Efficiency of fenofibrate in facilitating the reduction of central macular thickness in diabetic macular edema.

PURPOSE: The purpose of this study is to study the benefit of addition of oral fenofibrate to the current regimen of diabetic macular edema (DME) management and quantify its effect on macular thickness and visual function in DME.
METHODS: Fifty-three eyes of 50 patients were randomized into treatment (Group A) (oral fenofibrate 160 mg/day) and control groups (Group B). Both groups underwent treatment of DME as per the standard treatment protocol of our hospital including intravitreal injections (anti-vascular endothelial growth factor/steroid) and grid laser. Patients were followed up every 2 months to note the visual acuity and central macular thickness (CMT) for 6 months.
RESULTS: Our groups were matched with respect to age (P = 0.802), mean diabetic age (P = 0.878), serum HbA1C levels (P = 0.523), and serum triglyceride levels (P = 0.793). The mean reduction in CMT was 136 μ in Group A and 83 μ in Group B at the end of 6 months. This difference was statistically significant (P = 0.031). Visual acuity improvement was 0.15 in Group A and 0.11 in Group B at the end of 6 months (P = 0.186). On subgroup analysis in Group A, we found that there was no difference in reduction of CMT between hypertensives and normotensives (P = 0.916), in patients with normal triglyceride levels and increased triglyceride levels (P = 0.975).
CONCLUSION: Addition of fenofibrate to the standard protocol of DME management seems to facilitate reduction of CMT and probably have an added benefit on the visual functions.

RCT Entities:   Population Interventions Outcomes

The management of diabetic macular edema (DME) has seen a phenomenal change in the past decades: From the use of macular photocoagulation[1] alone for several decades to multiple intravitreal injection and implants.[2345] Established patients with DME undergo multiple intravitreal injections, few sessions of laser apart from a strict emphasis on control of systemic parameters. Despite this, the treatment is prolonged, recurrences are common, and failures are frequent.[6]

Systemic therapy for DME largely meant control of diabetes, hypertension, and lipid levels.[789] It had nothing in specific predominantly due to inadequate evidence, associated complications along with dramatic benefits and safety profile of the injectable drugs.[1011121314] A revival in the use of systemic medication for a local condition caused by a systemic disease does seem to have occurred in the recent times with hypolipidemic drugs, namely, Atorvastatin[15] and Fenofibrate. Two multicentric trials FIELD[16] and ACCORD[17] studies have established that fibrates can be used in the control of diabetic eye disease. Fenofibrate reduced the frequency of first laser treatment for macular edema by 31% and for proliferative retinopathy by 30%. Furthermore, it reduced the progression of diabetic retinopathy by 5.0% over 5 years (P = 0.022, FIELD) and 3.7% over 4 years (P = 0.006, ACCORD-eye). These two trials, however, have not evaluated the effect of fenofibrate on the preexisting macular edema requiring treatment.

Whether the addition of this drug in patients with DME will enhance the response of DME to the current available treatment and give better visual outcome is the question, we have set out to answer.

Methods

This is a prospective randomized controlled trial conducted in the retina department of a tertiary eye care center in the south Indian population. The study involved 50 patients with type 2 diabetes having treatment naïve, center-involving DME. The inclusion criteria were central macular thickness (CMT) measured on stratus optical coherence tomography (OCT) of equal to or more than 250 μ, HbA1c of <9, normal creatinine levels and systemic blood pressure <140/90 mmHg.

Patients with abnormal renal parameters, known liver disease, media opacities preventing good macular evaluation, macular ischemia, foveal hard exudates, coexisting macular diseases other than DME were excluded from the study. Patients with high-risk proliferative diabetic retinopathy (PDR) and those treated previously with laser photocoagulation or intravitreal injections were excluded from the study. Patients with known allergy to fenofibrate were also excluded. The study was conducted after approval from the institutional review board.

Methodology

Patients were included in the study after obtaining an informed written consent. All patients underwent complete ophthalmic examination including detailed history, best corrected visual acuity (Snellen), slit lamp examination, detailed fundus examination, Fundus fluorescein angiography (Topcon TRC.50EX retinal camera), and OCT on the stratus OCT (Carl Zeiss Meditec, USA). A complete physical evaluation by an internist and blood parameters including serum glucose, glycated hemoglobin, serum creatinine, blood urea, liver function tests, and lipid profile were done. Patients were randomized by coin tossing. Our study included 28 eyes of 25 patients in Group A, and 25 eyes of 25 patients in Group B. Twenty-two of our patients had DME only in one eye. In three patients having DME in both eyes, both eyes were included in the study in Group A. In Group B, all our patients had unilateral DME.

All patients underwent treatment for DME according to the current treatment protocols of our hospital [Fig. 1]. This protocol was modified from DRCR, net protocol to suit the needs of developing country population.

Figure 1

Treatment algorithm for patients with diabetic macular edema followed in our study. CMT=Central macular thickness; IVTA=Intravitreal triamcinolone acetate

Patients in Group A were in addition given oral fenofibrate 160 mg/day for 6 months as a single evening dose.

Anti-vascular endothelial growth factor (VEGF)/triamcinolone (IVTA) was given 3.5 mm/4 mm from limbus in the inferotemporal quadrant. One session of grid/focal laser photocoagulation using 532 nm frequency doubled Nd:Yag (Oculight, Iridex Inc., USA) was given in the leaking areas as seen on the FFA as per the DRCR guidelines[18] after edema subsided (achievement of normal foveal contour). Patients were given a trial of topical steroids for steroid responsiveness before IVTA and carefully followed up for the same.

Patients with nonhigh risk PDR were treated with pan-retinal photocoagulation in three sittings, power adjusted to obtain a light grey burn.

Patients were followed up every 2 months with complete ophthalmic evaluation and OCT. Results were analyzed after 6 months.

Statistical analysis

The sample size was calculated to give a power of 0.8 to the study. Qualitative data were analyzed by Chi-square test, Mann–Whitney test and Student's t-test. In all the above tests, a P ≤ 0.05 was accepted as indicative of statistical significance with 95% confidence interval.

Results

Demographic data

We studied 28 eyes of 25 patients in Group A and 25 eyes of 25 patients in Group B. Most of our patients were in age group of 51–70 years. Mean age in Group A was 60.16 years and Group B was 59.40 years. Males predominated in both groups (74% in Group A and 76% in Group B). Mean duration of diabetes was 11.38 years in Group A and 11.09 years in Group B. Most of our patients had diabetes for 5–15 years. More than 50% of the patients in each group were hypertensives on treatment, and there was no statistically significant difference in the distribution of hypertensives in the two groups. Our groups were comparable based on the demographic data.

Baseline laboratory parameters

Baseline laboratory data are shown in Table 1. The two groups were matched with respect to mean HbA1C, mean PPBS levels (P = 0.593), mean FBS levels (0.234), mean blood urea (P = 0.313), mean serum creatinine (P = 0.419), mean triglyceride levels (P = 0.793), mean low-density lipoprotein (LDL) levels, and mean high-density lipoprotein levels. Incidentally, we found a significant difference in the mean serum cholesterol and mean very LDL (VLDL) levels in the two groups in favor of Group B.

Table 1

Comparison of various laboratory parameters between the two groups

Despite the statistically significant difference the actual value of these parameters was within the normal limits and did not warrant any treatment.

Baseline clinical data

Seventy-five percent of the eye in Group A and 72% in Group B were phakic. 20 of the 28 eyes in Group A had non-PDR versus 16 of 25 eyes in Group B [Tables 2 and 3]. This difference was not statistically significant (P = 0.563). Rest of patients had early PDR. 71% and 60% of patients respectively in Group A and Group B had diffused macular edema and rest had focal edema.

Table 2

Baseline clinical data

Table 3

Results in control group (Group B)

Clinical results

Results in control group (Group B)

Group B received an average of 2.23 injections in 6 months. We found a statistically significant reduction in the macular edema from 404 μ (±91.59) to 319 μ (±57.29 μ) in the control group (Group B). Visual acuity improved from 0.41 log mar to 0.33 log mar. This was statistically significant with P < 0.001 at the end of 6 months. Three eyes (12%) had ≥2 line improvement 13 eyes (52%) had ≥1 line improvement in vision, and 9 eyes (36%) had stabilized vision [Table 3].

Results in fenofibrate group (Group A)

Clinical photograph and OCT of one of our patient in fenofibrate group is shown in Figs. 2–5. Patients in Group A received an average of 2.1 injections in 6 months. In the fenofibrate group (Group A), we found that CMT decreased from 429 μ (±130 μ) to 293.96 μ (±83.05 μ). This was statistically significant with a P < 0.001 at the end of 6 months. Visual acuity improved from 0.41 log mar to 0.26 log mar at the end of 6 months. More than two line improvement in vision was seen in 7 eyes (25%), more than one line improvement was seen in 14 eyes (50%) and stabilized in 7 (25%) eyes [Table 4].

Figure 2

Baseline photograph of patient with diabetic macular edema and PDR

Figure 5

Optical coherence tomography at six months

Table 4

Results in fenofibrate group (Group A)

Baseline optical coherence tomography

Clinical picture at six months

Comparative analysis of treatment outcomes in case and control group (Group B)

Effect on central macular thickness

Initial CMT was higher in the fenofibrate group (Group A) compared to control group (Group B) without statistical significance indicating that we could compare the two groups. The mean CMT was reduced by 136 μ in Group A and about 83 μ in Group B from the baseline at the end of 6 months. The results were statistically significant at the end of 6 months with a P value of 0.031 [Table 5]. The difference between the CMT of two groups steadily decreased from baseline to 6th month. Mean CMT in Group A became lower than Group B at month 4 and the difference was statistically significant at 6 months [Fig. 6].

Table 5

Comparison of mean central macular thickness between the two groups (Mann-Whitney test)

Figure 6

Comparative change in the central macular thickness between Group A and Group B over 6 months

Effect on visual acuity

The improvement in mean visual acuity in Group A was 0.15 log mar units as compared to 0.11 log mar units in Group B at the end of 6 months [Table 6 and Fig. 7]. Although the visual acuity was higher in Group A at 6th month this difference was not statistically significant with P value of 0.186. A higher number of patients achieved a two-line improvement in the vision in Group A compared to Group B (7 vs. 3).

Table 6

Comparison of mean best corrected visual acuity (logarithm of the minimum angle of resolution) between the two groups (Mann-Whitney test)

Figure 7

Comparative change in visual acuity between Group A and Group B over 6 months

Reduction of hard exudates

The reduction in hard exudate was noted in 6 eyes (19%) in Group A and 4 eyes (16%) in Group B at the end of the follow-up period, but the difference between the two groups was not statistically significant.

Results within the fenofibrate group (Group A)

Subgroup analysis in the Group A showed that there was no statistically significant difference in reduction of CMT between eyes with PDR and NPDR (NPDR – 150 μ, PDR – 120 μ) (P = 0.746). There was no statistically significant difference in CMT reduction between hypertensives and normotensives (hypertensives – 130 μ, normotensives – 141 μ) (P = 0.916). We also did not find a statistically significant difference between normal and abnormal triglyceride levels and normal and abnormal cholesterol levels (Normal TG - 136 μ, high TG - 134 μ, P = 0.975) (normal cholesterol - 122 μ, high cholesterol - 183 μ, P = 0.247) [Table 7].

Table 7

Subanalysis of fenofibrate group (Group A) with respect to central macular thickness

Discussion

Our study found that addition of fenofibrate to an established protocol for management of DME improves the reduction of CMT significantly irrespective of triglyceride, hypertensive, and the diabetic retinopathy status. In addition, it improves the visual acuity though the improvement is not statistically significant at 6 months.

In this study, two groups were matched with respect to chronological age, diabetic age, diabetic control, staging of diabetic retinopathy, and systemic parameters except serum cholesterol and serum VLDL levels (these two parameters were tilted in favor of Group B and these patients were not started on anti-cholesterol therapy as they were with in normal limits of serum cholesterol levels).

Both groups received the same treatment protocol for DME at the same institute under two ophthalmologists. The only difference thereby is the introduction of oral Fenofibrate in Group A.

Our study was done with a time domain OCT machine. This could have underestimated the CMT compared to spectral domain OCT. Since both initial and final scans were done in the same TD-OCT the decrease in the CMT could be evaluated.[19]

Statistically significant decrease in CMT and the improvement in the visual acuity was achieved in both groups. Independently, both the treatment protocols have shown improvement in the functional and anatomical outcome. Hence, any difference in the outcome either functional or anatomical in Group A compared to Group B can be attributed to the only difference between the groups, namely, 160 mg of oral Fenofibrate in Group A.

Fenofibrate and proposed mechanism of action in diabetes: Fenofibrate is a peroxisome proliferator-activated receptor α agonist. It is used in hypertriglyceridemia and mixed dyslipidemia.[20] Fenofibrate is converted to fibric acid in plasma and stimulates PPRFα which works through various mechanisms at the cellular level [Fig. 8].

Figure 8

Biochemical actions of fenofibrate after conversion to fibric acid in plasma. Fibric acid activates PPAR to bring about the above mentioned changes

Pathogenesis of diabetic retinopathy involves an increase in nitric oxide levels, increased inflammatory mediators through the cyclooxygenase pathway, increased production of oxygen free radicals due to advanced glycosylation end products and high intracellular glucose-induced capillary cell apoptosis.[21] Multiple biochemical actions after stimulation of PPRFα seems to counter these pathological mechanisms giving a sound biochemical basis for the use of fenofibrate in diabetes.[22232425262728293031323334353637]

Clinical evidence provided by FIELD and ACCORD studies also support the utility of this PPRFα agonist in diabetes.

FIELD study a multicentric study by Keech et al. concluded that requirement for the first laser was significantly lower in the fenofibrate group (P = 0.0002, HR – 0.69) and 2 step progression in retinopathy was significantly lower in patients with preexisting retinopathy (P = 0.004). In this study, laser treatment of macular edema specifically reduced by 31% (P = 0.002). Each pathology showed an accumulation of benefit over time with crossing over of curves at 8 months. Our study also showed statistically significant improvement in CMT at 6th month with a gradual widening of the gap between the two groups.

ACCORD study observed that the beneficial effects were maximum in mild-to-moderate NPDR (steps 2–4 at enrollment) (P = 0.00009) and addition of fenofibrate had synergistic effect with intensive glycemic control. In steps 5–7, fenofibrate group showed a significant benefit whereas the intensive glycemic control group did not. However, there was no significant benefit in the progression of DME in 3 steps (P = 0.78). This study involved only initial and the 4th year retinal photograph to comment on the progression of macular edema. The absence of frequent follow-up could result in loss of critical information underestimating the effect of the drug. The treatment used in ACCORD study was grid/focal laser photocoagulation alone. In our study, we found that addition of fenofibrate to an established regimen of anti-VEGF injection and laser reduced the CMT significantly at 6 months. The synergistic effect of fenofibrate may be enhanced when added to a regimen of anti-VEGF injection and laser.

The MacuFen study (n = 100) by Massin et al. found that use of 135 mg of fenofibrate moderately reduced the total macular volume in patients with non-center involving macular edema but the reduction was not statistically significant.[38] This could be due to lower drug dosage. Dosage of drug used in FIELD and ACCORD were 200 mg and 160 mg, respectively. There was a 5% and 3.7% risk reduction in FIELD and ACCORD, respectively. Our study found that 160 mg of drug was beneficial to the patients of DME. Whether the differences seen in various studies are due to different dosages of drug used needs to be studied, and there is need to zero in on the optimal dosage of the drug. Another possibility as highlighted by the FIELD and ACCORD studies is that higher stage of diabetic retinopathy shows higher benefit with addition of fenofibrate. This could be the case in our study where all patients had a center-involving DME requiring treatment.

FIELD and ACCORD studies observed that the effect of fenofibrate on slowing progression of diabetic retinopathy was seen irrespective of triglyceride levels which our study also corroborated.

Earlier trials with fibrates were limited due to the incidence of rhabdomyolysis.[39] FIELD and ACCORD did not report any musculoskeletal side effects. Our study also did not have any incidence of musculoskeletal side effects.

Limitations

Small sample size, single-center study, single dosage arm, and follow-up duration of 6 months are our limitations. Despite our limitations, our study provides important information regarding the utility of Fenofibrate in DME.

Conclusion

Fenofibrate seems to facilitate reduction of CMT in DME. Larger, multicentric, and longer duration studies with different dosage of fenofibrate would throw more light on this initial finding.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.