The American College of Physicians ( ACP ) has developed a guideline to present the evidence and provide clinical recommendations on the comparative effectiveness and safety of type 2 diabetes medications.
Hypoglycemia - No particular monotherapy or combination therapy has increased severe hypoglycemia ( generally defined as hypoglycaemia requiring assistance for resolution ) compared with the other treatments.
Monotherapy versus Monotherapy - Pooled results from monotherapy trials have shown that sulfonylureas increase the risk for mild to moderate hypoglycemia compared with Metformin ( odds ratio [ OR ], 4.60; I2=68%; high-quality evidence), thiazolidinediones ( OR, 3.88; I2=41%; high-quality evidence ), and meglitinides ( OR, 0.78; I2=18%; low-quality evidence ).
Data from RCTs ( randomised controlled trials ) have also indicated that other agents were favored over sulfonylureas for hypoglycemia: DPP-4 inhibitors ( data from 1 RCT showed that 21 of 123 patients treated with a sulfonylurea had mild or moderate hypoglycaemia compared with no patients treated with a DPP-4 inhibitor; moderate-quality evidence ) and GLP-1 agonists ( data from 3 RCTs; high-quality evidence ).
Monotherapy with meglitinides resulted in more hypoglycaemia compared with Metformin ( OR, 3.00; I2=0%; moderate-quality evidence ) or thiazolidinediones ( 2 RCTs: relative risk [ RR ], 1.2; RR, 1.6; low quality evidence ).
Monotherapy versus Combination Therapy - Compared with Metformin monotherapy, the combination of Metformin plus a thiazolidinedione ( OR, 1.57; I2=0%; moderate-quality evidence ), Metformin plus a sulfonylurea ( RR, 1.6 to 25 in 9 RCTs; moderate-quality evidence ), and Metformin plus meglitinides ( OR, 2.75; I2=21%; low-quality evidence; P greater than 0.05) resulted in an increase in hypoglycemia.
Combination Therapy versus Combination Therapy - The combination of Metformin plus a sulfonylurea has increased the risk for hypoglycemia by about 6 times compared with the combination of Metformin plus a thiazolidinedione ( OR, 5.80; I2=0%; high quality evidence ). One large RCT reported that Metformin plus a thiazolidinedione resulted in fewer hypoglycemic events compared with a thiazolidinedione plus a sulfonylurea ( 0.05 vs 0.47 event per 100 person-years of follow-up; low-quality evidence ).
Another study found more hypoglycemic symptoms in patients treated with the combination of Metformin plus a sulfonylurea than with the combination of a thiazolidinedione plus a sulfonylurea ( RR, 1.3; low-quality evidence ).
Other Adverse Effects - Evidence was insufficient to show any difference among the various type 2 diabetes medications on liver injury. Evidence from 51 studies was evaluated to determine gastrointestinal effects. Evidence examined from studies addressing these effects that compared Metformin monotherapy with thiazolidinediones ( high-quality evidence ), sulfonylureas ( moderate-quality evidence ), DPP-4 inhibitors ( moderate-quality evidence ), or meglitinides ( low-quality evidence ) report more gastrointestinal adverse effects with Metformin.
Trials comparing Metformin monotherapy with combination Metformin plus thiazolidinedione therapy ( moderate-quality evidence ) or Metformin plus sulfonylurea therapy ( moderate-quality evidence ), generally favored the combination therapy, although the Metformin dosage was typically lower in the combination group, possibly accounting for this difference.
One RCT reported more dyspepsia with a combination of Metformin plus a meglitinide than with Metformin plus a sulfonylurea ( 13% vs 3%; low-quality evidence ).
Two RCTs reported more diarrhea in combination treatment with Metformin plus a sulfonylurea than with a thiazolidinedione plus a sulfonylurea ( moderate-quality evidence ).
Although few studies reported on congestive heart failure, moderate-quality evidence from 5 observational studies favors Metformin over sulfonylureas, and moderate-quality evidence from 4 RCTs and 4 observational studies favors sulfonylureas over thiazolidinediones. One 6-month observational study reported higher rates of heart failure with the combination of a thiazolidinedione plus a sulfonylurea ( 0.47 per 100 person-years ) than with a thiazolidinedione plus Metformin ( 0.13 per 100 person-years ) ( low-quality evidence ). One RCT has reported that the combination of a thiazolidinedione plus a sulfonylurea or Metformin doubled the risk for heart failure compared with a sulfonylurea plus Metformin ( RR, 2.1; low-quality evidence ).
Evidence was insufficient to show any difference among the various type 2 diabetes medications on macular edema.
One RCT identified 1 person with cholecystitis out of 105 patients treated with a thiazolidinedione compared with none of 100 patients treated with Metformin ( low quality Evidence ). Another RCT identified 1 person with cholecystitis ( n=280 ) treated with Metformin monotherapy compared with no patients ( n=288 ) treated with a combination of Metformin plus a thiazolidinedione ( low-quality evidence ). Low-quality evidence for pancreatitis came from 1 trial that reported 1 patient ( n=242 ) with acute pancreatitis treated with a combination of Metformin plus a sulfonylurea compared with no patients receiving Metformin monotherapy ( n=121 ). The evidence was insufficient to show any difference in cholecystitis or pancreatitis with other monotherapies or combination therapies.
For bone fractures, high-quality evidence from 1 RCT showed more bone fractures with thiazolidinedione monotherapy than with Metformin monotherapy ( hazard ratio [ HR ], 1.57 ), and subgroup analysis showed that the risk is higher for women ( HR, 1.81; P = 0.008 ). Data were assessed from 2 RCTs and 1 observational study, and results showed fewer fractures with sulfonylureas than with thiazolidinediones ( high-quality evidence ). One RCT found an increase in fractures for patients treated with Rosiglitazone compared with a sulfonylurea ( HR, 2.13 ), whereas another study reported 2 ankle fractures ( n=251 ) with Pioglitazone monotherapy and no fractures with sulfonylurea monotherapy ( n=251 ) The observational study found statistically significantly more fractures in women treated with Pioglitazone ( HR, 1.70; P less than 0.001 ) and Rosiglitazone ( HR, 1.29; P 0.02 ) than with sulfonylurea. The combination of Metformin plus a sulfonylurea was favored over the combination of thiazolidinediones plus a sulfonylurea or thiazolidinediones plus Metformin ( RR, 1.57; P 0.001; high quality evidence ), and the RR for fractures was higher for women than men ( 1.82 vs 1.23 ). ( Xagena )
Source: Annals of Internal Medicine, 2012