Abstract
Purpose
To investigate the risk factors for subretinal fluid (SRF) in diabetic macular edema (DME).
Methods
We reviewed the records of 66 patients with DME. Systemic parameters, including hypertension, glycosylated hemoglobin, serum fasting glucose, total cholesterol, triglycerides, and diabetic kidney disease, were evaluated. Renal parameters for diabetic kidney disease included serum albumin, serum creatinine, albuminuria, and estimated glomerular filtration rate. Ocular factors included visual acuity and diabetic retinopathy, and the following parameters are evaluated through optical coherence tomography examination: disorganization of the retinal inner layers, loss of ellipsoid zone, central subretinal fluid thickness, central macular thickness, and presence of SRF.
Results
Higher albuminuria (odds ratio, 3.431; 95% confidence interval, 1.039–11.334; P=0.043) was associated with the presence of SRF in patients with DME. Lower serum albumin levels (beta = −14.028, se = 6.646, P=0.044) were associated with increased SRF thickness.
Conclusions
Poor kidney function was associated with the presence of SRF in DME. Screening for SRF in DME in patients with higher albuminuria and lower serum albumin levels should be routinely performed.
1. Introduction
Diabetic macular edema (DME) is a major cause of visual impairment in patients with diabetes [1] and can be classified into cystoid, diffuse, serous, or mixed types.
DME occurs because of the disruption of the blood-retinal barrier. Diabetic kidney disease and DME may share a similar microvascular pathophysiology [2]. Serum albumin, creatinine, albuminuria, and estimated glomerular filtration rate (eGFR) have been used as markers of diabetic kidney disease. Previous studies have shown that albuminuria and macroalbuminuria are closely associated with DME [2–4]. However, eGFR does not seem to have any effect on the severity or pattern of DME [5, 6].
Subretinal fluid (SRF) is found in 15%–30% of DME cases [2]. Persistent SRF may be detrimental to the retinal pigment epithelium and photoreceptors [7]. Few studies have reported the association between SRF in DME and diabetic kidney disease. Koo et al. [8] found a significantly higher frequency of SRF in DME in patients with albuminuria. In this study, we investigated the association between diabetic kidney disease and the presence and severity of SRF in patients with diabetes.
2. Materials and Methods
This study adhered to the tenets of the Declaration of Helsinki. Ethical approval for this study was waived by the Local Ethics Committee of Eye Hospital, China Academy of Chinese Medical Sciences in view of the retrospective nature of the study.
2.1. Data Collection
Eyes that had undergone laser photocoagulation, intravitreal injection, or intraocular surgery within 3 months prior to the study were excluded. Patients with macular edema of other causes, such as retinal vein occlusion and uveitis, were also excluded. Finally, a total of 66 patients with DME who visited our department between September 01, 2017, and September 30, 2020, were retrospectively enrolled.
Patient characteristics, including age, gender, duration of diabetes, hypertension, visual acuity, serum fasting glucose, glycosylated hemoglobin (HbA1c), triglycerides, total cholesterol, serum albumin, serum creatinine, urinary albumin creatinine ratio, eGFR, and diabetic retinopathy (DR) severity, were recorded. Disorganization of the retinal inner layers, loss of ellipsoid zone, central SRF thickness, central macular thickness (CMT), and presence of SRF were documented using optical coherence tomography. Disorganization of the retinal inner layers was defined as the inability to distinguish the boundaries between the ganglion cell—inner plexiform layer complex, inner nuclear layer, and outer plexiform layer [9]. DR severity was categorized as mild nonproliferative retinopathy (mild NPDR), moderate NPDR, severe NPDR, and proliferative diabetic retinopathy. Overnight first-void urine samples were collected. Albuminuria was defined as a urinary albumin-to-creatinine ratio >30 mg/g. eGFR was calculated by using the chronic kidney disease epidemiology collaboration equation and was categorized into the following five groups: <30 mL/min/1.73 m2 (stage 1), 30–44 mL/min/1.73 m2 (stage 2), 45–59 mL/min/1.73 m2 (stage 3), 60–89 mL/min/1.73 m2 (stage 4), or >90 mL/min/1.73 m2 (stage 5). SRF, a categorical variable, was divided into two categories: the presence and absence of SRF. Hypertension was analyzed as a categorical variable, with patients divided into two categories according to the presence or absence of hypertension. Data from the worst eye were used for analyses.
2.2. Statistical Analyses
Statistical analyses were performed using SPSS Statistics for Windows, version 26.0 (IBM, Armonk, NY). Summary statistics included the mean ± standard deviation, where appropriate. Binary and multiple logistic regression models were used to assess the risk factors for SRF. Univariate and multivariable linear regressions were used to evaluate the correlation between central SRF thickness and clinical parameters. Univariate linear regression was used to evaluate the correlation between CMT and clinical parameters. P values <0.05 were considered statistically significant.
3. Results
A total of 66 patients with DME were enrolled in this study. The baseline characteristics of the patients are shown in Table 1. None of the patients had undergone dialysis. Of the total patients, 39 (59%) were women. The mean age was 58.2 ± 11.2 years, the mean duration of diabetes was 10.6 ± 6.9 years, the mean HbA1c level was 8.0 ± 1.7%, and the mean CMT was 523.2 ± 154 μm. SRF in DME was observed in 30 patients.
Table 1.
Baseline clinical parameters of diabetic macular edema.
| Parameters | Diabetic macular edema (n = 66) |
|---|---|
| Age (years) | 58.2 ± 11.2 (32–86) |
| Female (no) | 39 |
| Diabetes duration (years) | 10.6 ± 6.9 (0.25–30) |
| Hypertension (patients) | 36 |
| Fasting glucose level (mmol/L) | 8.49 ± 3.32 (3.35–21.05) |
| HbA1c (%) | 8.0 ± 1.7 (5.4–11.7) |
| Triglyceride (mmol/L) | 1.9 ± 1.8 (0.71–14.6) |
| Total cholesterol (mmol/L) | 4.7 ± 1.3 (2.34–7.98) |
| Serum albumin (g/L) | 41.0 ± 3.7 (32.7–48.3) |
| Serum creatinine (mg/dL) | 1.3 ± 1.3 (0.5–10.69) |
| eGFR (ml/min/1.73 m2) | 64.8 ± 24.6 (4–122) |
| Urinary albumin creatinine ratio (mg/g) | 194.1 ± 223.7 (2.4–1201.5) |
| Disorganization of the retinal inner layers (patients) | 24 |
| Loss of the ellipsoid zone (patients) | 6 |
| CMT (μm) | 523.2 ± 154 (155–843) |
| Subretinal fluid (patients) | 30 |
| Diabetic retinopathy severity | |
| Mild NPDR | 0 |
| Moderate NPDR | 16 |
| Severe NPDR | 25 |
| PDR | 25 |
HbA1c: glycated hemoglobin; eGFR: estimated glomerular filtration rate; CMT: central macular thickness; NPDR: non-proliferative diabetic retinopathy; PDR: proliferative diabetic retinopathy.
Table 2 shows the factors associated with the presence of SRF. Binary logistic regression analysis revealed that lower HbA1c levels (odds ratio [OR], 0.727; 95% confidence interval [CI], 0.529–0.999; P=0.049) and higher albuminuria (OR, 3.571; 95% CI, 1.112–11.468; P=0.032) were associated with the presence of SRF among the 66 patients with DME. Multiple regression analysis revealed that only albuminuria was significantly associated (OR, 3.431; 95% CI, 1.039–11.334; P=0.043) with the presence of SRF. Visual acuity, DR severity, hypertension, serum fasting glucose, triglycerides, total cholesterol, eGFR, serum albumin, and serum creatinine showed no significant association with the presence of SRF. Table 3 shows the factors associated with SRF thickness according to the univariate linear regression among the 30 patients with SRF. Lower serum albumin levels (beta = −14.028, se = 6.646, P=0.044) were associated with a greater SRF thickness.
Table 2.
Correlating factors for presence of subretinal fluid in patients with diabetic macular edema.
| Parameters | Binary logistic regression analysis | Multiple logistic regression analysis | ||||
|---|---|---|---|---|---|---|
| OR | 95% CI | P value | OR | 95% CI | P value | |
| Age (years) | 0.970 | 0.927–1.014 | 0.181 | — | — | — |
| Diabetes duration (years) | 0.964 | 0.896–1.038 | 0.33 | — | — | — |
| Hypertension | 1.5 | 0.563–3.997 | 0.417 | — | — | — |
| Visual acuity | 2.62 | 0.853–8.05 | 0.093 | — | — | — |
| Fasting glucose level (mmol/L) | 0.889 | 0.751–1.051 | 0.169 | — | — | — |
| HbA1c (%) | 0.727 | 0.529–0.999 | 0.049 | 0.730 | 0.530–1.006 | 0.055 |
| Triglyceride (mmol/L) | 1.452 | 0.849–2.481 | 0.173 | — | — | — |
| Total cholesterol (mmol/L) | 0.897 | 0.621–1.295 | 0.561 | — | — | — |
| Serum albumin (g/L) | 0.951 | 0.831–1.089 | 0.469 | — | — | — |
| Serum creatinine (mg/dL) | 2.023 | 0.731–5.598 | 0.175 | — | — | — |
| Albuminuria | 3.571 | 1.112–11.468 | 0.032 | 3.431 | 1.039–11.334 | 0.043 |
| eGFR | 1.336 | 0.877–2.035 | 0.177 | — | — | — |
| DR severity | 0.991 | 0.531–1.849 | 0.977 | — | — | — |
Table 3.
Univariate linear regression for correlating factors of subretinal fluid thickness.
| Parameters | Beta | se | P value |
|---|---|---|---|
| Age (years) | 1.299 | 2.487 | 0.606 |
| Diabetes duration (years) | −1.157 | 5.770 | 0.843 |
| Fasting glucose level (mmol/L) | −11.242 | 12.289 | 0.368 |
| HbA1c (%) | −30.529 | 16.591 | 0.077 |
| Serum creatinine (mg/dL) | −1.07 | 15.695 | 0.946 |
| Serum albumin (g/L) | −14.028 | 6.646 | 0.044 |
| Total cholesterol (mmol/L) | −35.112 | 19.773 | 0.087 |
| Triglyceride (mmol/L) | −1.833 | 11.203 | 0.871 |
| Albuminuria | 53.360 | 75.140 | 0.483 |
| Hypertension | −48.972 | 56.926 | 0.397 |
| eGFR | 14.962 | 20.895 | 0.480 |
| Visual acuity | 81.498 | 58.892 | 0.177 |
| DR severity | 11.543 | 39.289 | 0.771 |
Table 4 shows the factors associated with CMT. Univariate linear regression revealed that HbA1c (beta = −42.17, se = 10.34, P=0.0001) and serum fasting glucose (beta = −13.24, se = 5.55, P=0.020) were associated with CMT among the 66 patients with DME. Multivariable linear regression revealed that a lower HbA1c was significantly associated with a higher CMT (beta = −42.06, se = 13.28, P=0.002). However, albuminuria, eGFR, serum albumin, and serum creatinine were not associated with CMT.
Table 4.
Correlating factors for CMT in patients with diabetic macular edema.
| Parameters | Univariate linear regression | Multivariable linear regression | ||||
|---|---|---|---|---|---|---|
| Beta | se | P value | Beta | se | P value | |
| Age (years) | −1.315 | 1.706 | 0.444 | — | — | — |
| Diabetes duration (years) | −3.585 | 2.753 | 0.198 | — | — | — |
| Fasting glucose level (mmol/L) | −13.24 | 5.55 | 0.020 | −0.09 | 6.69 | 0.989 |
| HbA1c (%) | −42.17 | 10.34 | 0.0001 | −42.06 | 13.28 | 0.002 |
| Serum creatinine (mg/dL) | 15.372 | 14.944 | 0.308 | — | — | — |
| Serum albumin (g/L) | −1.508 | 5.332 | 0.778 | — | — | — |
| Total cholesterol (mmol/L) | −25.39 | 13.93 | 0.073 | — | — | — |
| Triglyceride (mmol/L) | 2.4 | 10.48 | 0.820 | — | — | — |
| Albuminuria | 14.846 | 41.528 | 0.722 | — | — | — |
| Hypertension | −0.25 | 38.366 | 0.995 | — | — | — |
| eGFR | 6.523 | 16.118 | 0.687 | — | — | — |
| DR severity | 12.426 | 24.560 | 0.615 | — | — | — |
4. Discussion
Our study results suggest that diabetic kidney disease plays an important role in the occurrence of SRF in DME. Higher albuminuria showed a better association with the presence of SRF than lower HbA1c levels, although both were important risk factors for the presence of SRF.
Our data suggested that the presence of SRF had a good correlation with higher albuminuria, while serum albumin was not significantly different between the patients with or without SRF. In addition, we found that the severity of SRF was negatively correlated with serum albumin levels. Low serum albumin levels were associated with increased SRF thickness. Tsai et al. [7] reported that the presence of SRF was correlated with low serum albumin levels. This may be because the correlation of baseline SRF with serum albumin levels was performed to adjust for age and DR severity. Koo et al. [8] also reported a significantly higher frequency of SRF in DME in patients with albuminuria. Vascular hyperpermeability is a possible shared pathogenetic mechanism between the kidney (albuminuria) and the eye (SRF of DME). Fluid movement is governed by changes in pressure gradients. Advanced proteinuria with marked protein loss may result in lower intravascular osmotic pressure and higher hydrostatic pressure, leading to fluid retention in the subretinal space. In the early stages, serum albumin deficits may be compensated for by the increased production of albumin molecules in the liver, which may prevent hypoalbuminemia [10]. In the late stage, serum albumin deficits that could not be compensated for may be associated with increased SRF thickness. Additionally, according to our results, no associations could be found between albuminuria and CMT, which is similar to the findings of previous studies [11–13]. We inferred that patients with albuminuria may be prone to more fluid leakage from choroidal vessels and damaged retinal pigment epithelium, which eventually results in SRF in DME. Moreover, previously reported case series on SRF secondary to nephrotic syndrome demonstrated that SRF resolved following systemic furosemide treatment [14–16].
Contrary to the general belief [17], we observed that lower HbA1c levels were more likely to result in SRF and were associated with a higher CMT. Data from a more recent study also indicated that lower HbA1c levels are more likely to result in SRF in DME [11]. Other supportive evidence showed that better control of HbA1c did not lead to a greater reduction in CMT in DME [18]. Furthermore, Falavarjani et al. [19] observed reduced CMT after meals compared with that before meals. There are several possible explanations for this finding. First, it may be because of the early worsening effect, which can be caused by the rapid reduction in HbA1c levels [11]. Second, lowering serum glucose levels can lower the intravascular osmotic pressure. We inferred that the pressure gradient may have increased the tendency of the fluid to move from the intravascular to the retinal parenchyma and subretinal space. Thus, tight control of HbA1c may be associated with higher CMT and the presence of SRF. However, further research with a larger sample size is needed to clarify this.
In the present study, no associations were found between eGFR and the severity spectrum of DME or the presence of SRF. Previous studies have also shown that eGFR does not seem to have any effect on the severity or pattern of DME [5, 6]. However, Man et al. [2] and Temkar et al. [5] found that lower eGFR was not associated with the presence or severity of DME. Additionally, we did not find any significant correlation between DR severity and the presence of SRF. This may be because retinopathy and nephropathy may have a similar pathology, and eGFR and DR may fluctuate over time. However, further research is required to clarify this.
Our study has some limitations. It was a retrospective study with a small sample size.
5. Conclusions
According to our study results, higher albuminuria was associated with the presence of SRF in patients with DME. Lower serum albumin levels were associated with greater SRF thickness. Hence, screening for SRF in DME in all patients with diabetic kidney disease should be emphasized.
Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
References
- 1.Lee R., Wong T. Y., Sabanayagam C. Epidemiology of diabetic retinopathy, diabetic macular edema and related vision loss. Eye and Vision . 2015;2(1):p. 17. doi: 10.1186/s40662-015-0026-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Man R. E. K., Sasongko M. B., Wang J. J., et al. The association of estimated glomerular filtration rate with diabetic retinopathy and macular edema. Investigative Ophthalmology & Visual Science . 2015;56(8):4810–4816. doi: 10.1167/iovs.15-16987. [DOI] [PubMed] [Google Scholar]
- 3.Hammes H. P., Welp R., Kempe H. P., et al. Risk factors for retinopathy and DME in type 2 diabetes-results from the German/Austrian DPV database. PLoS One . 2015;10(7) doi: 10.1371/journal.pone.0132492.e0132492 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Hsieh Y. T., Tsai M. J., Tu S. T., Hsieh M. C. Association of abnormal renal profiles and proliferative diabetic retinopathy and diabetic macular edema in an asian population with type 2 diabetes. JAMA Ophthalmol . 2018;136(1):68–74. doi: 10.1001/jamaophthalmol.2017.5202. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Temkar S., Karuppaiah N., Takkar B., et al. Impact of estimated glomerular filtration rate on diabetic macular edema. International Ophthalmology . 2018;38(3):1043–1050. doi: 10.1007/s10792-017-0557-8. [DOI] [PubMed] [Google Scholar]
- 6.Jeng C. J., Hsieh Y. T., Yang C. M., Yang C. H., Lin C. L., Wang I. J. Diabetic retinopathy in patients with diabetic nephropathy: development and progression. PLoS One . 2016;11(8) doi: 10.1371/journal.pone.0161897.e0161897 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Tsai M. J., Hsieh Y. T., Shen E. P., Peng Y. J. Systemic associations with residual subretinal fluid after ranibizumab in diabetic macular edema. Journal of Ophthalmology . 2017;2017 doi: 10.1155/2017/4834201.4834201 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Koo N. K., Jin H. C., Kim K. S., Kim Y. C. Relationship between the morphology of diabetic macular edema and renal dysfunction in diabetes. Korean Journal of Ophthalmology . 2013;27(2):98–102. doi: 10.3341/kjo.2013.27.2.98. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Sun J. K., Lin M. M., Lammer J., et al. Disorganization of the retinal inner layers as a predictor of visual acuity in eyes with center-involved diabetic macular edema. JAMA Ophthalmol . 2014;132(11):1309–1316. doi: 10.1001/jamaophthalmol.2014.2350. [DOI] [PubMed] [Google Scholar]
- 10.Levitt D. G., Levitt M. D. Human serum albumin homeostasis: a new look at the roles of synthesis, catabolism, renal and gastrointestinal excretion, and the clinical value of serum albumin measurements. International Journal of General Medicine . 2016;9:229–255. doi: 10.2147/ijgm.s102819. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Klein R., Klein B. E., Moss S. E., Cruickshanks K. J. The Wisconsin epidemiologic study of diabetic retinopathy XV. Ophthalmology . 1995;102(1):7–16. doi: 10.1016/s0161-6420(95)31052-4. [DOI] [PubMed] [Google Scholar]
- 12.Kiire C. A., Porta M., Chong V. Medical management for the prevention and treatment of diabetic macular edema. Survey of Ophthalmology . 2013;58(5):459–465. doi: 10.1016/j.survophthal.2012.10.002. [DOI] [PubMed] [Google Scholar]
- 13.Aroca P. R., Espeso Sentís O., Del Castillo D. D. Prospective study of correlation between diabetic retinopathy and microalbuminuria in diabetes type 1 patients. Archivos de la Sociedad Espanola de Oftalmologia . 2000;75(5):307–314. [PubMed] [Google Scholar]
- 14.Wong M. H. Y., Cheung G. C. M., Chee S. P., Mathur R. Exudative detachment as a masquerader in hypoalbuminaemic patients. Clinical Kidney Journal . 2014;7(4):406–410. doi: 10.1093/ckj/sft161. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Koo N. K., Kim Y. C. Resolution of macular edema after systemic treatment with furosemide. Korean Journal of Ophthalmology . 2012;26(4):312–315. doi: 10.3341/kjo.2012.26.4.312. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Ciardella A. P. Partial resolution of diabetic macular oedema after systemic treatment with furosemide. British Journal of Ophthalmology . 2004;88(9):1224–1225. doi: 10.1136/bjo.2004.042580. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Romero P., Baget M., Mendez I., Fernandez J., Salvat M., Martinez I. Diabetic macular edema and its relationship to renal microangiopathy: a sample of Type I diabetes mellitus patients in a 15-yearfollow-up study. Journal of Diabetes and Its Complications . 2007;21(3):172–180. doi: 10.1016/j.jdiacomp.2006.07.008. [DOI] [PubMed] [Google Scholar]
- 18.Bansal A. S., Khurana R. N., Wieland M. R., Wang P. W., Van Everen S. A., Tuomi L. Influence of glycosylated hemoglobin on the efficacy of ranibizumab for diabetic macular edema: a post hoc analysis of the RIDE/RISE trials. Ophthalmology . 2015;122(8):1573–1579. doi: 10.1016/j.ophtha.2015.04.029. [DOI] [PubMed] [Google Scholar]
- 19.Falavarjani K. G., Khadamy J., Amirkourjani F., Safi H., Modarres M. Macular thickness measurement in clinically significant macular edema before and after meal. Journal of Current Ophthalmology . 2015;27(3-4):125–128. doi: 10.1016/j.joco.2015.11.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data used to support the findings of this study are available from the corresponding author upon request.