Combining sodium-dependent glucose co-transporter 2 inhibition with conventional diuretics

Abstract

See original paper on page 893 F amilial renal glucosuria resulting from a defect in the proximal tubular sodium-dependent glucose cotransporter (SGLT)2 is regarded as an innocent condition without associated morbidity [1,2]. To alleviate hyperglycemia in type 2 diabetes, for instance in the setting of obesity and hypertension, the induction of renal glucosuria by SGLT(2) inhibition therefore appears an interesting therapeutic option [3], which, importantly, does not require insulin. SGLT2 mediates the co-transport of glucose and sodium, effectively removing practically all glucose from the primary filtrate [4]. Although quantitatively, the resulting osmotic diuresis should be massive, in humans generally this is limited [5], probably because in humans inhibition of SGLT2 can lead to downstream increases in SGLT1 activity that quantitatively somewhat limits glucosuria [1]. Moreover, mild hypovolemia induced by osmotic diuresis can stimulate sodium reabsorption [6]. Thus, in order to reduce both hyperglycemia and hypertension in the metabolic syndrome a combination of SGLT2 inhibition with combined loop and thiazide diuretics should be very effective. This is the hypothesis tested by Rahman et al. as reported in this issue of the Journal of Hypertension. In a rat model of the metabolic syndrome, the spontaneously hypertensive/ NIH-corpulent rat (SHRcp) rat, they report very promising reductions in both blood glucose and blood pressure by the SGLT2 inhibitor luseogliflozin (L), especially in combination with the diuretics furosemide plus hydrochlorothiazide (D) [7]. Of course, the question is: at what cost are these benefits gained?As compared with vehicle (V), both L and L þ D similarly reduced postprandial blood glucose, HbA1C, as well as blood glucose and plasma insulin during an oral glucose tolerance test. Thus combining an SGLT2 inhibitor with diuretics clearly did not affect its antidiabetic efficacy in this model. However, although both D and L þ D decrease 24-h SBP and mean arterial pressure (MAP) a little more than L alone, the dipper pattern of MAP was restored best by L þ D, reaching pressure differences between active (dark) and nonactive (light) phases of about 7-8 mmHg. These pressure dips are similar to those observed in normotensive Wistar-Kyoto rats or nonobese hypertensive SHR rats. How is this achieved?Weight loss, osmotic diuresis or natriuresis were factors studied by Rahman et al. that could have contributed to the antihypertensive actions of the SGLT2 inhibitors. All three could be involved in the present study. Weight loss was not observed in SHRcp rats but the similar weight increase over the 5-week experimental period in the obese V, D and L rats (about 100 g) was reduced by about 50% in the L þ D rats (Supplemental figures). This is quite astounding because food intake was about 20% higher in the L and L þ D obese rats than in the V and D obese rats. Thus the L and L þ D rats were ingesting considerably more food but failed to put on more weight (L) or even lost weight (L þ D)! This woul...