Renal glucosuria is associated with lower body weight and lower rates of elevated systolic blood pressure: results of a nationwide cross-sectional study of 2.5 million adolescents

Kidneys play a major role in glucose homeostasis. Up to 50% of fasting gluconeogenesis may be attributed to the kidney cortex [1, 2]. In a normal state, the kidneys reabsorb all of the glucose filtered by the glomeruli (~ 180 g/day). Elevated blood glucose levels reaching the kidneys’ maximum reabsorption capacity, result in glucosuria. The process of glucose reabsorption occurs in the proximal tubules of the nephron by transmembrane transporters of two gene families: the glucose transporters (GLUTs) and the sodium coupled glucose transporters (SGLTs) [3, 4]. The former and specifically, the GLUT2 and GLUT1 found in the proximal tubule epithelial cells, are a group of transporters located on the basolateral membrane, facilitating the passive transport of glucose back into the circulation [5, 6]. The SGLT2, located on the apical side, is a transmembrane transporter responsible for > 90% of the glucose reabsorption from the glomerular filtrate, whereas, the SGLT1 reabsorbs the remaining 10% [4]. There are at least 50 autosomal dominant and autosomal recessive mutations in the SGLT2 coding gene (SLC5A2), underlying familial renal glucosuria (FRG) [7, 8]. Although FRG may cause glucosuria of > 100 g/day, the affected individuals are asymptomatic without an increased risk of diabetes mellitus (DM) or urinary tract infections (UTIs) [2, 8].

The benign nature of FRG, as well as the nearly exclusive expression of SGLT2 in the kidney, has rendered the SGLT2s an intriguing pharmacological target in the development of novel hypoglycemic medications [1, 9]. Subsequently, over the past few years, several SGLT2 inhibitors have been approved by the US Food and Drug Administration for treating patients with type 2 diabetes [10]. Empagliflozin, dapagliflozin and canagliflozin have been shown to be effective oral hypoglycemic agents, but more importantly, these agents were found to be beneficial in reducing cardiovascular morbidity and mortality as well as the progression of diabetic renal diseases [11‐14]. One of several proposed mechanisms underlying these positive dramatic effects on cardiovascular-related outcomes are weight reduction and blood pressure (BP) lowering effects of the SGLT2 inhibitors observed in pre-clinical and clinical trials [11‐13, 15].

The aim of our study was to illustrate in a large population-based cohort of Israeli adolescents, the prevalence of non-diabetic renal glucosuria (RG) and investigate whether these subjects were less likely to exhibit cardiovascular risk factors such as increased body weight and elevated BP levels.

Detailed baseline characteristics of the study cohort are presented in Table 1. The final study cohort was comprised of 2,506,830 individuals of whom 1108 (0.044%) were diagnosed with RG unrelated to DM (Table 1); 74.6% participants diagnosed with RG were males, whereas, in the non-RG population, males comprised only 58.6% of the cohort (Table 1). The study cohort diagnosed with RG were from the same SES and had similar distributions of educational levels and ethnic origins compared to the non glucosuric participants (Table 1). RG was more prevalent in males (0.056%) than in females (0.027%) with a crude OR of 2.08 (95% CI 1.81–2.38) and an aOR of 1.96 (95% CI 1.70–2.25) (Table 1).

This nationwide cross-sectional study comprising 2,506,830 individuals is the largest study to date describing the prevalence of non-DM associated RG amongst adolescents. The prevalence of RG in our entire cohort was 0.044%, whereas, RG was more common amongst males, with a nearly twofold increase in the adjusted risk for RG compared to females. Our findings slightly differ from the results of two recent Southeast Asian studies, where the rates of RG were 0.05–0.07%, with similar rates appearing amongst males and females [18, 19]. This difference may stem from the genetic variability of the Southeast Asian population compared to the Israeli subjects. Furthermore, we showed that glucosuric adolescents were less likely to be overweight, obese or exhibit an elevated systolic BP.

Renal glucosuria is an uncommon, albeit well-recognized medical condition, with a benign clinical course [5, 7]. Over the past three decades, enormous progress has been made in discovering the underlying mechanisms for glucose reabsorption from the ultra-filtrate in the proximal tubules of the nephron [6, 15]. Previous studies have demonstrated the mutations in the gene expressing SGLT2 transmembrane transporter responsible for most cases of RG [5, 7]. These inhibitors are beneficial in reducing cardiovascular morbidity, mortality and the progression of diabetic renal [11‐13, 20, 21]. One of the proposed mechanisms underlying these cardiovascular beneficial effects is reduced weight and BP values [15]. Dapagliflozin was shown to reduce total body weight and fat, particularly in visceral and subcutaneous adipose tissue [22‐24]. Furthermore, in patients with type 2 diabetes, canagliflozin treatment for 52 weeks generated a 4 kg reduction in weight, most attributable to fat tissue loss. Weight reduction effects remained significant even after 2 years of follow-up [25]. Similar effects on body weight reduction were also noted following empagliflozin treatment [24, 26]. Correspondingly, in our study, glucosuric participants in the general cohort, males in particular, were less likely to be overweight or obese. This effect is probably best explained by the caloric loss caused by vast glucose urination [10, 15].

This hypothesis is further supported by data showing that patients treated with dapagliflozin may lose up to 300 k/cal daily [27]. Moreover, treatment with empagliflozin [28] and dapagliflozin [29, 30] were both found to lower SBP and DBP. Furthermore, a recent large meta-analysis of 27 randomized clinical trials, including ~ 13,000 participants, revealed that treatment with several SGLT2 inhibitors was associated with a mean reduction in SBP and DBP of 4 mmHg and 1.9 mmHg, respectively, without significant postural adverse effects [31].

In our study, we found that glucosuric subjects were less likely to experience an increased SBP in the 130–139 mmHg range. There was no significant protective effect when SBP was > 139 mm/Hg, probably due to the low statistical power, since there were only < 50 glucosuric adolescents with a SBP of > 139 mmHg. The BP reduction may be attributed to an osmotic diuresis effect as evidenced by the large amounts of glucose and sodium measured in the patient’s urine treated with these agents [10, 31‐33].

It is noteworthy that all subjects in our study were examined to exclude DM and despite their normal blood glucose levels, they were less likely to be overweight or obese. Our results strengthen the assumption that most of the weight loss and BP reduction effects of the SGLT2 inhibitors are mediated through the glucosuria. This effect was manifested in all age groups, in non-diabetics and probably even more so in diabetic subjects whose potential loss of glucose is more significant and therefore, may have more significant effects on BMI and BP. A recent study of a Japanese middle-aged and elderly population found that glucosuric subjects (without a diagnosis of DM) had a higher mean BMI, SBP and DPB compared to non-glucosuric subjects [34]. However, the glucosuric group had a much higher mean creatinine level and higher rate of proteinuria and hematuria. Therefore, the glucosuria in this population was almost certainly the result of kidney disease in an older population. Contrary to this, our glucosuric subjects were adolescents and their glucosuria was probably due to a congenital abnormality rather than an acquired abnormality related to kidney disease. Our data raise the possibility of a beneficial effect of SGLT2 inhibitors in patients with prediabetes, mainly, those with a metabolic disease in order to prevent a progression to diabetes and a reduced cardiovascular risk. It is essential for future studies to prove this claim.

Our study has several notable limitations. Firstly, we had no data as to the severity of glucosuria amongst examinees diagnosed with RG. Secondly, study participants were analyzed as a homogenous group regardless of the specific genetic mutation associated with RG. Theoretically, different mutations may be associated with varying degrees of glucosuria, thereby, inducing variable body weight and BP reductions. Furthermore, our glucosuria estimation was qualitative and not quantitative, hence, we could not draw any conclusions as to the effects of increasing urine glucose and any measured outcomes. Thirdly, the BP values analyzed in our study were derived from a single BP office measurement taken in a potentially stressful environment (known to increase the probability of white coat-associated elevated BP measurement) [35]. Yet, most previous studies evaluating hypertension amongst adolescents, based their diagnosis upon only one BP measurement. Moreover, all examinees were uniformly exposed to the same stressful environment and therefore, the difference in BP values between subjects diagnosed with and without RG cannot be related to the stress.

To the best of our knowledge, the current study is the largest trial to date evaluating the prevalence of RG in a late-adolescent population without DM. Our data indicate that in adolescents, RG is more common in males and is significantly associated with lower body weight, less obesity and lower rates of SBP values in the hypertension range. These findings may support the assumption that SGLT2 inhibitors may reduce weight and BP in non-diabetic individuals. Well-powered large scale studies are essential in order to clarify whether these RG-associated beneficial metabolic effects in adolescence have any ramifications on future cardiovascular morbidity and mortality. Well-powered large scale studies are essential in order to clarify whether these RG-associated beneficial metabolic effects in adolescence have any ramifications on future cardiovascular morbidity and mortality rates.