A Cochrane analysis of the evidence for thiazolidinediones/glitazones in people with CKD and type 2 diabetes has found little to impress with respect to critical outcomes of cardiovascular death and severe hypoglycaemia.

The meta-analysis of 85 RCTs, representing 3,044 participants with all stage CKD and type 2 diabetes, found that the glucose-lowering agents may have little or no effect on the risk of cardiovascular death compared to placebo or standard care alone.

The absolute risk of CV death with thiazolidinediones, mostly pioglitazone and rosiglitazone, reduced from 18 per 1,000 to 4 per 1000 at 14 weeks (RR 0.20).

Meanwhile, it found their effects on severe hypoglycaemia or other cardiovascular and kidney outcomes were not estimable in people with CKD and type 2 diabetes given no events in these categories.

“Combining all the studies, we found that we still don’t know whether treatment with thiazolidinediones helps to lower a person’s chances of death directly due to a heart problem or severe hypoglycaemic event…,” it said in a plain language summary.

Among the findings, the Cochrane review [link here] found:

• Thiazolidinediones may have little or no effect on CV death compared to sulphonylureas in people with CKD stages 1 and 2 and type 2 diabetes, but no studies evaluated the effects of severe hypoglycaemia, or other cardiovascular or kidney critical outcomes.
• The effects of thiazolidinediones compared to other hypoglycaemic agents (e.g. DPP‐4 inhibitors, metformin, alpha‐glucosidase inhibitors, or meglinitides) or different doses of thiazolidinediones were uncertain.
• Adverse events, including oedema, lactic acidosis, blindness, diabetic ketoacidosis, hypoglycaemia, and fracture were rarely and inconsistently reported in the included studies.

It said KDIGO guidelines [link here] currently recommend thiazolidinediones as an additional treatment option in people with CKD and type 2 diabetes in combination with first‐line therapy of SGLT2 inhibitors and metformin.

Yet the uncertainty of the evidence, including safety concerns, warranted more research, it said.

“The included studies reported an overall high risk of bias because of a lack of blinding, uncertainty in the randomisation process, bias in deviations from the intended interventions, or selective reporting of key critical and important outcomes.”

“In some studies, the risk of bias for cardiovascular death and severe hypoglycaemia was considered at high risk of bias due to concerns about blinding, or these critical outcomes were not available for some treatment comparisons.”

The Cochrane review said included studies also had a small sample size or focused on short‐term follow‐up treatment. As well, studies rarely reported key patient‐reported outcomes such as fatigue or life participation.

In combination, the low quality evidence limited the ability to inform clinical decision‐making.

“Future well‐designed and powered RCTs should be specifically designed and conducted to assess the effects of thiazolidinediones on treating people with CKD and type 2 diabetes. Future studies on head‐to‐head comparisons are needed, and they could increase our certainty of the current evidence,” it said.

The review was led by assistant managing editor of Cochrane Kidney and Transplant Patrizia Natale from the University of Sydney. Senior investigator was nephrologist Professor Giovanni Strippoli, who also has affiliations with the University of Sydney and is deputy coordinating editor of Cochrane Kidney and Transplant. Dr David Tunnicliffe from the University of Sydney Centre for Kidney Research based at The Children’s Hospital at Westmead was another local author on the review.