Introduction
Methods
Literature search
Literature selection
Data extraction
Assessment of methodological quality and outcome-specific certainty of evidence
Definition of the outcomes
Rating the overall certainty of the evidence
Overall certainty of evidence | Underlying criteria | Definition/explanation |
---|---|---|
Convincing | At least one SR with or without MA of prospective studies available | There is high level of confidence that the true effect lies close to that of the estimate(s) of the effect |
If more than one SR with or without MA are available: all overall results must be consistent1 | ||
In case of a positive or negative association, biological plausibility is given | ||
All included SRs with or without MA must reach at least a “moderate” outcome-specific certainty of evidence2; in addition, all included SRs must reach at least a methodological quality3 of “moderate” | ||
Probable | At least one SR with or without MA of prospective studies available | There is moderate confidence in the effect estimate(s): the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different |
If more than one SR with or without MA are available, the majority of overall results must be consistent1 | ||
In case of a positive or negative association, biological plausibility is given | ||
The majority4 of included SRs with or without MA must have reached at least a “moderate” certainty of evidence2; in addition, all included SRs must reach at least a methodological quality3 of “moderate” | ||
Possible | At least one SR with or without MA of prospective studies available | Confidence in the effect estimate(s) is limited: the true effect may be substantially different from the estimate of the effect |
If more than one SR with or without MA are available, the majority of overall results must be consistent1 | ||
In case of a positive or negative association, biological plausibility is given | ||
The majority4 of included SRs with or without MA must reach at least a “low” certainty of evidence2; in addition, the majority4 of all included SRs must reach at least a methodological quality3 of “moderate” | ||
Insufficient | No SR is available | There is very little confidence in the effect estimate (s): the true effect is likely to be substantially different from the estimate of effect |
OR | ||
The majority4 of included SRs with or without MA reach a “very low” certainty of evidence2; in addition, the majority of all included SRs reach a methodological quality3 of “low” |
Results
Characteristics of the included systematic reviews
Author, year | Study type, study period | Study population | Exposition | Protein intake | Outcome | Effect estimates | Heterogeneity estimators | NutriGrade rating | AMSTAR 2 rating |
---|---|---|---|---|---|---|---|---|---|
Asoudeh 2022 [22] | SR with MA of cohort studies Published until July 2021 Study duration: 6–26 yrs | Both sexes General population Aged ≥ 20 yrs | NP | Risk of kidney stones | Pooled RR (95% CI), random/fixed effect model | Moderate | |||
2 cohort studies | n = 154,221 n = 2,982 cases | Total protein | Fixed effect model: 1.04 (0.92, 1.18) Dose-response analysis: 0.99 (0.98, 1.01) per 10g/d increase Pnonlinearity = 0.11 | I2 = 68% P = 0.08 I2 = 44.7% P = 0.18 | Low: 4.0 | ||||
6 cohort studies | n = 466,109 n = 8,494 cases | Animal protein | Random effect model: 1.00 (0.89–1.14) | I2 = 62% P = 0.02 | Low: 4.9 | ||||
5 cohort studies | n = 387,816 n = 6,542 cases | Dose–response analysis: 0.99 (0.97, 1.02) per 10 g/d increase Pnonlinearity = 0.26 | I2 = 72% P < 0.01 | ||||||
2 cohort studies | n = 320,896 n = 8,971 cases | Nondairy animal protein | Fixed effect model: 1.11 (1.03, 1.20) Dose–response analysis: 1.01 (1.00, 1.02) per 10 g/d increase Pnonlinearity = 0.38 | I2 = 0% P = 0.47 I2 = 42% P = 0.16 | Low: 5.0 | ||||
2 cohort studies | n = 320,896 n = 8,971 cases | Dairy protein | Fixed effect model: 0.91 (0.84, 0.99) Dose-response analysis: 0.96 (0.93, 0.99) per 10g/d increase Pnonlinearity = 0.86 | I2 = 0% P = 0.55 I2 = 0% P = 0.57 | Moderate. 6.0 | ||||
Hengeveld 2022 [23] | SR without MA of RCTs Published before 04/2020 Study duration: 12 wks to 18 mos | Both sexes Relatively healthy and/or people with (moderate) physical function limitations, overweight, obesity and/or (pre-)frailty Aged ≥ 65 yrs | High vs low animal protein intake (2 RCTs with concomitant exercise in control and intervention group) | 1.06–1.4 g/kg BW/d vs. 0.81-1.05 g/kg BW/d | NA | Moderate | |||
4 RCTs | n = 347–361 | Serum creatinine | 3 out of 4 RCTs reported no significant effect of increased protein | Low: 4.5 | |||||
5 RCTs | n = 622 | eGFR | None of the RCTs reported a significant effect of increased protein | Moderate: 5.5 | |||||
1 RCT | n = 111 | Albumin/creatinine ratio | The RCT reported no significant effect of increased protein | Low: 4.5 | |||||
Kelly 2021 [21] | SR with MA of cohort studies Published until June 2019 Follow-up: 5.5–21 yrs | Both sexes General population, partly people with T2D, hypertension, dyslipidaemia and/or CVD Aged 30–67 yrs | Higher vs. lower protein intake | NP | OR (95% CI), random effects model | High | |||
3 cohort studies | n = 19,835 | Incident CKD | 1.08 (0.91, 1.28) P = 0.36 | I2 = 47% Chi2 = 3.75 tau2 = 0.01 df = 2 P = 0.15 | Very low: 3.0 | ||||
5 cohort studies | n = 18,507 | GFR decline | 1.07 (0.96, 1.19) P = 0.25 | I2 = 42% Chi2 = 12.16 tau2 = 0.01 df = 7 P = 0.10 | Very low: 3.5 | ||||
Lin 2020 [20] | SR with MA of 4 cohort studies Published until May 2019 follow-up: 20–26 yrs | Both sexes Aged 25–79 yrs n = 271,969 | High vs. low animal protein intake | NP | Nephrolithiasis (incident stones) | RR (95% CI), random effects model 1.1 (1.02–1.19) P = NP | No heterogeneity | Low: 4.5 | High |
Devries 2018 [16] | SR with MA of RCTs Published from 1975 until 2016 Intervention duration: 4 d–104 wks | Both sexes Healthy and/or people with obesity and/or hypertension Aged ≥ 18 yrs | High vs. normal/low protein intake | 1.2–3.3 g/kg BW/d vs. 0.3–2.6 g/kg BW/d 20–40 En% vs. 12–24 En% 123–150 g/d vs. 46–75 g/d | SMD (95% CI), random effects model | Moderate | |||
28 RCTs | n = 1409 | GFR | 0.19 (0.07, 0.31) P = 0.002 Dose–response analysis (linear) protein intake (g/kg BW/d) r = 0.332, P = 0.03 | I2 = 0% Chi2 = 15.77 tau2 = 0.00 df = 29 P = 0.98 | Moderate: 7.05 | ||||
14 RCTs | n = 1307 | ΔGFR | 0.11 ( – 0.05, 0.27) P = 0.16 Dose–response analysis (linear) protein intake (g/kg BW/d) r = 0.184, P = 0.33 | I2 = 44% Chi2 = 30.26 tau2 = 0.05 df = 17 P = 0.02 | Moderate: 6.35 | ||||
Van Elswyk 2018 [15] | SR without MA of RCTs and cohort studies Published before 08/2017 Intervention duration: 4 d–8 wks Follow-up: 15–21 yrs | Both sexes Healthy and/or people with metabolic risk factors Aged ≥ 18 yrs | Protein intake consistent with the US RDA (≥ 0.8 g/kg BW/d or 10–15 En%) or higher protein intake (20–35 En% or 10% higher than comparison intake) | 1.8–2.5 g/kg BW/d vs. 0.7–1.5 g/kg BW/d | NA | Low | |||
1 cohort study | n = 3798 | Total protein, animal protein, plant protein | eGFR | “Intake of total, plant, and animal protein are not associated with changes in eGFR over time.” | Low: 4.0 | ||||
13 RCTs | n = 235 | Total protein, animal protein, plant protein | GFR | 8 of 11 RCTs reported Significantly higher GFR in response to increased protein intake | Very low: 2.0 | ||||
4 RCTs | n = 60 | Total protein, animal protein, plant protein | Uric acid | 3 of 4 RCTs reported elevated uric acid in response to increased protein intake | Very low: 3.0 | ||||
Schwingshackl 2014 [18] | SR with MA of RCTs Published until 02/2014 Intervention duration: 1 wk–24 mos | Both sexes Generally healthy and/or people with overweight, obesity and/or T2D - Mean age: 22.3–67 yrs | High vs. normal/low protein diet (≥ 5% difference in total energy intake) | 12.5–40 En% vs. 5.4–24 En% or 1.0–2.4 g/kg BW/d vs. 0.5–1.2 g/kg BW/d | Pooled WMD (95% CI), random effects model | Moderate | |||
21 RCTs | n = 1599 | GFR | 7.18 ml/min/1.73 m2 (4.45, 9.91) P = < 0.001 | I2 = 52% tau2 = 16.53 Chi2 = 47.76 df = 23 P = 0.002 | Moderate: 6.6 | ||||
22 RCTs | n = 1764 | Serum creatinine | – 1.42 µmol/l ( – 3.50, 0.65) P = 0.18 | I2 = 57% tau2 = 10.13 Chi2 = 51.42 df = 22 P = 0.0004 | Low: 5.6 | ||||
13 RCTs | n = 910 | Serum urea | 1.75 mmol/l (1.13, 2.37) P = < 0.00001 | I2 = 88% tau2 = 1.13 Chi2 = 112.57 df = 13 P < 0.00001 | Moderate: 6.6 | ||||
8 RCTs | n = 295 | Serum uric acid | 0.18 µmol/l ( – 0.08, 0.44) P = 0.17 | I2 = 3% tau2 = 0.00 Chi2 = 7.25 df = 7 P = 0.40 | Low: 5.4 | ||||
7 RCTs | n = 210 | Urinary pH | – 0.39 ( – 0.82, 0.03) P = 0.07 | I2 = 95% tau2 = 0.33 Chi2 = 148.51 df = 7 P < 0.00001 | Low: 5.3 | ||||
11 RCTs | n = 783 | Urinary albumin/protein | 0.50 mg/24 h ( – 2.83, 3.82) P = 0.77 | I2 = 63% tau2 = 13.51 Chi2 = 32.58 df = 12 P = 0.001 | Low: 5.8 | ||||
10 RCTs | n = 708 | Urinary calcium excretion | 25.43 mg/24 h (13.62, 37.24) P < 0.001 | I2 = 90% tau2 = 172.58 Chi2 = 100.09 df = 10 P < 0.00001 | Moderate: 6.6 | ||||
Pedersen 2013 [17] | SR without MA of RCTs and cohort studies Published between 01/2000 and 12/2011 Intervention duration: 7d–3 wks follow-up: 7–10 yrs | Both sexes Generally healthy and/or people with overweight or obesity Mean age: 24–70 yrs | RCTs: high vs. normal/low protein intake Cohort studies: En% from protein or total protein intake (in kcal or servings) | 1–1.5 g/kg BW/d vs. 2–3 g/kg BW/d | NA | Moderate | |||
2 RCTs | n = 48 | Total protein, animal protein | GFR | "The evidence is assessed as inconclusive regarding the relation of protein intake to renal function based on GFR.” | low: 5.0 | ||||
2 cohort studies | n = 10,216 | Total protein | eGFR | Low: 3.0 | |||||
2 RCTs | n = 48 | Total protein | Microalbuminuria | "The evidence is assessed as inconclusive regarding the relation of protein intake to renal function based on microalbuminuria.” | Low: 4.5 | ||||
Animal protein | Low: 5.0 | ||||||||
2 cohort studies | n = 10,216 | Total protein | Low: 3.0 | ||||||
2 cohort studies | n = 141,864 | Animal protein | Kidney stones | "The evidence is assessed as inconclusive regarding the relation of protein intake to risk of kidney stones” | Low: 3.0 | ||||
Santesso 2012 [19] | SR with MA of 2 RCTs Published before 08/2011 Intervention duration: 84 d | Both sexes People with overweight or obesity Mean age: 46–58 yrs n = 67 | High vs. low protein (≥ 5% difference in total energy intake) | Median: 27 En% vs. 18 En% Range: 16–45 En% vs. 5–23 En% | Serum creatinine | Pooled SMD (95% CI), random effects model 6.14 (2.49, 9.79) P = 0.001 | I2 = 0% tau2 = 0.00 Chi2 = 0.10 df = 2 P = 0.95 | Very low: 2.0 | High |