Low Blood Glucose Index, High Blood Glucose Index, mean of daily differences (MODD), and mean amplitude of glucose excursion (MAGE) were extracted using GlyCulator 2.0. (Department of Biostatistics and Translational Medicine, Medical University of Lodz, Poland) which calculates the glycemic variability indices from the raw CGM data (https://apps.konsta.com.pl/modules/glyculator/).

Results

Clinical characteristics of the participants are shown in [Table 1]. Age and gender showed no significant differences, though the mean age of participants with retinopathy was numerically higher. The duration of diabetes was significantly longer in people with retinopathy. The use of oral anti-diabetic drugs (OADs) alone for glucose control was significantly less common in people with retinopathy (P < 0.001). While the use of insulin only was similar in the two groups, the use of insulin + OADs was more common in the DR group (P = 0.001). The mean HbA1c was significantly higher in cases than controls. The mean e-GFR was significantly low in people with retinopathy. It was observed that the mean TIR was lower in cases compared to controls, but the difference was statistically insignificant. The other parameters including the CGM-derived average glucose and glucose management indicator (GMI) were also similar in the two groups. Diabetes-related complications and co-morbidities were not different between the two groups [Table 2].

Table 1: Clinical characteristics of the diabetic retinopathy and no diabetic retinopathy group

Click here to view

Table 2: Diabetes-related complications and comorbidities among diabetic retinopathy and no-diabetic retinopathy group

Click here to view

Among the 37 people with retinopathy, the disease was classified as mild (8/37 [22%]), moderate (12/37 [32%]), severe (10/37 [27%]) and proliferative (7/37 [19%]), based on the worse among the two eyes. Involvement of the center was seen in (7/37 [19%]) people with retinopathy. About 27/37people with retinopathy were younger than 65 years and a TIR of >70% was achieved in 9 adults (33%). Furthermore, 25/37people without retinopathy were younger than 65 years and a TIR of >70% was achieved in 13 adults (52%). The differences between people (<65 years.) with and without retinopathy achieving TIR targets were not significant (p = 0.173). 10/37 people with retinopathy were at or older than 65 years and a TIR of >50% was achieved in 8 adults (80%). 12/37 people without retinopathy were at or older than 65 years and a TIR of >50% was achieved in 8 adults (66.7%). The differences between people (≥65 years.) with and without retinopathy achieving targets were not significant (p = 0.484) as noted in [Table 3]. As shown in [Table 3], in the age group >65 years, the TBR target of <1% was achieved in 91.7% of people without retinopathy and 50% of people with DR (p = 0.029).

Table 3: Type 2 diabetes mellitus with diabetic retinopathy and no-diabetic retinopathy for age <65 years and ≥65 years for time-in-range and time-below-range

Click here to view

To assess glycemic variability using the CV cutoff of 36, Chi-square test was carried out between the two groups comparing people with CV <36 and CV ≥36 in both DR and no-DR. 12/37 participants with retinopathy and 14 out of the 37 controls had a CV ≥36. These differences were not significant (P = 0.626).

The GMI, average glucose, TAR, and TBR were all numerically higher in people with retinopathy (P = not significant). Other indices of glycemic variability were numerically higher in people with DR; at the same time, statistical significance was not reached [Table 4].

Table 4: Glycemic variability between diabetic retinopathy and no-diabetic retinopathy group

Click here to view

Discussion

The present study is the first case-control, hospital-based study from Western India, undertaken to assess the association of glycemic variability in participants with DR. Glycemic variability was assessed by CGM (Freestyle Libre Pro). In our study, people with DR had a longer duration of diabetes, a lower estimated glomerular filtration rate (eGFR), and a higher HbA1c when compared to controls. Our findings were similar to the results of a recently published longitudinal study from Chennai, India including 19,909 individuals with T2 diabetes where people with DR had diabetes for a longer duration, lower body mass index, higher FPG, higher HbA1c, higher systolic blood pressure, higher serum creatinine, lower eGFR and higher levels of albuminuria, than those without DR.^[17]

There are various CGM devices available in India which provide data on blood glucose levels. Devices that are used in India are; (a). The FreeStyle Libre Pro which can be used up to 14 days and is affordable meant for healthcare professionals, (b). FreeStyle Libre system which can be used up to 14 days, meant for people with diabetes (c). Dexcom G6 - Real - time CGM, (d). Medtronic i Pro2 and, (e). Medtronic Guardian Connect which can be used for up to 6 days. Among them, the FreeStyle Libre Pro, is a retrospective CGM. Dexcom G6 is a Real-time CGM which can be used up to 7 to 10 days, with trends and alerts and it continuously measures glucose levels every 5 minutes. Medtronic i Pro 2 is a professional CGM sensor which works for 5 to 7 days where the real-time glucose cannot be assessed by the patient. Medtronic Guardian Connect continuously measures glucose levels every 5 min which can be assessed via phone. Among the options, we studied the Freestyle Libre Pro version, which is affordable, commonly used, and importantly is the masked system which is suitable for study purposes as it overcomes bias due to changes made to diet, exercise, or medications, which can be seen in real-time CGM systems.

Furthermore, in our study, people with DR were less likely to be managed with OADs (p < 0.05) alone, as they were more likely to need insulin in combination with oral drugs. Our observations were similar to another study, wherein authors reported more than 75% of patients with PDR were on insulin (in addition to OAD) for management of diabetes as compared with only 20% of those without DR.^[17]

CGM variables such as GMI, average glucose, TAR, MAGE, and MODD were numerically higher in cases compared to controls, but statistical significance was not reached with the small sample size. The TIR was numerically lower in people with DR, showing that people with diabetes and retinopathy spent less time in euglycemia when compared with people who have diabetes and no retinopathy.

As per a recent consensus, in people with diabetes below and above 65 years, the TIR targets are >70% and >50%.^[15] Attainment of consensus-driven targets for the time-in range was similar between the two groups. For TBR targets, it has been suggested that for people with diabetes aged >65 years and <65 years, respectively, should have TBR values of <1% and <4%, respectively. Our analysis suggests that in age >65 years the TBR targets of <1% could be achieved in 92% of people with diabetes and no retinopathy, but only in 50% of people with diabetes and retinopathy (P < 0.05). In other words, elderly people with DR may also have a greater predisposition to hypoglycemia. This could limit the clinician's efforts to achieve euglycemia. These findings may reflect greater glycemic variability below the normal ranges in elderly people with diabetes and retinopathy, needing more dose monitoring and frequent titration of therapy in this age group. These results also suggest the need for further research into the links between hypoglycemia and DR in the elderly, and its implications on achieving glycemic targets.

In another study from China (n = 3262), increasing severity of retinopathy was inversely correlated with TIR quartiles.^[7] This study assessed TIR using the shorter duration CGM, whereas in our study, the 14- day long CGM was the tool used. In addition to CGM systems, other forms of glycemia assessments have also been measured to study glycemic variability in DR.^[18] In the Rio De Janerio Diabetes Cohort Study, visit-to-visit glycemic variability, estimated by either HbA1c or fasting plasma glucose, could predict retinopathy progression.^[19] Taken together, these studies suggest the need for a longer-term, prospective study with a large sample size to ascertain the influence of glycemic variability on DR.

One of the limitations of our study is the small sample size. However, to the best of our knowledge, this is the first case-control study from India of ambulatory glucose profiles and time in range assessments in people with and without DR. This is a hypothesis-generating study which could be corroborated with similar studies with large sample size. Control of potential risk factors for DR, especially glycemic variability continues to be an important area for future research.

Conclusion

To conclude, indices of hyperglycemia and glycemic variability were numerically higher in people with DR compared to controls. In elderly people > 65 years of age, a significantly lesser proportion of people with DR (compared to control) could reach a TBR of < 1%, suggesting a higher predisposition to hypoglycemia. This pilot study needs corroboration with greater sample size and more multicenter studies to understand glycemic variability and its impact on DR.

Acknowledgment

We thank Mr. Shrivallabh Sane, Lecturer in Biostatistics, Maharashtra Institute of Mental Health (MIMH) Pune for his help rendered in the analysis of the data, and Dr. Arpana Sathe Shendge for her contribution to data extraction.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

References

1.	Leasher JL, Bourne RR, Flaxman SR, Jonas JB, Keeffe J, Naidoo K, et al. Global estimates on the number of people blind or visually impaired by diabetic retinopathy: A meta-analysis from 1990 to 2010. Diabetes Care 2016;39:1643-9.
2.	Yau JW, Rogers SL, Kawasaki R, Lamoureux EL, Kowalski JW, Bek T, et al. Global prevalence and major risk factors of diabetic retinopathy. Diabetes Care 2012;35:556-64.
3.	Khan R, Singh S, Surya J, Sharma T, Kulothunga V, Raman R. Age of onset of diabetes and its comparison with prevalence and risk factors for diabetic retinopathy in a rural population of India. Ophthalmic Res 2019;61:236-42.
4.	Maa AY, Sullivan BR. Relationship of hemoglobin A1C with the presence and severity of retinopathy upon initial screening of Type II diabetes mellitus. Am J Ophthalmol 2007;144:456-7.
5.	Sartore G, Chilelli NC, Burlina S, Lapolla A. Association between glucose variability as assessed by Continuous Glucose Monitoring (CGM) and diabetic retinopathy in type 1 and type 2 diabetes. Acta Diabetol 2013;50:437-42.
6.	Suh S, Kim JH. Glycemic variability: How do we measure it and why is it important? Diabetes Metab J 2015;39:273-82.
7.	Lu J, Ma X, Zhou J, Zhang L, Mo Y, Ying L, et al. Association of time in range, as assessed by continuous glucose monitoring, with diabetic retinopathy in type 2 diabetes. Diabetes Care 2018;41:2370-6.
8.	Leelarathna L, Wilmot EG. Flash forward: A review of flash glucose monitoring. Diabet Med 2018;35:472-82.
9.	American Diabetes Association. 2. Classification and Diagnosis of Diabetes: Standards of Medical Care in Diabetes-2018. Diabetes Care 2018;41:S13-27.
10.	Early Treatment Diabetic Retinopathy Study Research Group. Grading diabetic retinopathy from stereoscopic color fundus photographs – An extension of the modified Airlie House classification. ETDRS report number 10. Early Treatment Diabetic Retinopathy Study Research Group. Ophthalmology 1991;98:786-806.
11.	Wilkinson CP, Ferris FL 3^rd, Klein RE, Lee PP, Agardh CD, Davis M, et al. Proposed international clinical diabetic retinopathy and diabetic macular edema disease severity scales. Ophthalmology 2003;110:1677-82.
12.	Early Treatment Diabetic Retinopathy Study Research Group. Fundus photographic risk factors for progression of diabetic retinopathy. ETDRS report number 12. Early Treatment Diabetic Retinopathy Study Research Group. Ophthalmology 1991;98:823-33.
13.	Monnier L, Colette C, Owens DR. The application of simple metrics in the assessment of glycaemic variability. Diabetes Metab 2018;44:313-9.
14.	Czerwoniuk D, Fendler W, Walenciak L, Mlynarski W. GlyCulator: A glycemic variability calculation tool for continuous glucose monitoring data. J Diabetes Sci Technol 2011;5:447-51.
15.	Battelino T, Danne T, Bergenstal RM, Amiel SA, Beck R, Biester T, et al. Clinical targets for continuous glucose monitoring data interpretation: Recommendations from the international consensus on time in range. Diabetes Care 2019;42:1593-603.
16.	Danne T, Nimri R, Battelino T, Bergenstal RM, Close KL, DeVries JH, et al. International consensus on use of continuous glucose monitoring. Diabetes Care 2017;40:1631-40.
17.	Rajalakshmi R, Shanthi Rani CS, Venkatesan U, Unnikrishnan R, Anjana RM, Jeba Rani S, et al. Correlation between markers of renal function and sight-threatening diabetic retinopathy in type 2 diabetes: A longitudinal study in an Indian clinic population. BMJ Open Diabetes Res Care 2020;8:e001325.
18.	Sun B, Luo Z, Zhou J. Comprehensive elaboration of glycemic variability in diabetic macrovascular and microvascular complications. Cardiovasc Diabetol 2021;20:9.
19.	Cardoso CR, Leite NC, Moram CB, Salles GF. Long-term visit-to-visit glycemic variability as predictor of micro- and macrovascular complications in patients with type 2 diabetes: The Rio de Janeiro type 2 diabetes cohort study. Cardiovasc Diabetol 2018;17:33.