The effects of omega-3 fatty acids on diabetic nephropathy: A meta-analysis of randomized controlled trials.

OBJECTIVE: To evaluate the effects of omega-3 long-chain polyunsaturated fatty acids on proteinuria, estimated glomerular filtration rate (eGFR) and metabolic biomarkers among patients with diabetes. STUDY
DESIGN: Meta-analysis of randomized controlled clinical trials (RCTs). SETTING &
SUBJECTS: Patients with diabetes. SELECTION CRITERIA FOR STUDIES: We conducted electronic searches in PubMed, Embase and Cochrane Central Register of Controlled Trials from January 1960 to April 2019 to identify RCTs, which examined the effects of omega-3 fatty acids on proteinuria, eGFR and metabolic biomarkers among diabetic patients.
RESULTS: Ten RCTs with 344 participants were included in our meta-analysis. Omega-3 fatty acids reduced the amount of proteinuria among type 2 diabetes mellitus (type 2 DM) and type 1 diabetes mellitus (type 1 DM). This association was only significant among type 2 DM (SMD = -0.29 (95% CI: -0.54, -0.03; p = 0.03). Only studies with duration of intervention of 24 weeks or longer demonstrated a significant lower proteinuria among omega-3 fatty acids compared to control group (SMD = -0.30 (95% CI: -0.58, -0.02; p = 0.04). There was a higher eGFR for both type 1 and type 2 DM groups among omega-3 fatty acids compared to control group, however, the effect was not statistically significant. Regarding serum total cholesterol, LDL-cholesterol and HbA1C, there was no significant difference comparing omega-3 fatty acids to control group. There was a non-significant systolic blood pressure reduction in the omega-3 fatty acids supplementation group compared to control.
CONCLUSION: Omega-3 fatty acids could help ameliorate proteinuria among type 2 DM who received omega-3 supplementation for at least 24 weeks without adverse effects on HbA1C, total serum cholesterol and LDL-cholesterol.

Introduction

The prevalence of diabetes around the world has reached an unprecedented level in recent decades. While diabetes is already estimated to afflict more than 350 million people around the world, this is predicted to grow to over 550 million people by the year 2035[1, 2]. More importantly, 30–40% of patients with diabetes mellitus will develop diabetic nephropathy[2] which is characterized by proteinuria in advanced stages. The degree of proteinuria reflects the severity of glomerular damage and is associated with a faster decline in the estimated glomerular filtration rate (eGFR) [3-5]. Additionally, proteinuria in this population is associated with hyperuricemia, stroke, and cardiovascular disease morbidity/mortality [5-8].

Long-chain omega-3 polyunsaturated fatty acids, including eicosapentaenoic acid (EPA) and docosahexaenoic acids (DHA), have shown anti-inflammatory, antithrombotic properties and benefits on kidney function[9-11]. There is a number of clinical trials studying in various types of kidney diseases including IgA nephropathy[12], lupus nephritis[13, 14] and polycystic kidney disease[15]. However, the information about the effects of omega-3 fatty acids on kidney function, particularly in diabetic kidney disease still lacks consensus. Currently, the data from Diabetes Control and Complications Trial showed that higher dietary eicosapentaenoic acid and docosahexaenoic acid consumption was associated with a lower risk of proteinuria among diabetic patients[16]. Nonetheless, the meta-analysis on the effect of n–3 long-chain polyunsaturated fatty acid supplementation on urine protein excretion and kidney function by Miller et al.[17] in 2009 suggested that there was no sufficient evidence to conclude that n–3 long-chain polyunsaturated fatty acid supplementation could reduce albuminuria among diabetic patients subgroup (7 studies, 222 patients). Since then, 3 new studies were published including 344 patients (55% increases in sample size). Moreover, another meta-analysis on omega-3 fatty acid supplementation as adjunctive therapy in the treatment of chronic kidney disease by Jing et al.[11] in 2017 suggested that omega-3 fatty acid supplementation is associated with a significantly reduced risk of end-stage renal disease and delays the progression of this disease, but in this study, diabetic patients were not included.

The aim of this meta-analysis was to investigate the effects of omega-3 fatty acid supplementation in reducing proteinuria in diabetic patients by using all available evidence from the published literature. All eligible studies assessed proteinuria, the serum creatinine clearance rate, the estimated glomerular filtration rate, or the occurrence of end-stage renal disease.

Methods

Data sources and searches

The protocol for this systematic review is registered with PROSPERO (International Prospective Register of Systematic Reviews; no.CRD42019134873). We conducted electronic searches in PubMed, Embase and Cochrane Central Register of Controlled Trials from January 1960 to April 2019 to identify randomized controlled trials (RCTs), which explored the effects of omega-3 fatty acid supplementation on proteinuria, eGFR and metabolic biomarkers among diabetic patients. The same search strategy was used for EMBASE and Cochrane Central Register of Controlled Trials using the corresponding terms. Manual searches of the reference lists from all relevant original and review articles were also conducted to identify additional eligible studies. This study was conducted by the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement[18].

Selection criteria

RCTs examining the effect of omega-3 fatty acid supplementation compared to control on proteinuria or albuminuria were included. There were no restrictions on sample size or study duration. Retrieved articles were individually reviewed for eligibility by two investigators (A.C. and A.R.). Disagreements were addressed and solved by mutual consensus.

Data extraction and quality assessment

The following data were extracted: study design, year of publication, country of origin, sample size, duration of follow-up, type of omega-3 fatty acid, dose, frequency, mean age and type of diabetes. The following outcomes of interest were examined: change in kidney outcomes [proteinuria and eGFR], serum lipids and glucose control biomarkers [triglyceride, total cholesterol (TC), high density lipoprotein (HDL), low density lipoprotein (LDL), hemoglobin A1C (HbA1C]) and blood pressure parameters [systolic blood pressure (SBP), diastolic blood pressure (DBP)] between baseline and at the study end.

Revised Cochrane risk-of-bias tool for randomized trials (RoB 2)[19] was used to evaluate the risk of bias for RCTs. The assessment included the following components: risk of bias arising from randomization process, risk of bias due to deviation from the intended interventions, missing outcome data, risk of bias in measurement of the outcome and risk of bias in the selection of the reported result. A judgment about the risk of bias arising from each domain is generated by an algorithm, based on answers to the signaling questions. Judgment could be high risk of bias, low risk of bias, or some concerns.

Data synthesis and statistical analysis

Random effects models were used due to the expected clinical heterogeneity in the included populations. We also compared the results with the fixed effect model. Adjusted point estimates from each study were consolidated by the generic inverse variance approach of DerSimonian and Laird, which designated the weight of each study based on its variance[20]. We also applied fixed effects models to compare the results. We computed standardized mean difference (SMD) in mean values for proteinuria at the study end because this particular outcome was measured on a different scale across studies. However, for other continuous variables that were measured on the same scale, we used weight mean difference (WMD) for the mean values at the study end. We assumed that there were no significant differences in baseline characteristics for each variable in randomized controlled trials. All pooled estimates were displayed with 95% confidence intervals (CI). Heterogeneity among effect sizes estimated by individual studies was described with the I2 index and the chi-square test. A value of I2 of 0%-25% represents insignificant heterogeneity, 26%-50% low heterogeneity, 51%-75% moderate heterogeneity and 76–100% high heterogeneity[21]. Meta-regression was used to assess the association between change in proteinuria and change in eGFR as well as the change in proteinuria and combined dose of DHA and EPA.

Publication bias was formally assessed using funnel plots and the Egger test to assess for asymmetry of the funnel plot. A p-value of less than 0.05 indicates the presence of publication bias[22]. The meta-analysis was performed by STATA/IC 15.1 (StataCorp LLC, Texas, USA).

Results

Characteristics and quality of the studies

A total of 1,277 potentially relevant citations were identified and screened. Seventy citations were evaluated in detail, of which 10 trials [23-32] with 344 participants fulfilled the eligibility criteria and were included in this meta-analysis. The literature retrieval, review, and selection process are demonstrated in Fig 1.

Fig 1

Search methodology and selection process.

Characteristics of the individual trials are displayed in Table 1. Briefly, the trials varied in sample size from 18 to 79 patients. From 10 trials, three followed a cross-over design[24, 25, 27]. There were 3 trials conducted in North America [23-25], 3 trials conducted in Europe [26, 27, 29], 3 trials conducted in Asia [28, 31, 32] and 1 trial conducted in Australia [30]. There were 5 trials that included only type 2 DM [23, 24, 27, 28, 32], 3 trials that included only type 1 DM [25, 26, 29] and 2 trials that included both type 1 and type 2 DM [30, 31]. However, only one from those two studies reported outcomes in each group separately[31]. For the analysis purposes, the study by Hamazaki et al. was divided into 2 separate studies based on the type of DM. Therefore, we had 11 study arms from 10 original studies and no duplicate populations. The mean age of patients ranged from 33 to 67.4 years old. The duration of follow up spanned from 6 weeks to 52 weeks.

Table 1

Main characteristics of studies included in the meta-analysis of the effects of omega-3 fatty acids on proteinuria among patients with diabetes.

Author	Country	Number of patients	Mean age	% Female		Fish oil		Control	Follow-up	Type of DM
					Baseline eGFR (ml/min/1.73 m2)	EPA	DHA
Haines 1986 [26]	UK	41	42.3	26.8	NA	2.7 g	1.9 g	olive oil	6 wk	Type 1
Jensen 1989 [25]	USA	18	37.0	22.2	82±5	2 g	2.6 g	olive oil	8 wk	Type 1
Hamazaki 1990 [31]	Japan	9	59.3	55.6	NA	1.8 g	-	no omega-3 supplementation	24 wk	Type 1
		17	66.0	58.8	NA	1.8 g	-	no omega-3 supplementation	24 wk	Type 2
Shimizu 1995 [28]	Japan	45	63.6	51.2	NA	900 mg	-	healthy	52 wk	Type 2
Rossing 1996 [29]	Denmark	29	33.0	34.5	116±7	2.0 g	2.6 g	olive oil	52 wk	Type 1
Lungershausen 1997 [30]	Australia	32	55.0	25.0	116±11	2.0 g	1.4 g	corn oil	12 wk	Type 1 andtype 2
Zeman 2005 [27]	Czech	24	48.8	45.8	NA	2.07 g	1.53 g	olive oil	52 wk	Type 2
Miller 2013 [24]	USA	31	67.4	45.1	78±22	2.26 g	1.13 g	placebo	8 wk	Type 2
Lee 2015 [32]	Korea	19	60.4	36.8	58±8	1.38 g	1.14 g	olive oil	12 wk	Type 2
Elajami 2017 [23]	USA	79	63.4	18.9	79±22	1.86 g	1.5 g	no omega-3 supplementation	52 wk	Type 2

Risk of bias

According to the revised Cochrane risk-of-bias tool for randomized trials, with respect to the overall risk of bias, five studies had low risk of bias [24, 25, 29, 30, 32]; one study with some concerns for risk of bias [23] and another four studies had high risk of bias [26–28, 31]. In terms of risk of bias arising from the randomization process, four studies had high risk of bias [26–28, 31]. For risk of bias due to deviations from the intended interventions, five studies raised some concerns [23, 26–28, 31]. Five studies raised some concerns for missing outcome data and risk of bias in selection of the reported result [23, 26–28, 31]. All of the studies had low risk of bias in the measurement of the outcome. There were five studies that had some concerns for the risk of bias in selection of the reported result [23, 26–28, 31]. There was no study that had high risk of bias in all domains (Table 2).

Table 2

Risk of bias according to revised Cochrane risk-of-bias tool for randomized trials.

	Risk of bias arising from the randomization process	Risk of bias due to deviations from the intended interventions	Missing outcome data	Risk of bias in measurement of the outcome	Risk of bias in selection of the reported result	Overall risk of bias
Haines 1986 [26]	High	Some concerns	Some concerns	Low	Some concerns	high
Jensen 1989 [25]	Low	Low	Low	Low	Low	Low
Hamazaki 1990 [31]	High	Some concerns	Some concerns	Low	Some concerns	high
Shimizu 1995 [27]	High	Some concerns	Some concerns	Low	Some concerns	high
Rossing 1996 [29]	Low	Low	Low	Low	Low	Low
Lungershausen 1997 [30]	Low	Low	Low	Low	Low	Low
Zeman 2005 [27]	High	Some concerns	Some concerns	Low	Some concerns	high
Miller 2013 [24]	Low	Low	Low	Low	Low	Low
Lee 2015 [32]	Low	Low	Low	Low	Low	Low
Elajami 2017 [23]	Low	Some concerns	Some concerns	Low	Some concerns	Some concerns

Effect of omega-3 fatty acids on kidney outcomes

As shown in Fig 2 and Table 3, 11 study arms (342 patients) reported proteinuria as the primary outcome. We found that proteinuria among diabetic patients receiving omega-3 fatty acids was lower than control group (SMD = -0.19 (95% CI: -0.38, 0.01); p = 0.06, I2 = 0%) but this was not statistically significant. Six study arms (208 patients) showed a higher eGFR among omega-3 fatty acids group but the effect was not significant (WMD = 1.56 mL/min/1.73m2 (95% CI:-1.53, 4.65); p = 0.32, I2 = 5.6%).

Fig 2

Forest plots of the included studies assessing proteinuria among diabetic patients.

Table 3

Summary effects of omega-3 fatty acids on outcomes of interest among diabetic patients.

Outcomes	No of study arms	No of patients	Weighted mean difference/Standardized mean difference*	Confidence interval	I2	P-value
Proteinuria	11	342	-0.19*	(-0.38, 0.01)	0%	0.06
eGFR	6	208	1.56 mL/min/1.73 m2	(-1.53, 4.65)	5.6%	0.32
SBP	10	318	-2.10 mmHg	(-4.48, 0.28)	0%	0.08
DBP	10	318	1.04 mmHg	(-1.81, 3.89)	39.8%	0.48
Triglyceride	10	313	-24.24 mg/dL	(-36.40, -12.10)	0%	<0.001
TC	6	168	3.72 mg/dl	(-4.63, 12.06)	80.2%	0.38
HDL-c	6	242	4.57 mg/dl	(0.79, 8.34)	82.5%	0.02
LDL-c	6	215	2.29 mg/dL	(-2.45, 7.03)	0%	0.34
HbA1C	10	313	-0.03%	(-0.45, 0.39)	66.2%	0.89

eGFR, estimated glomerular filtration rate; TC, total cholesterol; LDL-c, low density lipoprotein cholesterol; HDL-c, high density lipoprotein cholesterol; HbA1C, hemoglobin A1C; SBP, systolic blood pressure; DBP, diastolic blood pressure

* indicates standardized mean differences

Effect of omega-3 fatty acids on blood pressure parameters

Ten study arms with 318 patients reported that there were no differences in both SBP (WMD = -2.10 mmHg (95% CI:-4.48, 0.28); P = 0.08, I2 = 0%), and DBP (WMD = 1.04 mmHg (95% CI:-1.81, 3.89); P = 0.48, I2 = 39.8%) between treatment group and control group as shown in Figs 3A and 3B.

Fig 3

a Forest plots of the included studies assessing systolic blood pressure among diabetic patients. b Forest plots of the included studies assessing diastolic blood pressure among diabetic patients.

Effect of omega-3 fatty acids on serum lipids and glucose control

Regarding triglycerides, ten study arms with 313 patients showed that omega-3 fatty acids significantly diminished triglycerides (WMD = -24.24 mg/dL (95% CI:-36.40, -12.10); P < 0.001, I2 = 0%). While in lights of total cholesterol, six study arms with 168 participants demonstrated no significant difference for total cholestrol between omega-3 fatty acids group and control (WMD = 3.72 mg/dl (95% CI:-4.63, 12.06); P = 0.38, I2 = 80.2%). In terms of serum LDL-cholesterol, six study arms with 215 patients demonstrated no significant difference in serum LDL-cholesterol (WMD = 2.29 mg/dL (95% CI:-2.45, 7.03); P = 0.34, I2 = 0%). However, for HDL-cholesterol, six study arms with 242 participants illustrated that omega-3 fatty acids group had a higher HDL-cholesterol compared to control group (WMD = 4.57 mg/dL (95% CI: 0.79, 8.34); P = 0.02, I2 = 82.5%). Moreover, ten study arms with 313 patients illustrated no significant difference in HbA1C between omega-3 fatty acids group and control group (WMD = -0.03% (95% CI: -0.45, 0.39); P = 0.89, I2 = 66.2%). Forrest plots were shown in

Fixed effects models

We also performed the analyses using fixed effects models. DBP, total cholesterol and HbA1C became significantly different between omega-3 fatty acids and control group as shown in . However, a random effects model will yield more conservative results than the fixed effect when tau2 is not equal to zero.

Subgroup analysis and meta-regression

In the subgroup analysis for type of DM, we excluded the study by Lungershausen et al.[30] since they did not provide separate results according to type of DM. Among type 2 DM group with 213 participants, omega-3 fatty acids could significantly reduce proteinuria (SMD = -0.29 (95% CI: -0.54, -0.03); P = 0.03, I2 = 3.9%) when compared to control group. However, among type 1 DM group with 97 participants, there was no significant difference in proteinuria (SMD = 0.01 (95% CI -0.36, 0.38); P = 0.95, I2 = 0%) between omega-3 fatty acids group and control group (Fig 4). For serum triglyceride, lower serum triglyceride was found among omega-3 fatty acids group in both type 1 diabetes with 97 participants (WMD = -29.35 mg/dl (-55.53, -3.18); p-value = 0.03, I2 = 0%) and type 2 diabetes with 213 participants (WMD = -21.36 mg/dl (-39.24, -3.47); p-value = 0.02, I2 = 32.1%). However, for HDL cholesterol, 70 participants with type 1 diabetes demonstrated a higher HDL compared to control group (WMD = 8.07 mg/dl (0.45, 15.70); p-value = 0.04, I2 = 86.1%) while type 2 DM with 172 participants failed to reveal significant difference in HDL between omega-3 fatty acids group and control group (WMD = 2.59 mg/dl (-1.40, 6.57); p-value = 0.20, I2 = 67.5%). Other parameters of interest are shown in Table 4.

Fig 4

Forest plots of the included studies assessing proteinuria among diabetic patients categorized by type of diabetes.

Table 4

Summary effects of subgroup analysis on the type of diabetes and follow-up period on omega-3 fatty acids on outcomes of interest among diabetic patients.

Outcomes	Mean difference	95% CI	P-value	I2
Proteinuria	SMD
Type of diabetes
Type 2	-0.29	(-0.54, -0.03)	0.03	3.9%
Type 1	0.01	(-0.36, 0.38)	0.95	0%
Follow-up period
< 24 weeks	-0.06	(-0.35, 0.23)	0.68	0%
> = 24 weeks	-0.30	(-0.58, -0.02)	0.04	6.1%
eGFR	WMD
Type of diabetes
Type 2	1.34 mL/min/1.73m2	(-4.94, 7.61)	0.68	59.2%
Type 1	1.88 mL/min/1.73m2	(-2.90, 6.67)	0.44	0%
Follow-up period
< 24 weeks	-0.70 mL/min/1.73m2	(-4.80, 3.40)	0.74	0%
> = 24 weeks	4.35 mL/min/1.73m2	(-1.39, 10.09)	0.14	40.6%
SBP	WMD
Type of diabetes
Type 2	-0.95 mmHg	(-6.69, 4.79)	0.75	37.1%
Type 1	-2.19 mmHg	(-5.21, 0.84)	0.16	0%
Follow-up period
< 24 weeks	-2.93 mmHg	(-7.15, 1.28)	0.17	0%
> = 24 weeks	-1.36 mmHg	(-6.08, 3.36)	0.57	26.7%
DBP	WMD
Type of diabetes
Type 2	0.78 mmHg	(-4.40, 5.97)	0.77	31.9%
Type 1	-0.16 mmHg	(-4.44, 4.13)	0.94	55.9%
Follow-up period
< 24 weeks	0.29 mmHg	(-3.98, 4.55)	0.90	36.6%
> = 24 weeks	1.67 mmHg	(-2.51, 5.85)	0.43	38.3%
Triglyceride	WMD
Type of diabetes
Type 2	-21.36 mg/dl	(-39.24, -3.47)	0.02	32.1%
Type 1	-29.35 mg/dl	(-55.53, -3.18)	0.03	0%
Follow-up period
< 24 weeks	-23.10 mg/dl	(-41.33, -4.87)	0.01	0%
> = 24 weeks	-24.60 mg/dl	(-43.99, -5.20)	0.01	14.7%
TC	WMD
Type of diabetes
Type 2	1.96 mg/dl	(-11.36, 15.28)	0.77	86.6%
Type 1	6.79 mg/dl	(-4.52, 18.10)	0.24	62.2%
Follow-up period
< 24 weeks	-0.91 mg/dl	(-8.47, 6.64)	0.81	0%
> = 24 weeks	5.99 mg/dl	(-5.51, 17.48)	0.31	88.2%
HDL-c	WMD
Type of diabetes
Type 2	2.59 mg/dl	(-1.40, 6.57)	0.20	67.5%
Type 1	8.07 mg/dl	(0.45, 15.70)	0.04	86.1%
Follow-up period
< 24 weeks	1.53 mg/dl	(-1.43, 4.50)	0.31	0%
> = 24 weeks	6.60 mg/dl	(1.47, 11.72)	0.01	88.0%
LDL-c	WMD
Type of diabetes
Type 2	-0.26 mg/dl	(-7.08, 6.56)	0.94	0%
Type 1	4.67 mg/dl	(-1.92, 11.25)	0.17	0%
Follow-up period
< 24 weeks	2.23 mg/dl	(-5.59, 10.04)	0.58	0%
> = 24 weeks	1.86 mg/dl	(-5.50, 9.23)	0.62	27.8%
HbA1C	WMD
Type of diabetes
Type 2	-0.14%	(-0.55, 0.26)	0.50	20.5%
Type 1	0.27%	(-0.73, 1.27)	0.60	81.9%
Follow-up period
< 24 weeks	0.33%	(-0.18, 0.83)	0.21	0%
> = 24 weeks	-0.22%	(-0.73, 0.29)	0.40	69.5%

eGFR, estimated glomerular filtration rate; TC, total cholesterol; LDL-c, low density lipoprotein cholesterol; HDL-c, high density lipoprotein cholesterol; HbA1C, hemoglobin A1C; SBP, systolic blood pressure; DBP, diastolic blood pressure; WMD, weighted mean differences; SMD, standardized mean difference

Stratified by the duration of follow-up, we used 24 weeks as a cut point since this value was a median. We found that study with follow-up time at least 24 weeks (203 participants) demonstrated a significant reduction in proteinuria comparing omega-3 fatty acids to control group (SMD = -0.30 (-0.58, -0.02); p-value = 0.04, I2 = 6.1%) while study with follow-up period less than 24 weeks (139 participants) failed to show significant difference in proteinuria (SMD = -0.06 (-0.35, 0.23); p-value = 0.68, I2 = 0%). Other parameters were shown in Table 4.

Moreover, we found that only type 2 DM patients who received omega-3 fatty acids for at least 24 weeks (165 participants) had a significant decrease in proteinuria comparing to control group (SMD = -0.38 (-0.73, -0.03); p-value = 0.04, I2 = 24.8%). While among type 1 DM patients, there was no significant difference in decreasing proteinuria even supplementing with omega-3 fatty acids for more than 24 weeks (38 participants) (SMD = 0.03 (-0.61, 0.67); p-value = 0.93, I2 = 0%). In a meta-regression analysis, the change in proteinuria was not associated with change in GFR (-0.01 (-0.09, 0.07); p-value = 0.69) and the change in proteinuria was not associated with combined dose of EPA and DHA (0.03 (-0.17, 0.24); p-value = 0.73).

Assessment of publication bias

As Egger’s test for proteinuria as our primary outcome was not significant (P > 0.05), together with a funnel plot for proteinuria of the studies included in this meta-analysis without significant asymmetry. Therefore, publication bias was less likely to occur. (Fig 5)

Fig 5

Funnel plot of standardized mean difference of proteinuria.

Discussion

Even though several meta-analyses have previously investigated the effects of omega-3 fatty acids on proteinuria, the possible benefits of omega-3 fatty acids remain unclear, especially among diabetic patients. This is the largest meta-analysis to assess the treatment effect of omega-3 fatty acids on proteinuria and other outcomes among different types of diabetic patients. Our meta-analysis demonstrated that omega-3 fatty acids could ameliorate proteinuria among type 2 DM who received this supplementation for at least 24 weeks. However, there were no significant effects on eGFR, serum LDL-cholesterol, serum HbA1C and blood pressure parameters. We included 344 patients with both type 2 DM and type 1 DM in RCTs from 1960 to April 2019. A previous meta-analysis by Miller et al.[17] in 2009 included only 222 diabetic patients, which suggested insufficient data to confirm the efficacy of omega-3 fatty acid treatments for proteinuria in diabetic patients. Moreover, we also performed subgroup analysis in terms of type of diabetes and follow-up period to gain more insight on the exploration of heterogeneity and we found a significant effect of omega-3 fatty acids on reducing proteinuria among type 2 DM and among patients with a follow-up period of at least 24 weeks.

The mechanisms through which omega-3 fatty acids diminish proteinuria are not clear. Evidence suggests that omega-3 fatty acids may act via renal hemodynamic effects[33]. However, in our meta-analysis, the observed effects of omega-3 fatty acids supplementation on proteinuria are not likely the result of blood pressure or renal perfusion effects because we did not observe any significant differences in blood pressure parameters.The effect of omega-3 fatty acids in ameliorating proteinuria may be beyond hemodynamic parameters. One of the hypotheses is that omega–3 fatty acids may reduce urine protein excretion through anti-inflammatory effects and oxidative stress. As hyperglycemia among diabetic patients induces podocyte injury as well as endothelial cell and tubulointerstitial injury through the formation of advanced glycation end‐products (AGE), activation of protein kinase C (PKC) and generation of reactive oxygen species, this process plays a pivotal role in initiation and progression of proteinuria and diabetic nephropathy[34].

Our meta-analysis demonstrated only the benefits in delaying proteinuria among type 2 DM patients. This could be explained by a small sample size of type 1 DM patients (213 vs 97). Additionally, the pathophysiology of diabetic nephropathy in type 2 DM and type 1 DM patients is somewhat different. For type 2 DM, proteinuria could be caused by various etiologies including but not limited to insulin resistance, concomitant hypertension and obesity. One of the possible explanations would be that among type 2 diabetes there are pro-inflammatory cytokines generated from abundant adipose tissue as a part of obesity in type 2 diabetes. This inflammatory response leads to proteinuria among diabetic nephropathy. Omega-3 fatty acids help reduce insulin resistance as well as pro-inflammatory responses from adipose tissue. This effect might result in lower proteinuria compared to patients with type 1 diabetes which proteinuria is mainly through polyol, hexosamine, advanced glycation end product and protein kinase C (PKC) pathways [35, 36]. Nevertheless, any meta-analyses could not derive explanations for any mechanistic pathways or derive a hypothesis. Hence, future studies designed to examine mechanisms of omega-3 fatty acids on proteinuria or kidney functions are needed as well as to assess the effect of omega-3 fatty acids on inflammatory cytokines among type 1 and type 2 diabetes.

We found that omega-3 fatty acids did not provide any effects on GFR decline. This could be explained by a low sample size as well as short period of follow-up. Furthermore, we knew that there were about one-third of proteinuric patients who did not develop end-stage renal disease (ESRD) after 20 years of follow-up and about 10% of diabetic patients without proteinuria whose kidney function kept declining and led to ESRD[37, 38]. Therefore, proteinuria and GFR decline is loosely correlated as we also found by meta-regression. However, proteinuria is still a predictor of cardiovascular and stroke events among diabetic patients. We hypothesized that omega-3 fatty acids could help diminish proteinuria and reduce cardiovascular complications and stroke incidence among type 2 DM.

In terms of effects on lowering blood pressure of omega-3 fatty acids, our findings are consistent with the previous meta-analysis of the effects of omega-3 acids on cardiometabolic biomarkers in type 2 diabetes by Lauren et al. in 2018 [39] which included 2674 patients. With respect to HbA1C, the effect of omega-3 fatty acids on HbA1C is controversial. A meta-analysis by Zhou et al.[40] found that intake of omega-3 fatty acids might be associated with increased type 2 diabetes risk. It raised the concern that omega-3 fatty acids intake might interfere with HbA1C control. However, our meta-analysis revealed no significant difference in HbA1C between treatment arms and control group which is congruent with the latest meta-analysis on the same topic for HbA1C by Chen et al [41]. Lastly, regarding the effects of omega-3 fatty acids on blood lipid level, it aligns with the previous meta-analysis [39, 42] which showed a significant reduction in serum triglyceride. However, our meta-analysis did not find a significant reduction in LDL. This might be explained by our small sample size to conclude the effect on serum lipid profile. Additionally, we found that omega-3 fatty acids significantly raised serum HDL only among type 1 diabetes. This could be explained by higher doses of omega-3 fatty acids in each trial supplemented among type 1 diabetic patients.

Our meta-analysis had several strengths that are worth mentioning. First, only RCTs were included. Hence, the bias would be smaller than observational studies due to less confounding. Second, we quantified the association between omega-3 fatty acids and amount of proteinuria and examined it within subgroups. The subgroup analyses allowed the effect of omega-3 fatty acids to be evaluated in specific type of diabetes and follow-up period. In the meanwhile, several limitations of our study should be highlighted. Although, we have the largest sample size, 344 participants were still considered as fairly small number of patients particularly when we performed subgroup analysis. We acknowledged that even after we performed random effects model in our meta-analysis as well as explored for heterogeneity, there are still possible residual confounding such as different background diets of patients or concurrent medications in each trial which were not described. Moreover, different doses and components of omega-3 fatty acids in each trial as well as different control group could lead to heterogeneity and we did not have enough data to perform a dose response meta-analysis. However, EPA and DHA had similar biological actions and properties[43, 44]. Regarding the time of follow-up, median of 24 weeks were relatively short to detect the GFR decline. Furthermore, we had insufficient data on certain clinical parameters regarding duration of diabetes, concurrent medications particularly ACEI/ARB and different methods using to measure urine protein or albumin excretion as an endpoint. Moreover, it was also difficult to conclude whether the effects on proteinuria or other outcomes were caused by EPA or DHA. Furthermore, some biomarkers such as hs-CRP that reflects inflammation were lacking.

In conclusion, the present meta-analysis of 10 RCTs encompassing 344 participants demonstrated that omega-3 fatty acids could ameliorate proteinuria among type 2 DM patients who received omega-3 supplementation for at least 24 weeks without adverse effects on HbA1C, total serum cholesterol and LDL-cholesterol. However, there were no significant difference in change in eGFR between omega-3 fatty acids and placebo group. Clinical trials with more participants and longer time of follow-up should be conducted to better understanding the effects of omega-3 fatty acids on kidney outcomes as well as cardiovascular complications and incidence of stroke among diabetic patients. Markers of oxidative stress, inflammation and urine protein fingerprinting which could reflect severity of glomerular or tubulointerstitial injury should be extensively studied in order to address the potential mechanism of omega-3 fatty acids on delaying proteinuria.

PRISMA 2009 checklist.

(DOC)

Click here for additional data file.

PubMed search strategy.

(DOCX)

Click here for additional data file.

Forrest plots of the included studies assessing HbA1C among diabetic patients.

(TIF)

Click here for additional data file.

Forrest plots of the included studies assessing total cholesterol among diabetic patients.

(TIF)

Click here for additional data file.

Forrest plots of the included studies assessing HDL cholesterol among diabetic patients.

(TIF)

Click here for additional data file.

Forrest plots of the included studies assessing LDL cholesterol among diabetic patients.

(TIF)

Click here for additional data file.

Forrest plots of the included studies assessing serum triglyceride among diabetic patients.

(TIF)

Click here for additional data file.

Forrest plots of the included studies assessing eGFR among diabetic patients.

(TIF)

Click here for additional data file.

Summary effects of omega-3 fatty acids on outcomes of interest among diabetic patients (Fixed effects model).

(DOCX)

Click here for additional data file.

10 Oct 2019

PONE-D-19-24289

The Effects of Omega-3 Fatty Acids on Proteinuria among Patients with Diabetes: A Meta-analysis of Randomized Controlled Trials

PLOS ONE

Dear Dr. Chewcharat,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

We would appreciate receiving your revised manuscript by 9 December 2019. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Tomislav Bulum, MD, PhD

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We noticed you have some minor occurrence of overlapping text with the following previous publication(s), which needs to be addressed:

- Hu, Jing, Zuoliang Liu, and Hao Zhang. "Omega-3 fatty acid supplementation as an adjunctive therapy in the treatment of chronic kidney disease: a meta-analysis." Clinics 72.1 (2017): 58-64.

- https://www.ahajournals.org/doi/pdf/10.1161/jaha.117.006020?download=true

The text that needs to be addressed involves some sentences  of the Introduction.

In your revision ensure you cite all your sources (including your own works), and quote or rephrase any duplicated text outside the methods section. Further consideration is dependent on these concerns being addressed.

3. Please include your tables as part of your main manuscript and remove the individual files. Please note that supplementary tables (should remain/ be uploaded) as separate "supporting information" files

4. Please upload a copy of Figure 5, to which you refer in your text. If the figure is no longer to be included as part of the submission please remove all reference to it within the text.

5. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

Reviewers' comments:

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Well done. The author had performed a good meta analysis to answer an important clinical question. Being a meta-analysis , I understand that some of the data eg the use of ACE-I/ARB were not available

Reviewer #2: Title: usually term ‘proteinuria among patients with diabetes’ denotes diabetic nephropathy.

Abstract: No issue except for conclusion. Please refer.

Introduction: The introduction states the gaps in knowledge which justifies a new metaAnalyses .

However, in the 2nd paragraph in reference to the line ‘ However, the effects of omega-3 fatty acids on kidney function still remain controversial particularly in diabetic kidney disease.’ I do not think word 'controversial' is right here. n-3 FAs are perceived to have a role in mediating inflammation, dyslipidemia etc. A more appropriate term would be 'lack of consensus’.

1st Line/Last paragraph/ 1st pp of introduction. The aim of this meta-analysis was to evaluate the benefits of omega-3 fatty acid supplementation in reducing proteinuria in diabetic patients by using all available evidence from the published literature. Isn’t ‘benefits’ pre-conclusive???

Methods: no issue

Results:

2nd line/ 1st paragraphs.. A total of 1277 potentially relevant citations were identified and screened. Seventy citations were evaluated in details, of which 10 trials (refs) with 344 participants…..

Issue- [ref] not stated.

Risk of bias section.. my comment—are the same studies ticking for high risk of bias as per overall risk? Randomization process? Deviation from intended intervention? Missing data?

Subgroup analyses..please quote patient numbers for each group. Also 2nd paragraph/ last line ..should be ‘are’ instead of ‘were’.

Discussion:

There is considerable room for improvement. I would like to detail some points which should be considered:

The major limitation of meta-analyses should be acknowledged. This is it cannot be used to derive explanations for mechanistic pathways or even derive a hypothesis. Further, the construct of this meta-Analyses depends on studies that are poor in design, small in patient numbers and lacking data on more robust biomarkers relating to inflammation [CRP], microalbuminemia status and blood urea levels. There is no point referencing animal studies. There are many studies with CKD population to provide enough inputs as to limitations in these n-3 fatty acid feeding trials.

In addition in reference to Table 1 which describes the selected studies:

There is heterogeneity in treatment based on dose and components of n-3 PUFAs [EPA, DHA or EPA+DHA..an either-and situation]. Further there is also heterogeneity of control treatment. Placebo [not stated] or oleic acid or linoleic acid. The control and treatments DO NOT match. This should be discussed.

Secondly, in these trials the background diets of the patients are not described. As we all know, once a patient is counseled a protein diet [1st line of management to treat proteinuria], then this affects urine protein status. The 2nd aspect is tighter blood glucose control which as its effect on proteinuria, eGFR and HbA1C.

The discussion should raise all the above points.

Conclusion: There is an issue with this statement- ‘omega-3 fatty acids could ameliorate proteinuria among type 2 DM patients who received omega-3 supplementation for at least 24 weeks without adverse effects on HbA1C and serum LDL-cholesterol.’

This conclusion cannot be supported from the evidence reported:

• Overall diabetic patients [NIDDM+IDDM] proteinuria- not significant; eGFR- not significant

• NIDDM- proteinuria- yes ; eGFR- not significant

• IDDM- proteinuria- not significant; eGFR- not significant

• In a meta-regression analysis, the change in proteinuria was not associated with change in GFR (- 0.01 (-0.09, 0.07); p-value = 0.69) and the change in proteinuria was not associated with combined dose of EPA and DHA (0.03 (-0.17, 0.24); p-value = 0.73).

Reviewer #3: Comments:

This manuscript aims to investigate the effects of omega-3 long- chain polyunsaturated fatty acids on proteinuria, eGFR and metabolic biomarkers in diabetic patients. This is a meta-analysis including 10 RCTs with 344 participants, and the authors report that Omega-3 supplementation for 24 weeks or longer could help alleviated proteinuria in patients with type 2 diabetes.

There are some questions should be addressed:

1. Introduction:

(1) Please provide the related references in the paragraph 2.

2. Methods:

(1) Data extraction and quality assessment: how about other serum lipids and glucose control biomarkers, such as the HDL, total cholesterol or fasting glucose?

(2) The results form fixed-effects models should also be presented.

3. Results:

(1) In the flow chart (Figure1), a total of 1277 articles are screened for retrieval, and 179 excluded. However,1089 included in the next stage (missing 9 articles), please check the number carefully.

(2) In table 1, mean age of patients ranged from 33 to 67.4 years old. The duration of follow up spanned from 6 weeks to 52 weeks. It is inconsistent with that in the results section, please check.

(3) Please provide the related tables and figures about the effect of omega-3 fatty acids on eGFR, serum lipids and glucose control.

(4) Why choose 24 weeks as a cut of duration of intervention in subgroup? Please explain. If possible, please provide the results using meta-regression analysis.

(5) Page 13, Paragraph 4: the results were not found in the table 4. Please provide.

(6) Please provide Figure 5.

(7) Each table or figure should be cited in the manuscript. Please check.

(8) Please improve the resolution and clarity of figures.

(9) The authors should provide the mean (SD) of the study outcomes for each treatment group in the figures or tables. It is inappropriate to present MD only.

4. Discussion:

(1) The first paragraph: It is inappropriate to present the result with “this is the first meta-analysis to…among diabetic patients in all aspects.”, because in 2009, Miller et al. conducted a similar meta-analysis.

(2) Please discuss the result that omega-3 fatty acids could help ameliorate serum triglyceride among type 1 DM who received omega-3 supplementation less than 24 weeks.

(3) If possible, to evaluate the optimal dosage of Omega-3 fatty acids for prevention of the study outcomes.

(4) Please further discuss the possible mechanism for effect of omega-3 supplementation on the different diabetes types.

(5) The authors first stated: “the observed effects of omega-3 fatty acids supplementation on proteinuria are not likely the result of blood pressure or renal perfusion effects only because we did not observe simultaneous changes in GFR. Hence, the effect of omega-3 fatty acids in ameliorating proteinuria may be beyond hemodynamic parameters”, while in the followed text, stated: “We found that omega-3 fatty acids did not provide any effects on GFR decline. This could be explained by low sample size as well as short period of follow-up.” Is it reasonable?

5. Please indicate the full names the first time you use the abbreviations in the text.

6. There are some spelling and grammatical errors that should be checked carefully and corrected throughout the manuscript.

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

5 Dec 2019

Responses to Journal:

Thank you for reviewing our manuscript . We edited the whole manuscript according to your format.

Responses to reviewer#1

Reviewer #1: Well done. The author had performed a good meta analysis to answer an important clinical question. Being a meta-analysis, I understand that some of the data e.g. the use of ACE-I/ARB were not available

Response: Thank you for reviewing our manuscript and for highlighting this important point. As we mentioned in the discussion, there was no information about ACE-I/ARB in the primary studies, therefore we were not able to evaluate the effect of this parameter by performing subgroup analysis.

Responses to reviewer#2

Comment#1 Title: usually term ‘proteinuria among patients with diabetes’ denotes diabetic nephropathy.

Response: Thank you for your suggestion. The manuscript’s title has now been changed to “The Effects of Omega-3 Fatty Acids on Diabetic Nephropathy: A Meta-analysis of Randomized Controlled Trials” as suggested

Comment#2 Abstract: No issue except for conclusion. Please refer.

Comment#3 Introduction: The introduction states the gaps in knowledge which justifies a new meta-analyses. However, in the 2nd paragraph in reference to the line ‘ However, the effects of omega-3 fatty acids on kidney function still remain controversial particularly in diabetic kidney disease.’ I do not think word 'controversial' is right here. n-3 FAs are perceived to have a role in mediating inflammation, dyslipidemia etc. A more appropriate term would be 'lack of consensus’.

1st Line/Last paragraph/ 1st pp of introduction. The aim of this meta-analysis was to evaluate the benefits of omega-3 fatty acid supplementation in reducing proteinuria in diabetic patients by using all available evidence from the published literature. Isn’t ‘benefits’ pre-conclusive???

Response: Thank you for your suggestions. We edited these points in the manuscript as you suggested.

Comment#4 Methods: no issue

Comment#5

Results: 2nd line/ 1st paragraphs. A total of 1277 potentially relevant citations were identified and screened. Seventy citations were evaluated in details, of which 10 trials (refs) with 344 participants…..

Issue- [ref] not stated.

Risk of bias section.. my comment—are the same studies ticking for high risk of bias as per overall risk? Randomization process? Deviation from intended intervention? Missing data?

Subgroup analyses..please quote patient numbers for each group. Also 2nd paragraph/ last line ..should be ‘are’ instead of ‘were’.

Response:

- Thank you for your suggestion. We edited the references as suggested.

- Regarding risk of bias, different studies suffered in various domains and Table 2 provides all the necessary information for the specific domains described in the text for each study separately. For overall risk, we combined all of the five elements to summarize the overall risk as recomended from the Revised Cochrane risk-of-bias tool for randomized trials (RoB 2) as referenced. We have now added a more detailed description of the results of the quality assessment in the text as well.

Comment#6

Discussion:

There is considerable room for improvement. I would like to detail some points which should be considered:

The major limitation of meta-analyses should be acknowledged. This is it cannot be used to derive explanations for mechanistic pathways or even derive a hypothesis. Further, the construct of this meta-Analyses depends on studies that are poor in design, small in patient numbers and lacking data on more robust biomarkers relating to inflammation [CRP], microalbuminemia status and blood urea levels. There is no point referencing animal studies. There are many studies with CKD population to provide enough inputs as to limitations in these n-3 fatty acid feeding trials.

In addition in reference to Table 1 which describes the selected studies:

There is heterogeneity in treatment based on dose and components of n-3 PUFAs [EPA, DHA or EPA+DHA..an either-and situation]. Further there is also heterogeneity of control treatment. Placebo [not stated] or oleic acid or linoleic acid. The control and treatments DO NOT match. This should be discussed.

Secondly, in these trials the background diets of the patients are not described. As we all know, once a patient is counseled a protein diet [1st line of management to treat proteinuria], then this affects urine protein status. The 2nd aspect is tighter blood glucose control which as its effect on proteinuria, eGFR and HbA1C.

The discussion should raise all the above points.

Conclusion: There is an issue with this statement- ‘omega-3 fatty acids could ameliorate proteinuria among type 2 DM patients who received omega-3 supplementation for at least 24 weeks without adverse effects on HbA1C and serum LDL-cholesterol.’

This conclusion cannot be supported from the evidence reported:

• Overall diabetic patients [NIDDM+IDDM] proteinuria- not significant; eGFR- not significant

• NIDDM- proteinuria- yes ; eGFR- not significant

• IDDM- proteinuria- not significant; eGFR- not significant

• In a meta-regression analysis, the change in proteinuria was not associated with change in GFR (- 0.01 (-0.09, 0.07); p-value = 0.69) and the change in proteinuria was not associated with combined dose of EPA and DHA (0.03 (-0.17, 0.24); p-value = 0.73).

Response: Thank you for your suggestions. We have reformed the Discussion section accordingly.

We acknowledge the fact that the patient populations in the synthesized studies are different and this heterogeneity led us to apply a random-effects model to analyze the data, incorporating the additional sources of variability between the studies. Regarding background diets, we understand that this is an important confounder in our exposure and outcome association. Unfortunately, there was no available information for this characteristic. However, the randomized nature of the data we are synthesizing provides a reasonable amount of confidence that the comparisons performed are not heavily affected by these factors. We included this lack of information as a limitation in our Discussion section. We are also aware that some RCTs were not optimal regarding study design. We have highlighted this in the Discussion section and also include that as a limitation of our synthesis.

Regarding the heterogeneity of the treatments and control groups, previous meta-analyses such as Miller et al and Jing et al. performed analyses similar to the current one with the same comparison groups. EPA and DHA both belong to the omega-3 polyunsaturated fatty acids (PUFAs) family and they share common biological properties not only regarding their metabolic and cardiovascular effects but even for anti-cancer effects. They have been evaluated together in the past for various conditions, therefore we considered that it would be reasonable to examine them as equivalent in this meta-analysis. It would be useful to study them separately if we had available evidence, but such data are not available.

Our conclusion: “omega-3 fatty acids could ameliorate proteinuria among type 2 DM patients who received omega-3 supplementation for at least 24 weeks without adverse effects on HbA1C and serum LDL-cholesterol” takes into account all the points that were raised from the reviewer:

• Overall diabetic patients [NIDDM+IDDM] proteinuria- not significant; eGFR- not significant

• We only refer to type 2 DM, not all diabetics. We refer only to proteinuria, not eGFR or other markers of kidney function.

• NIDDM- proteinuria- yes ; eGFR- not significant

• We do not refer to eGFR in our conclusion

• IDDM- proteinuria- not significant; eGFR- not significant

• We do not refer to IDDM patient in our conclusion

• In a meta-regression analysis, the change in proteinuria was not associated with change in GFR (- 0.01 (-0.09, 0.07); p-value = 0.69) and the change in proteinuria was not associated with combined dose of EPA and DHA (0.03 (-0.17, 0.24); p-value = 0.73).

• Our conclusion does not refer to the effect of the variables we considered for the exploration of heterogeneity. It only refers to our main analysis.

Responses to reviewer#3

Reviewer #3: Comments:

This manuscript aims to investigate the effects of omega-3 long-chain polyunsaturated fatty acids on proteinuria, eGFR and metabolic biomarkers in diabetic patients. This is a meta-analysis including 10 RCTs with 344 participants, and the authors report that Omega-3 supplementation for 24 weeks or longer could help alleviated proteinuria in patients with type 2 diabetes.

There are some questions should be addressed:

1. Introduction:

(1) Please provide the related references in the paragraph 2.

Response: Thank you for your suggestion. We added the references.

2. Methods:

(1) Data extraction and quality assessment: how about other serum lipids and glucose control biomarkers, such as the HDL, total cholesterol or fasting glucose?

(2) The results form fixed-effects models should also be presented.

Response: Thank you for your suggestion.

(1) We added as suggested.

(2) The use of random effects model to synthesize our data was based on an a-priori expectation of substantial between study heterogeneity, mainly due to clinical and methodological reasons. It is expected that the random effects model will yield more conservative results than the fixed effect when tau2 is not equal to zero. Below we demonstrated the results of the fixed-effect model. We added these results of this analysis in the supplementary material. As expected, some outcomes that were not significant in the random effects analysis became significant in the fixed effects model but this is only due to their mathematical properties. We present this table in the supplementary material.

Outcomes No of study arms No of patients Weighted mean difference/Standardized mean difference* Confidence interval I2 P-value

Proteinuria 11 342 -0.19* (-0.38, 0.01) 0% 0.06

eGFR 6 208 1.54 mL/min/1.73 m2 (-1.40, 4.48) 5.6% 0.31

SBP 10 318 -2.10 mmHg (-4.48, 0.28) 0% 0.08

DBP 10 318 2.10 mmHg (0.57, 3.63) 39.8% 0.007

Triglyceride 10 313 -24.24 mg/dL (-36.40, -12.10) 0% <0.001

TC 6 168 9.07 mg/dl (6.44, 11.71) 80.2% <0.001

HDL-c 6 242 5.97 mg/dl (4.73, 7.22) 82.5% <0.001

LDL-c 6 215 2.29 mg/dL (-2.45, 7.03) 0% 0.34

HbA1C 10 313 -0.42% (-0.60, -0.24) 66.2% <0.001

3. Results:

(1) In the flow chart (Figure1), a total of 1277 articles are screened for retrieval, and 179 excluded. However,1089 included in the next stage (missing 9 articles), please check the number carefully.

Response: Thank you for your suggestion. It was a typo.

(2) In table 1, mean age of patients ranged from 33 to 67.4 years old. The duration of follow up spanned from 6 weeks to 52 weeks. It is inconsistent with that in the results section, please check.

Response: Thank you for pointing this out. The data in Table 1 are correct, and we have corrected the numbers in the main text

(3) Please provide the related tables and figures about the effect of omega-3 fatty acids on eGFR, serum lipids and glucose control.

Response: Thank you for your suggestion. The Tables and Figures for eGFR, serum lipids, and glucose control can be found as part of the online Supplement .

(4) Why choose 24 weeks as a cut of duration of intervention in subgroup? Please explain. If possible, please provide the results using meta-regression analysis.

Response: Thank you for your suggestion. Since there is no formal consensus on the minimum duration of RCTs for this topic and there is large variability in the duration of the studies, we used the median (24 weeks) as a measure of the central tendency of the duration of the studies. This duration (24 weeks) is clinically meaningful as well because the results of supplementation are not expected to be seen in a concise period of time. Therefore, we wanted to differentiate between a true null effect and a null effect because of an inadequate duration of the treatment. We also performed a meta-regression analysis on the duration of follow up as a continuous variable in weeks. The change in proteinuria was not associated with duration of follow-up (-0.01 (-0.02, 0.005); p-value = 0.23). The interpretation of the meta-regression coefficient is the change in proteinuria per 1-week increase in the duration of the treatment which may not be directly interpreted in clinical practice. These results are shown in Table 4.

(5) Page 13, Paragraph 4: the results were not found in the table 4. Please provide.

(6) Please provide Figure 5.

(7) Each table or figure should be cited in the manuscript. Please check.

(8) Please improve the resolution and clarity of figures.

Response: Thank you for your suggestion. We resolved these issues.

(9) The authors should provide the mean (SD) of the study outcomes for each treatment group in the figures or tables. It is inappropriate to present MD only.

Response: Thank you for your suggestion. In the Forest plots of our analysis, we do report the WMD or SMD of each study along with the 95% CI, which are calculated directly from the SD. In the Tables of the manuscript, because of space limitations, we could not include the individual estimates for all the outcomes. Therefore this information is easily derived from the forest plots.

4. Discussion:

(1) The first paragraph: It is inappropriate to present the result with “this is the first meta-analysis to…among diabetic patients in all aspects.”, because in 2009, Miller et al. conducted a similar meta-analysis.

Response: Thank you for your suggestion. We downtoned the statement to being the largest meta-analysis among diabetic patients as per your suggestion.

(2) Please discuss the result that omega-3 fatty acids could help ameliorate serum triglyceride among type 1 DM who received omega-3 supplementation less than 24 weeks.

Response: Thank you for pointing this out. After careful review of the numbers, we realized that this was en error at the data entry where the numbers for the control group were placed at the intervention group and vice-versa. With this opportunity, we went back and checked all the data thoroughly, reassuring that there is no other error.

(3) If possible, to evaluate the optimal dosage of Omega-3 fatty acids for the prevention of the study outcomes.

Response: Thank you for your suggestion. Due to lack of available data, we could not perform dose-response analysis to conclude the optimal dose with the least side effects. We added that as a potential limitation to our analysis.

(4) Please further discuss the possible mechanism for effect of omega-3 supplementation on the different diabetes types.

Response: Thank you for your suggestion. We added this part as suggested.

(5) The authors first stated: “the observed effects of omega-3 fatty acids supplementation on proteinuria are not likely the result of blood pressure or renal perfusion effects only because we did not observe simultaneous changes in GFR. Hence, the effect of omega-3 fatty acids in ameliorating proteinuria may be beyond hemodynamic parameters”, while in the followed text, stated: “We found that omega-3 fatty acids did not provide any effects on GFR decline. This could be explained by low sample size as well as short period of follow-up.” Is it reasonable?

Response: Thank you for pointing this out. In this sentence, we tried to provide alternative explanations for the null association between omega-3 fatty acids and GFR, besides the association being indeed null (lack of power or not adequate follow-up). We re-phrased this part of the manuscript to be clearer.

5. Please indicate the full names the first time you use the abbreviations in the text.

Response: Thank you for your suggestion. We edited as suggested

6. There are some spelling and grammatical errors that should be checked carefully and corrected throughout the manuscript.

Response: Thank you for your suggestion. We went through the manuscript carefully and corrected all the errors.

Submitted filename: Response_AC&SP_final.docx

Click here for additional data file.

20 Dec 2019

PONE-D-19-24289R1

The Effects of Omega-3 Fatty Acids on Diabetic Nephropathy: A Meta-analysis of Randomized Controlled Trials

PLOS ONE

Dear Dr. Chewcharat,

Thank you for submitting your revised manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

We would appreciate receiving your revised manuscript by 20 January 2020. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Tomislav Bulum

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #2: All comments have been addressed

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #2: Partly

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #2: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Reviewer #2: Dear Authors,

These details are required:

[1] Table 1 should provide eGFR data

[2] Check reference citation...eg. No.4 is it WE Mitch?? should be.

Reviewer #3: The authors have answered all the questions. However, they still should make a further discussion for the following question in the revised manuscript.

1. The reference 37 is not a meta-analysis, and did not support the related discussion. Please check the order of the references.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #2: Yes: Tilakavati Karupaiah

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

7 Jan 2020

January 6, 2020

Re: Manuscript entitled “The Effects of Omega-3 Fatty Acids on Diabetic Nephropathy: A Meta-analysis of Randomized Controlled Trials ”

Submission ID: PONE-D-19-24289

Dear Editor,

Thank you for the thoughtful input and review of our manuscript. We believe as a result of this review, our study will have more value for your readers. We revised the manuscript based on the reviewers’ suggestions. We have attached our point by point response.

Thank you for your time and consideration. If you have any additional questions or comments, please let us know.

Sincerely,

Api Chewcharat, MD, MPH

Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, MA 02115, USA

Email: Api.che@hotmail.com

Responses to reviewer#2

Reviewer #2: Dear Authors,

These details are required:

[1] Table 1 should provide eGFR data

Response: Thank you for your suggestion. We added eGFR data in table 1.

[2] Check reference citation...eg. No.4 is it WE Mitch?? should be.

Response: We appreciated the reviewer’s valuable input. We found the suggested reference very helpful to support our introduction. Hence, we added the reference as suggested.

Responses to reviewer#3

Reviewer #3: The authors have answered all the questions. However, they still should make a further discussion for the following question in the revised manuscript.

1. The reference 37 is not a meta-analysis, and did not support the related discussion. Please check the order of the references.

Response: Thank you for your comment. To the best of our knowledge, there were no other meta-analyses to support our findings as to why there was only significant reduction in proteinuria only among type 2 diabetes. We postulated that because in type 2 DM, there are higher inflammatory cytokines generated from adipose tissue compared to type 1 DM. Omega-3 fatty acids might help diminish these inflammatory cytokines that eventually leads to lower proteinuria in only type 2 DM. However, we did not have enough information to confirm this hypothesis since the inflammatory markers were not available in many trials that we included. Moreover, it might be explained by the inadequate power to detect the significant difference of proteinuria among type 1 DM because there were only 97 participants in type 1 DM compared to 213 participants for type 2 DM. Future studies are needed to assess whether omega-3 fatty acids could reduce the inflammatory cytokines differently between type 1 and type 2 DM. The following text has been added to the discussion, as suggested. We also removed the reference 37 in the previous manuscript version.

“Our meta-analysis demonstrated only the benefits in delaying proteinuria among type 2 DM patients. This could be explained by a small sample size of type 1 DM patients (213 vs 97). Additionally, the pathophysiology of diabetic nephropathy in type 2 DM and type 1 DM patients is somewhat different. For type 2 DM, proteinuria could be caused by various etiologies including but not limited to insulin resistance, concomitant hypertension and obesity. One of the possible explanations would be that among type 2 diabetes there are pro-inflammatory cytokines generated from abundant adipose tissue as a part of obesity in type 2 diabetes. This inflammatory response leads to proteinuria among diabetic nephropathy. Omega-3 fatty acids help reduce insulin resistance as well as pro-inflammatory responses from adipose tissue. This effect might result in lower proteinuria compared to patients with type 1 diabetes which proteinuria is mainly through polyol, hexosamine, advanced glycation end product and protein kinase C (PKC) pathways (reference 35, 36). Nevertheless, any meta-analyses could not derive explanations for any mechanistic pathways or derive a hypothesis. Hence, future studies designed to examine mechanisms of omega-3 fatty acids on proteinuria or kidney functions are needed as well as to assess the effect of omega-3 fatty acids on inflammatory cytokines among type 1 and type 2 diabetes.”

For the sentence after that, we would like to raise the point that any meta-analysis could not derive or proof any mechanistic pathway.

We greatly appreciated the editors’ and reviewers’ time and comments to improve our manuscript.

Submitted filename: R2_Omega-3.docx

Click here for additional data file.

14 Jan 2020

The Effects of Omega-3 Fatty Acids on Diabetic Nephropathy: A Meta-analysis of Randomized Controlled Trials

PONE-D-19-24289R2

Dear Dr. Chewcharat,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Tomislav Bulum

Academic Editor

PLOS ONE

23 Jan 2020

PONE-D-19-24289R2

The Effects of Omega-3 Fatty Acids on Diabetic Nephropathy: A Meta-analysis of Randomized Controlled Trials

Dear Dr. Chewcharat:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Tomislav Bulum

Academic Editor

PLOS ONE