The predictive value of diabetic retinopathy on subsequent diabetic nephropathy in patients with type 2 diabetes: a systematic review and meta-analysis of prospective studies

Figures & data

Figure 1. Flowchart of literature search and the selection process of the studies.

Table 1. Baseline characteristic of included studies.

Study	Country	Sample size	Mean age (years)	Percentage male (%)	Duration of diabetes	DR diagnostic criteria	DN diagnostic criteria	True positive	False positive	False negative	True negative	NOS score
[19]	India	80	47.4	60.0	Not mentioned	Fundoscopy by a qualified ophthalmologist using a clinical ophthalmoscope	Diffuse intercapillary glomerulosclerosis	6	7	8	53	7
[20]	Italy	53	NA	66.0	7–14 years	Ophthalmoscopy after mydriasis	‘typical’ diabetic glomerulopathy by light microscopy	14	15	0	24	8
[21]	China	51	54.7	70.6	7.2 years	Fundoscopy by the attending physician or ophthalmologist with pupils adequately dilated: presence of background retinopathy with or without proliferative changes	The presence of mesangial expansion and diffuse intercapillary glomerulosclerosis	20	10	14	7	6
[22]	Denmark	34	52.5	94.1	13–18 years	Direct ophthalmoscopy	Biopsy-proven diabetic glomerulosclerosis	17	0	9	8	6
[23]	Japan	109	48.1	67.9	7.3 years	Direct ophthalmoscopy with pupils dilated by an ophthalmologist	Pathological evaluation of renal biopsy specimens, including light microscopy, electron microscopy, and immunofluorescence staining	46	8	24	20	7
[24]	Denmark	49	54.7	81.6	5.0 years	Direct ophthalmoscopy after pupillary dilation and afterwards by fundus photography and graded: nil, simplex, or proliferative retinopathy	Diffuse diabetic glomerulosclerosis, hyaline arteriolosclerosis, tubular atrophy, and interstitial fibrosis	17	11	17	4	6
[25]	China	68	48.7	55.9	6.4 years	Presence of proliferative or background changes. About 60–70% of patients were assessed by an ophthalmologist and the rest by physician alone.	Presented 3 of following features: morphological criteria of diabetic glomerular lesions included glomerular Hyaline arteriolosclerosis; global mesangial sclerosis with or without Kimmelstiel–Wilson nodule or nodular mesangial sclerosis; exudative lesions such as ‘fibrin cap’, ‘capsular drop’, or ‘hyaline thrombus’; microaneurysm; uniform glomerular capillary basement membrane thickening; or ultrastructural thickness 350 nm	17	8	7	36	7
[26]	Spain	35	59.0	62.9	Not mentioned	Formal ophthalmological examination and fluorescein angiography was performed in all the patients.	Glomerulosclerosis either of the nodular or diffuse types, interstitial fibrosis with infiltration by mononuclear inflammatory cells, hyalinization of the renal arterioles, linear deposits of Igs (Immunoglobulins) in glomerular basement membrane on immunofluorescence studies and diffuse thickening of glomerular basement membrane on electron microscopy examination	9	3	17	6	7
[27]	China	110	46.3	69.7	5.0 years	Not mentioned	Diffuse mesangial expansion with predominance of increased mesangial matrix, Kimmelstiel–Wilson nodular lesions, hyaline exudative lesions, and glomerular basement membrane thickening	46	5	14	45	8
[28]	Iran	46	48.9	41.3	7.2 years	Background ± proliferative changes on funduscopy and/or fluorescein angiography	Kidney biopsy samples were examined by light microscopy, immunofluorescence, and electron microscopy	17	1	9	19	6

Figure 2. The summary sensitivity and specificity of DR on subsequent DN in patients with T2D.

Table 2. Subgroup analyses.

Parameters	Factors	Subgroup	Effect estimate and 95%CI	I^2 (%)	p Value for Q statistic	Ratio between subgroups	p Value between subgroups
Sensitivity	Country	Europe	0.67 (0.32–0.90)	83.4	<.01	1.00 (0.59–1.69)	1.000
		Asia	0.67 (0.60–0.74)	29.5	.21
	Mean age (years)	≥50.0	0.53 (0.42–0.63)	47.5	.13	0.78 (0.62–0.98)	.032
		<50.0	0.68 (0.61–0.75)	38.3	.17
	Percentage male (%)	≥65.0	0.67 (0.58–0.75)	66.7	.01	1.24 (0.90–1.72)	.193
		<65.0	0.54 (0.38–0.69)	65.7	.03
	Study quality	High	0.69 (0.45–0.85)	80.1	<.01	1.15 (0.80–1.66)	.452
		Low	0.60 (0.49–0.70)	0.0	.51
Specificity	Country	Europe	0.65 (0.32–0.88)	76.0	.01	0.79 (0.47–1.34)	.389
		Asia	0.82 (0.67–0.91)	80.2	<.01
	Mean age (years)	≥50.0	0.62 (0.24–0.89)	76.6	.01	0.72 (0.37–1.39)	0.331
		<50.0	0.86 (0.79–0.91)	48.0	.10
	Percentage male (%)	≥65.0	0.67 (0.43–0.85)	84.7	<.01	0.78 (0.55–1.10)	.161
		<65.0	0.86 (0.78–0.91)	38.1	.18
	Study quality	High	0.81 (0.71–0.88)	68.5	.01	1.04 (0.53–2.02)	.912
		Low	0.78 (0.26–0.97)	88.4	<.01
PLR	Country	Europe	1.93 (0.67–5.56)	72.0	<.01	0.52 (0.15–1.83)	.308
		Asia	3.72 (1.87-7.38)	71.1	<.01
	Mean age (years)	≥50.0	1.38 (0.46–4.12)	15.5	.06	0.29 (0.09–0.94)	.038
		<50.0	4.79 (3.12–7.36)	4.2	.08
	Percentage male (%)	≥65.0	2.02 (0.95–4.28)	79.8	<.01	0.53 (0.21–1.35)	.185
		<65.0	3.78 (2.20–6.50)	11.9	.07
	Study quality	High	3.54 (2.35–5.32)	12.5	.05	1.32 (0.19–9.03)	.780
		Low	2.69 (0.41–17.70)	79.6	<.01
NLR	Country	Europe	0.50 (0.16–1.58)	79.7	<.01	1.25 (0.38–4.07)	.711
		Asia	0.40 (0.30–0.53)	63.8	.02
	Mean age (years)	≥50.0	0.77 (0.37–1.60)	77.2	<.01	2.08 (0.96–4.50)	.062
		<50.0	0.37 (0.29–0.47)	57.1	.05
	Percentage male (%)	≥65.0	0.50 (0.29–0.84)	78.6	<.01	0.94 (0.49–1.80)	0.860
		<65.0	0.53 (0.37–0.77)	66.8	.03
	Study quality	High	0.39 (0.20–0.73)	73.5	<.01	0.75 (0.29–1.92)	.548
		Low	0.52 (0.26–1.01)	79.3	<.01
DOR	Country	Europe	3.57 (0.35–36.20)	81.0	<.01	0.46 (0.04–5.85)	.553
		Asia	7.68 (2.78–21.24)	73.2	<.01
	Mean age (years)	≥50.0	1.19 (0.32–4.38)	62.0	.05	0.11 (0.02–0.50)	.004
		<50.0	11.06 (4.97–24.59)	50.8	.09
	Percentage male (%)	≥65.0	5.20 (1.09–24.77)	85.9	<.01	0.81 (0.11–5.94)	.835
		<65.0	6.43 (1.86–22.22)	63.1	.04
	Study quality	High	7.81 (3.06–19.95)	65.9	.01	2.26 (0.22–23.19)	.493
		Low	3.46 (0.41–29.20)	83.6	<.01
AUC	Country	Europe	0.71 (0.67–0.75)	–	–	0.97 (0.90–1.05)	.492
		Asia	0.73 (0.68–0.76)	–	–
	Mean age (years)	≥50.0	0.55 (0.50–0.59)	–	–	0.71 (0.64–0.77)	<.001
		<50.0	0.78 (0.74–0.81)	–	–
	Percentage male (%)	≥65.0	0.70 (0.66–0.74)	–	–	0.81 (0.76–0.87)	<.001
		<65.0	0.86 (0.82–0.88)	–	–
	Study quality	High	0.83 (0.79–0.86)	–	–	1.32 (1.22–1.42)	<.001
		Low	0.63 (0.59–0.67)	–	–

Figure 3. The summary PLR and NLR of DR on subsequent DN in patients with T2D.

Figure 4. The summary DOR of DR on subsequent DN in patients with T2D.

Figure 5. The summary AUC on subsequent DN in patients with T2D.

Figure 6. Publication bias.

John GT, Date A, Korula A, et al. Nondiabetic renal disease in noninsulin-dependent diabetics in a south Indian Hospital. Nephron. 1994;67(4):441–443.

 PubMed Web of Science ®Google Scholar

Brocco E, Fioretto P, Mauer M, et al. Renal structure and function in non-insulin dependent diabetic patients with microalbuminuria. Kidney Int Suppl. 1997;63:S40–S44.

 PubMedGoogle Scholar

Mak SK, Gwi E, Chan KW, et al. Clinical predictors of non-diabetic renal disease in patients with non-insulin dependent diabetes mellitus. Nephrol Dial Transplant. 1997;12(12):2588–2591.

 PubMed Web of Science ®Google Scholar

Christensen PK, Gall MA, Parving HH. Course of glomerular filtration rate in albuminuric type 2 diabetic patients with or without diabetic glomerulopathy. Diabetes Care. 2000;23(2):B14–20.

 PubMedGoogle Scholar

Suzuki D, Takano H, Toyoda M, et al. Evaluation of renal biopsy samples of patients with diabetic nephropathy. Intern Med. 2001;40(11):1077–1084.

 PubMed Web of Science ®Google Scholar

Christensen PK, Larsen S, Horn T, et al. Renal function and structure in albuminuric type 2 diabetic patients without retinopathy. Nephrol Dial Transplant. 2001;16(12):2337–2347.

 PubMed Web of Science ®Google Scholar

Wong TY, Choi PC, Szeto CC, et al. Renal outcome in type 2 diabetic patients with or without coexisting nondiabetic nephropathies. Diabetes Care. 2002;25(5):900–905.

 PubMed Web of Science ®Google Scholar

Serra A, Romero R, Bayés B, et al. Is there a need for changes in renal biopsy criteria in proteinuria in type 2 diabetes? Diabetes Res Clin Pract. 2002;58(2):149–153.

 PubMed Web of Science ®Google Scholar

Zhou J, Chen X, Xie Y, et al. A differential diagnostic model of diabetic nephropathy and non-diabetic renal diseases. Nephrol Dial Transplant. 2008;23(6):1940–1945.

 PubMed Web of Science ®Google Scholar

Soleymanian T, Hamid G, Arefi M, et al. Non-diabetic renal disease with or without diabetic nephropathy in type 2 diabetes: clinical predictors and outcome. Ren Fail. 2015;37(4):572–575.

 PubMed Web of Science ®Google Scholar

Data availability statement

The authors confirm that the data supporting the findings of this study are available within the article.