We read the paper by van der Aart-van der Beek et al¹ with great interest and understand that a core message of this secondary analysis of three relatively small-sized randomized controlled trials is that high dietary protein intake does not offset the benefits of sodium-glucose cotransporter-2 (SGLT2) inhibition, also known as “flozination”. However, it may also be true that low protein intake could offer additional synergistic benefits to patients undergoing flozination. We feel that the presented data could have been interpreted differently given that there appear to be clinically relevant and biologically plausible trends in support of combining a low-protein diet with SGLT2 inhibition to improve the salutary effects of this pharmacotherapy approach to reducing albuminuria and protecting kidney health and longevity.

In the DELIGHT trial (n = 233), dapagliflozin versus placebo reduced urinary albumin to creatinine ratio (UACR) by 21% in the high-protein-diet group versus 28% in the low-protein-diet group, suggesting 39% more effect of the SGLT2 inhibitor if combined with low protein intake. The interaction P value was not significant, probably because of the limited sample size. Interestingly, the IMPROVE trial (n = 30) showed a very similar trend in that there was 38% more effect of restricted protein intake added to the effect of dapagliflozin as compared to the pharmacotherapy alone or in combination with high dietary protein intake (Figure 1), although the P value for interaction was not significant. The third study, the DIAMOND trial (n = 53), significant showed a marginal effect of SGLT2 inhibition on UACR (P = 0.072). Dichotomization of dietary protein intake decreases statistical power. Treating dietary protein intake as a continuous variable could be more useful in this context, as methodologically implemented in other studies.²

We feel that these consistent trends from two separate trials of dapagliflozin, even though statistically nonsignificant, are consistent with the well-established data in the past, showing that a low-protein diet enhanced the effect of angiotensin-converting-enzyme inhibition or angiotensin receptor blockade on renal protection.^{3, 4} Indeed, a low-protein diet leads to contraction of the afferent arteriole of the glomeruli, resulting in reduced intraglomerular pressure, hence, opposing the renal hyperfiltration that is invariably harmful to kidneys in the long term⁵; this mechanism is similar to what flozination with such SGLT2 inhibitors as dapagliflozin appears to engender, given the modulation of the tubuloglomerular feedback that abrogates intraglomerular pressure under SGLT2 inhibition.⁶ High dietary protein intake, on the contrary, can aggravate glomerular hyperfiltration and weaken the salutary effect of SGLT2 inhibition on renal haemodynamics.⁷

The additional observation that dapagliflozin exhibited more estimated glomerular filtration rate preservation benefit in patients whose dietary protein intake was classified as high, suggests that high protein intake can cause an even larger effect size, probably by worsening glomerular hyperinflation,⁸ so that the effect of SGLT2 inhibitors becomes more evident, whereas low-protein diets may dilute this effect in patients with chronic kidney disease given mitigation of hyperfiltration by lower dietary protein intake.⁵ However, it is important to note that diet can be wrong in many ways other than its high protein or high meat content, including having excessive calories, inadequate fibre intake or high added preservatives. These and other important nutritional questions can be examined in future studies involving SGLT2 inhibitors in kidney disease with more formal assessments of dietary protein intake on a background of such medications as renin-angiotensin-aldosterone system inhibition and SGLT2 inhibition. What remains even less known is whether plant-based or plant-dominant low-protein diets, including the DASH and PLADO diets, would have an even stronger additive effect on renal outcomes under the SGLT2 inhibitory effects of dapagliflozin.^{9, 10} Given the important data and trends presented in their article,¹ an alternative title for the study by van der Aart-van der Beek et al¹ could have been: “Biologically plausible trends suggesting that a low-protein diet may enhance the effect of flozination on albuminuria”. We advocate well-designed, randomized controlled trials examining the potentially synergistic effect of low-protein and plant-dominant diets on improving renal outcomes including under SGLT2 inhibition therapy.

我们怀着极大的兴趣阅读了 van der Aart-van der Beek 等人^{1 的论文}，并了解到对三个相对较小规模的随机对照试验进行的二次分析的核心信息是，高膳食蛋白质摄入量并不能抵消钠的益处-glucose cotransporter-2 (SGLT2) 抑制，也称为“flozination”。然而，低蛋白质摄入量也可能为接受 flozination 的患者提供额外的协同益处。我们认为所提供的数据可能有不同的解释，因为似乎有临床相关性和生物学合理性支持将低蛋白饮食与 SGLT2 抑制相结合以改善这种药物治疗方法对减少蛋白尿和保护肾脏健康和长寿的有益效果的趋势。

在 DELIGHT 试验 (n = 233) 中，达格列净与安慰剂相比，高蛋白饮食组的尿白蛋白与肌酐比 (UACR) 降低了 21%，而低蛋白饮食组降低了 28%，表明效果提高了 39%如果结合低蛋白质摄入，SGLT2 抑制剂的作用。交互作用P值不显着，可能是因为样本量有限。有趣的是，IMPROVE 试验（n = 30）显示出非常相似的趋势，与单独使用药物疗法或与高膳食蛋白质摄入量相结合，限制蛋白质摄入量与达格列净的作用相比增加了 38%（图 1） )，虽然P交互价值不显着。第三项研究，DIAMOND 试验（n = 53），显着显示了 SGLT2 抑制对 UACR 的边际效应（P  = 0.072）。膳食蛋白质摄入量的二分法降低了统计功效。在这种情况下，将膳食蛋白质摄入量视为连续变量可能更有用，正如其他研究中实施的方法一样。²

我们认为，来自达格列净的两项独立试验的这些一致趋势，尽管在统计上不显着，但与过去公认的数据一致，表明低蛋白饮食增强了血管紧张素转换酶抑制或血管紧张素受体的作用阻断肾脏保护。^{3, 4}事实上，低蛋白饮食会导致肾小球传入小动脉收缩，从而导致肾小球内压力降低，从而对抗肾脏高滤过，长期来看这对肾脏总是有害的⁵; 考虑到在 SGLT2 抑制下消除肾小球内压力的肾小球反馈的调节，这种机制类似于像达格列净这样的 SGLT2 抑制剂似乎产生的絮凝作用。⁶相反，高膳食蛋白质摄入会加重肾小球高滤过并减弱 SGLT2 抑制对肾血流动力学的有益作用。⁷

达格列净在饮食蛋白质摄入量被归类为高的患者中表现出更多估计的肾小球滤过率保留益处的额外观察表明，高蛋白质摄入量可能导致更大的效应大小，可能是通过恶化肾小球过度充气，⁸因此 SGLT2 的作用抑制物变得更加明显，而低蛋白质饮食可能会稀释慢性肾病患者的这种影响，因为通过降低饮食蛋白质摄入量可以减轻过度滤过。⁵然而，重要的是要注意，除了高蛋白质或高肉类含量外，饮食在许多方面都可能是错误的，包括卡路里过多、纤维摄入量不足或添加过多防腐剂。这些和其他重要的营养问题可以在涉及肾脏疾病中的 SGLT2 抑制剂的未来研究中进行检查，并在诸如肾素-血管紧张素-醛固酮系统抑制和 SGLT2 抑制等药物的背景下对膳食蛋白质摄入进行更正式的评估。鲜为人知的是，在达格列净的 SGLT2 抑制作用下，植物性或植物性低蛋白饮食（包括 DASH 和 PLADO 饮食）是否会对肾脏结果产生更强的累加作用。^{9, 10}鉴于他们文章中提出的重要数据和趋势，¹ van der Aart-van der Beek 等人¹的研究的另一个标题可能是：“生物学上合理的趋势表明低蛋白饮食可能会增强 flozination 对蛋白尿的影响”。我们提倡设计良好的随机对照试验，检查低蛋白和植物主导饮食对改善肾脏结局的潜在协同作用，包括在 SGLT2 抑制治疗下。