Table 1.

Some comments on the main RCTs con LPDs enrolling at least 50 patients per arm

Study	Comparators	N and groups	Main results	Limitations	Other comments
Klahr S et al., N Engl J Med. 1994 (MDRD) [39]	LPD vs No diet; LPD vs LPD	Study 1, 585 patients: usual diet: 1.3 g/Kg/day LPD: 0.58 Study 2, 255 patients on or vLPD (plus BP control)	moderate CKD: small benefit LPDs. severe CKD: no difference in ESRD progression on LPDs and vLPD	Highly complex study. The results are given as ITT; however PP analysis shows a significant effect of LPDs, thus highlighting the role of compliance.	The largest RCT on LPDs leading to inconclusive results: it may be also read as measure of the limitations of RCTs analysed as ITT, due to compliance issues
Brunori G, et al. Am J Kidney Dis. 2007 [53]	vLPDs versus dialysis in the elderly (stage 5)	56 patients in each group (296 screened)	vLPDs are effective in delaying the need for dialysis without increasing mortality	Only about 30 % of the initial population accepted being randomised. No information on the follow-up and outcomes of the excluded patients.	The only study randomizing dialysis vs vLPDs; highly relevant even if randomizing such intrusive issues may be perceived as “unethical”
Cianciaruso B, et al. Nephrol Dial Transplant 2008 [57]	0.55 LPD and 0.8 LPD in CKD stage 4–5	200 patients on 0.55 diet, 192 on 0.8 diet (screened 753; initial randomization: 423 pts)	LPD at 0.55 g/kg/day guarantees better metabolic control than a 0.8 diet.	Relatively low compliance in the 0.55 study group (compliant patients: 27 % in the 0.55- Group and 53 % in the 0.8-Group), thus blunting the conclusions. At present 0.8 should be a “normal protein”	Very large study, on two “moderately restricted LPDs: it shows that even within the “moderate restriction range” the lower the better, without risk of malnutrition
Garneata L et al. JASN 2016 [59]	vLPD vs LPD in CKD stages 4–5	vLPD: 104 patients LPD: 103 patients (Screened 1413; non compliance is the main reason for non being randomised)	Better correction of metabolic abnormalities and lower need for dialysis in the vLPD cohort.	Only 14 % of screened patients were randomized. Optimal compliance is a requisite for randomization, indirectly suggesting that these diets are an option for relatively few CKD patients.	The largest recent RCT targeted on supplemented vLPDs vs LPDs. Underlines the importance of vegan diet and of supplementation.

Legend: LPD low protein diet, vLPD very low protein diet, CKD chronic kidney disease, BP blood pressure; numbers indicate the prescribed protein intake per Kg per day