Effects of antidiabetic drugs on the level of serum uric acid in patients who have type 2 diabetes

Metformin

Metformin is the most commonly administered anti-diabetic drug, although very few studies have examined its effects on the SUA level in patients with T2DM. However, there is some evidence that metformin decreases the SUA level of these patients. For example, Gregorio et al²⁰ assessed the effect of the addition of metformin to 76 elderly patients with T2DM (31 men, 45 women; mean age: 76.44±1.03 years) that was poorly controlled by sulfonylurea for a one-year period. Their results showed that metformin decreased the SUA level from 5.88±0.25 to 5.02±0.28 mg/dL (p<0.05). Another controlled study examined the effect of metformin in patients with gout and insulin resistance (IR; 28 men and 2 women; mean age: 51 years). Their results showed that it lowered the SUA concentration from 9.57±1.84 to 7.44±1.81 mg/dL (p<0.01), and that 11 patients developed normouricemia (<6.0 mg/dL) after 12 months of metformin therapy.²¹ Notably, the authors of this study found that the normouricemic effect of metformin was unrelated to the renal excretion of UA or decreased body weight, and they hypothesized that metformin decreased UA production in these patients due to metformin-mediated decrease in the synthesis of free fatty acids.²¹ Another study that examined 50 obese females (mean age: 43.58±1.40 years) reported that the SUA level decreased from 5.52±0.20 to 4.78±0.12 mg/dL (p<0.0001) after 6 months of metformin treatment.²² Krzystek-Korpacka et al²³ implemented a one-year weight reduction program for children and adolescents who were overweight or obese, and examined its effect on the SUA level. They reported that the efficacy of metformin in decreasing the SUA level was apparently independent of weight loss.

There is also evidence that a high SUA level is linked with IR. For example, the decreased renal clearance of UA in patients with hyperinsulinemia is related to IR, and other evidence suggested that IR led to increased synthesis of free fatty acids, and that these were eventually metabolized into UA.^24-26 Metformin treatment of patients with T2DM leads to decreased IR and hyperinsulinemia, and this may lead to increased urinary clearance of UA, decreased synthesis of free fatty acids, and a decreased level of SUA.²⁷ There is also evidence that metformin can lead to weight reduction and appetite suppression. Taken together, normouricemia appears to be one of the many beneficial effects of metformin.

Sulfonylureas

The sulfonylureas have uricosuric effects, but they differ in the strength of this effect.²⁸ However, none of the sulfonylureas have known normouricemic effects.^29-31 A study of 29 patients with T2DM (19 men, 10 women; mean age: 56.1 years) showed that 12 months of gliclazide treatment did not alter the SUA level.³⁰ Another study of T2DM patients (17 men, 3 women; mean age: 65 years) showed that a 12-week regimen of gliclazide led to a slightly increased level of plasma UA in the fasting state (5.2±0.9 to 5.6±0.9 mg/dL; p=0.02), but the UA level was unchanged during hyperglycemic conditions.³² Hussain et al³¹ carried out a case-control study of 40 patients with T2DM and showed that glibenclamide did not have normouricemic effects. Kitazawa et al³³ carried out a prospective randomized controlled trial (RCT) of Japanese patients with T2DM (19 men, 12 women; mean age: 57.6±9.3 years) to compare glimepiride with tofogliflozin as the third oral agent added to a regimen consisting of metformin with a dipeptidyl peptidase 4 inhibitor (DPP4i). They found that the 24-week glimepiride regimen did not significantly alter the SUA level. Skillman et al²⁹ hypothesized that sulfonylureas did not have uricosuric or normouricemic effects because these agents simply lacked this activity or because the serum concentrations were too low. It is also possible sulfonylureas may initially decrease the SUA level, but that this effect decreases over time.

Dipeptidyl peptidase 4 inhibitors

A study of 120 patients with T2DM (72 men, 48 women; mean age: 63.2±13.8 years) reported that sitagliptin significantly increased the SUA level after 12 weeks (5.08±1.14 to 5.30±1.24 mg/dL; p=0.001).³⁴ Another study of 64 drug-naïve patients with T2DM (50 men; 14 women; mean age: 56.0±12.3 years) found that 3 months of sitagliptin increased the SUA level 4.91±1.28 to 5.42±1.43 mg/dL (p<0.00001).³⁵ Tojikubo et al³⁶ studied the effect of sitagliptin in 73 patients with T2DM and found that it led to a significant increase in the SUA level (5.10±1.13 to 5.63±1.24 mg/dL; p<0.001) and significant decrease in the estimated glomerular filtration rate (eGFR: 78.0±24.4 to 71.8±23.3 mL/min/1.73 m²; p<0.001) after at least one year of treatment. A RCT of 163 patients with T2DM found that the SUA increased 5.4±1.4 to 5.6±1.3 mg/dL (p=0.004) after 24 weeks.³⁷ Another study of alogliptin therapy in 55 drug-naïve individuals with T2DM (46 men, 9 women; mean age: 52.4±13.2 years) reported an increased SUA level from 4.69±1.60 to 5.24±1.61 mg/dL (p<0.00001) after 3 months of treatment.³⁵

In contrast, a study that assessed the effect of changing from sitagliptin to linagliptin in 73 subjects who had T2DM (41 men, 32 women; mean age: 66±13 years) showed that linagliptin lowered the SUA concentration after one year (5.63±1.24 to 5.24±1.10 mg/dL; p<0.001) and led to no significant decrease in the eGFR.³⁶ Yamagishi et al³⁸ examined 26 patients with T2DM (18 men, 8 women; mean age: 69.4±12.4 years) and reported that linagliptin decreased the SUA level after 24 weeks (5.5±1.2 to 5.1±1.2 mg/dL; p<0.05). These same researchers also confirmed that linagliptin inhibited xanthine oxidase activity in vitro, and suggested that this effect was partly responsible for the UA-lowering effect of linagliptin.³⁸ Another retrospective study of T2DM patients found that switching from sitagliptin to vildagliptin led to a decreased serum level of UA.³⁹

The precise mechanism by which DPP4is alter the SUA level remains unknown, and it is also unknown why some of these drugs increase the UA level but others decrease the UA level. One suggestion is that the inconsistent effects of different DPP4is may be attributed to their different chemical structures and pharmacological properties.^36,38 Regardless of the mechanisms of the different drugs and the reasons for their differences, it seems that linagliptin is more effective than other DDP4is for controlling hyperuricemia in patients with T2DM.

Alpha-glucosidase inhibitors

Moriwaki et al⁴⁰ provided evidence that acarbose can attenuate the increase of plasma UA induced by sucrose ingestion. More specifically, these authors examined 6 healthy subjects and reported that the pre-administration of acarbose attenuated the increase of plasma UA that occurred after sucrose ingestion, although this treatment did not alter the urinary excretion or fractional clearance of UA. These authors hypothesized that this effect may be due to the inhibition of intestinal sucrose absorption, which delayed its conversion into fructos.⁴⁰ One study of 121 T2DM patients reported that voglibose had no effect on the SUA level.³⁴ Based on the mechanism and site of action of these drugs, we speculate that they probably do not have a clinically significant impact on the SUA concentration. However, further studies of the mechanisms of these drugs and their effects on the SUA level are warranted.

Thiazolidinediones

A review article that analyzed the pleiotropic effects of thiazolidinediones reported that these drugs had urate-lowering effects.⁴¹ More specifically, a study of 95 T2DM patients (61 men, 34 women; mean age: 61.1±10.3 years) found that troglitazone significantly decreased the SUA level (5.5±1.2 to 5.0±1.2 mg/dL; p<0.0001) after 9.7±7.6 months of treatment.⁴² Similarly, a study of 40 T2DM patients reported that a 12 week regimen of rosiglitazone decreased the SUA level from 4.78±1.1 to 4.41±1.1 mg/dL (p=0.001).⁴³ Another study of 21 T2DM patients with metabolic syndrome found that a 12-week regimen of rosiglitazone decreased the SUA level from 6.22±1.31 to 6.02±1.27 mg/dL (p<0.015).⁴⁴ Kutoh et al⁴⁵ examined 19 drug-naïve patients with newly diagnosed T2DM whose baseline SUA levels were above 6.0 mg/dL. They reported that a 12-week regimen of pioglitazone significantly lowered the UA level (6.87±0.73 to 5.90±0.77 mg/dL; p<0.00001).⁴⁵

It is possible that thiazolidinediones decrease the SUA level due to their attenuation of IR, because IR is associated with an elevated level of SUA.⁴⁵ An alternative mechanism was suggested in a study of UA stone formers.⁴⁶ This study showed that pioglitazone therapy increased urine pH, and the authors suggested that the resulting increased solubility of UA led to enhanced urinary excretion of UA and a lower serum level of UA.

Insulin

A matched cohort study reported that insulin significantly increased the SUA level by 1.25 mg/dL (p=0.02).⁴⁷ This effect is likely because insulin increases the renal reabsorption of urate due to its activation of urate transporter-1 (URAT-1) or the sodium-dependent organic anion co-transporter (SOAT) in the renal proximal tubules.⁴⁷ Another study of healthy individuals found that exogenous insulin led to an acute decrease of fractional renal urate excretion in a euglycemic clamp experiment.⁴⁸ Other studies reported that the insulin-induced increase of SUA may increase the risk of gout.^49,50

Glucagon-like peptide-1 receptor agonists

A study of T2DM patients with obesity (13 men, 9 women; mean age: 51±2 years) showed that a 26-week regimen of exenatide had no significant effect on the SUA level.⁵¹ Similarly, Muskiet et al⁵² determined the effect of a 52-week regimen of twice-daily exenatide in T2DM patients (16 men and 10 women; mean age: 59.7±8.1 years) and found no effect on the SUA level. Tonneijck et al⁵³ carried out a post-hoc analysis of 4 previous clinical trials and examined the short-term response to exenatide. The acute effect was a mild elevation of SUA in healthy individuals, although there was no effect in patients with T2DM.⁵³ This post-hoc analysis also examined the effect of a 12-week liraglutide regimen in T2DM subjects (27 men, 9 women; mean age: 63±7years) and found no significant changes relative to placebo.⁵³ A RCT by Nakaguchi et al⁵⁴ evaluated the effect of a 24-week liraglutide regimen on the SUA level of patients with T2DM (21 men, 9 women; mean age: 67.2±9.0 years) and reported no significant effect. In agreement, Kurir et al⁵⁵ and Liakos et al⁵⁶ carried out before-after studies in populations of T2DM patients and reported that liraglutide had no significant effect on the SUA level.

A randomized, open-label, comparator-controlled trial of T2DM patients that examined an 8-week regimen of lixisenatide injections showed that this treatment had no effect on the SUA level.⁵³ Several before-after studies of patients with T2DM reported that dulaglutide had no effect on the SUA level. In particular, Kuchay et al⁵⁷ investigated the effect of dulaglutide on the SUA level of patients who had T2DM and nonalcoholic fatty liver disease (NAFLD) and found it had no significant effect. Hirai et al⁵⁸ and Iwasaki et al⁵⁹ reported similar findings. However, an animal study that examined a rat model of diabetes showed that rats receiving GLP-1 RAs had lower SUA levels than untreated rats.⁶⁰ Taken together, the current data provide no support for the view that GLP-1 RAs can lower the SUA level. However, very few studies were specifically designed to examine the effects of GLP-1 RAs on the SUA level, not all GLP-1 RAs have yet been studied, many of the available studies had limited durations, and different GLP-1 RAs may have different effects on SUA. Further studies that examine multiple variables, including different GLP-1 RAs, treatment duration, and dosage, are required to ascertain the impact of individual GLP-1 RAs on the SUA level.

Sodium-glucose cotransporter 2 inhibitors

There is evidence that multiple drugs in the class of sodium-glucose cotransporter 2 inhibitors (SGLT2is) can promote normouricemia. For example, a meta-analysis that examined 10 RCTs and 2 observational studies of T2DM patients (n=7801) found that empagliflozin significantly decreased the SUA level compared with placebo (standardized mean difference [SMD]: -1.34 mg/dL, 95% CI: [ -2.05, -0.63]).⁶¹ A RCT that assessed the effect of dapagliflozin in patients with T2DM (20 men, 9 women; mean age: 57.77±12.29 years) showed that this drug significantly lowered the SUA level (5.86±1.72 to 4.71±1.20 mg/dL; p<0.001).⁶² Similarly, Fuchigami et al³⁷ carried out a prospective randomized trial of 168 patients with T2DM and showed that a 24-week dapagliflozin regimen significantly decreased the SUA level (5.4±1.3 to 4.9±1.1 mg/dL; p<0.001). A 2022 RCT of 24 patients with T2DM measured the plasma UA level, fractional UA excretion, and hemodynamic function of the kidneys during fasting state and during clamped euglycemia or hyperglycemia before and after a 12-week dapagliflozin regimen.³² The results showed that dapagliflozin decreased the plasma UA level by 0.8±0.8 mg/dL during fasting, by 1.0±1.0 mg/dL during hyperinsulinemia-euglycemia, and by 08±0.7 mg/dL during the hyperglycemia (all p<0.001). This study also showed that the fractional daily UA excretion increased by 3.0±2.1% (p<0.001) and by 2.6±4.5% during hyperinsulinemia-euglycemia (p=0.003).³²

A post-hoc study of data from 4 placebo-controlled phase III trials of 2313 patients with T2DM found that canagliflozin decreased the SUA level by ~13% (0.7 mg/dL) relative to placebo at week 26.⁶³ Similarly, Seino et al⁶⁴ reported decreases in the SUA after luseogliflozin administration in T2DM subjects. Chino et al⁶⁵ stratified T2DM patients into 3 groups according to baseline level of glycosylated hemoglobin (HbA1c) and into 4 groups according to baseline level of SUA. Their multivariate analysis showed that the luseogliflizin-mediated decline of SUA was associated with a higher baseline SUA level and a lower HbA1c level.⁶⁵ One RCT of Japanese subjects who had T2DM showed that tofogliflozin decreased the SUA level.⁶⁶ Similarly, 3 other studies in Japan also found that ipragliflozin had significant SUA-lowering effects compared with controls.^67-69

Several meta-analyses also examined the results of administering drugs in this class on the SUA level. For example, Zhao et al⁷⁰ reviewed 62 RCTs (34,941 patients) that examined patients with T2DM, and found that multiple SGLT2is (empagliflozin, canagliflozin, dapagliflozin, ipragliflozin, luseogliflozin, and tofogliflozin) significantly decreased the SUA level relative to controls (range: -0.29 to -0.77 mg/dL, total weighted mean difference [WMD]: -0.63 mg/dL). Among these many SGLT2is, Zhao et al⁷⁰ reported that empagliflozin had the strongest effect (WMD: -0.77 mg/dL). A different systematic review and meta-analysis that included 19 RCTs (4218 patients) examined the response to SGLT2is in Asian T2DM patients. This study showed that SGLT2is significantly decreased the SUA concentration compared with the control treatment (total SMD: -0.965; 95% CI: [-1.029, -0.901]; p<0.001; I2=98.7%).⁷¹ Furthermore, a network meta-analysis by these researchers that compared different specific drugs with placebo showed that the UA-lowering effects of these drugs were as follows: luseogliflozin (SMD: -1.28, 95% credible interval [CrI]: [-1.70, -0.86]), dapagliflozin (SMD: -0.97, 95% CrI: [-1.40, -0.55]), empagliflozin (SMD: -0.67, 95% CrI: [-1.10, -0.24]), tofogliflozin (SMD: -0.59, 95% CrI: [-1.02, -0.16]), ipragliflozin (SMD: -0.57, 95% CrI: [-0.99, -0.14]), and canagliflozin (SMD: -0.51, 95% CrI: -0.95, -0.07]). Their further analysis showed that luseogliflozin at 1 or 10 mg/day and dapagliflozin at 5 mg/day had superior effects.

Akbari et al⁷² carried out a meta-analysis of 55 placebo-controlled clinical trials of 36,215 patients who received SGLT2is (n=23,494) or placebo (n=12,721). In agreement with the other studies, they found that all examined SGLT2is significantly decreased the SUA level compared with placebo. They reported the UA-lowering effects were as follows: empagliflozin (mean difference [MD]: -0.69 mg/dL, 95% CI: [-0.80, -0.58]), dapagliflozin (MD: -0.59 mg/dL, 95% CI: [-0.67, -0.52]), canagliflozin (MD: -0.61 mg/dL, 95% CI: [-0.70, -0.52]), luseogliflozin (MD: -0.41 mg/dL, 95% CI: [-0.56, -0.26]), tofogliflozin (MD: -0.33 mg/dL, 95% CI: [-0.46, -0.19]), and ipragliflozin (MD: -0.32 mg/dL, 95% CI: [-0.46, -0.18]).⁷²

A meta-analysis by Yip et al⁷³ examined 43 RCTs (31,921 subjects) to compare the effect of SGLT2is relative to placebo on the SUA level in patients who did or did not have T2DM. They identified decreases in the SUA level in subjects with diabetes (-0.53 mg/dL; 95% CI: [-0.63, -0.43]) and in those without diabetes (-1.54 mg/dL; 95% CI: [-2.13, -0.95]). Furthermore, a 2024 meta-analysis of 5 trials that had a median follow-up time of 2.2 years examined the effect of SGLT2is in 31,535 patients, 76% with T2DM and 54% with heart failure. Overall, the mean change of SUA was -0.79 mg/dL (95% CI: [-1.03, -0.54]) and each 1 mg/dL decrease of SUA was linked to a lower composite risk of cardiovascular death and hospitalization for heart failure (HR=0.64; 95% CI: [0.46-0.88]).¹⁷ Two recent studies of canagliflozin and erutgliflozin found that the magnitude of the decrease in the SUA level had a negative relationship with adverse kidney outcomes.^74,75 Three other large-scale clinical trials of patients who had diabetes and heart failure reported that SGLT2is decreased the risk for gout by 30-50%.^76-78

Several studies have examined the possible mechanism by which SGLT2is decrease the SUA level. One study suggested that this effect may be attributable to the induction of decreased flux through the pentose phosphate pathway and increased renal excretion of urate.⁷⁹ In particular, because drugs in this class block the reabsorption of glucose by SGLT-2, which has high expression in the apical membranes of renal segments S1 and S2 (regions of the proximal convoluted tubules), this leads to a decreased level of glucose, activation of sirtuin-1, inhibition of xanthine oxidase, and decreased synthesis of UA.⁸⁰ Sodium-glucose cotransporter 2 inhibitors may also increase the urinary excretion of urate by altering the function of UA transporters in the renal tubules, such as glucose transporter 9 (GLUT9) and URAT-1. A study of healthy male volunteers in Japan and Xenopus oocytes suggested that the SGLT2i-mediated increase in tubular glucose leads to trans-activation of a GLUT9 isoform, and then to urate efflux in the renal proximal tubules.⁸¹ In addition, the results from clinical trials and animal studies suggest that SGLT2is may increase the excretion of UA due to their blockade of URAT1, which is located on the apical surface of the renal proximal tubules and functions in urate reabsorption after glomerular filtration.³² Finally, SGLT2is may also upregulate the ATP-binding cassette transporter G-2 (ABCG-2), thus promoting urate secretion.⁸² Further studies are needed to examine other potential mechanisms by which drugs in this class decrease the SUA level and compare the effects of different specific drugs.

Conclusion

The pathological processes of T2DM and hyperuricemia are interconnected, and each disorder increases risk for many other diseases, including cardiovascular disease and kidney diseases. Because so many other diseases are associated with hyperuricemia, it is likely that patients with T2DM will benefit by maintaining normouricemia. Fortunately, many of the available antidiabetic drugs also decrease the serum level of UA. Among the various types of antidiabetic drugs, SGTL2is have the best-documented effects on SUA in patients with T2DM. The many studies that support the use of SGTL2is for these patients include well-designed before-after studies, large RCTs, and several meta-analyses. However, for some other classes of antidiabetic drugs, there are relatively few clinical studies that examined SUA as a primary outcome; instead, these studies examined SUA levels as a secondary research objective or using a secondary analysis. Therefore, more well-designed studies are required to assess the effects of these other classes of antidiabetic drugs on the serum level of UA. Moreover, further investigations are needed to examine the underlying mechanisms by which the different antidiabetic drugs decrease SUA.

This review provides strong evidence that administration of SGTL2is to patients with T2DM has an added benefit of decreasing SUA levels. This indicates a significant advantage for this class of drugs when considering treatments for T2DM patients, especially for patients who also have hyperuricemia. Studies of the mechanism of SGLT2is suggest they may lower the SUA by inhibiting the pentose phosphate pathway, and by increasing the excretion of urate by the kidneys due to alteration of renal tubular UA transporters (GLUT9 and URAT1). Overall, the GLP-1 RAs had no significant impact on the SUA level; however, further studies are required to examine the effects of individual GLP-1 RAs on SUA. In addition, thiazolidinediones, metformin, and linagliptin potentially decrease the SUA, whereas insulin, sitagliptin, and alogliptin all appear to increase the SUA. Sulfonylureas and alpha-glucosidase inhibitors appear to have no significant impact on the SUA. We suggest that the results presented here be considered when developing future therapeutic guidelines for patients who have T2DM with hyperuricemia.