Summary of findings 4. Glitazone versus placebo for treating people with diabetes and chronic kidney disease (CKD).
| Glitazone versus placebo for treating people with diabetes and CKD | ||||||
| Patient or population: people with diabetes and CKD Intervention: glitazone Comparison: placebo | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Effect estimate (95% CI) | No. of participants (studies) | Quality of the evidence (GRADE) | Comments | |
| Risk with placebo | Risk with glitazone | |||||
| HbA1c (%) HbA1c (mmol/mol) |
The mean HbA1c was 0.41% lower (1.15 lower to 0.32 higher) with glitazone agonists compared to placebo The mean HbA1c was 4.5 mmol/mol lower (12.6 lower to 3.5 higher) with glitazone agonists compared to placebo |
MD ‐0.41 (‐1.15 to 0.32) MD ‐4.5 (‐12.6 to 3.5) |
88 (2) | ⊕⊝⊝⊝ VERY LOW 1 2 3 | ‐ | |
| FBG (mmol/L) | ‐ | ‐ | ‐ | 233 (5) | ⊕⊝⊝⊝ VERY LOW5 6 | Qualitative synthesis of studies showed that glitazones, particularly pioglitazone lowered FBG compared to placebo in patients with an eGFR < 60 mL/min/1.73 m2, including patients on HD. Two studies (total of 71 participants) in people with HD reported that pioglitazone lowered FBG at the end of the study compared to the start, and also lower than in placebo group (both P < 0.05). Similarly another study (39 participants) in HD patients reported that pioglitazone reduced the FBG by 2.91 mmol/L (‐5.44 to ‐0.38 mmol/L); P = 0.02 compared to placebo. Conversely another study in HD patients (63 participants) reported that pioglitazone resulted in a lower FBG (mean ± SD) at the end of the study compared to the start (7.72 ± 2.50 versus 6.89 ± 2.67 mmol/L P < 0.05), but this was not statistically lower than placebo (6.89 ± 2.67 versus 7.33 ± 2.56 mmol/L, P > 0.05). One study of people with earlier stages of CKD (60 participants) showed that in people with stage 3 CKD who were treated with pioglitazone‐losartan, there were higher rates of decline in blood glucose values compared with people treated with losartan only. This difference was significant after 12 months ( change (mean ± SD) after 12 months –22.7 ± 6.9% for pioglitazone‐losartan therapy as compared with –15.1 ± 6.3% for losartan alone; P < 0.01). Larger reductions in FBG concentrations were observed for people in this study with stage 4 CKD after 12 months of the combined as compared with the single‐drug treatment (i.e. –22.9 ± 8.9% versus –17.6 ± 5.9%; P = 0.07), but the difference was not significant. |
| Death (all causes) | 77 per 1,000 | 38 per 1,000 (4 to 398) | RR 0.50 (0.05 to 5.18) | 52 (1 RCT) | ⊕⊝⊝⊝ VERY LOW 1 4 | ‐ |
| All cardiovascular death | ‐ | ‐ | ‐ | ‐ | ‐ | Not reported. |
| Weight (kg) | ‐ | ‐ | ‐ | 222 (5) | ⊕⊝⊝⊝ VERY LOW5 6 | From qualitative synthesis of data from 3 studies (total of 110 participants), pioglitazone did not result in a significant increase of dry weight compared to placebo in patients receiving HD (Abe 2007; Abe 2008a; Pfutzner 2011) or a significant increase of body weight compared to placebo in patients with an eGFR 15 to < 60 mL/min/1.73 m2 (Jin 2007: 60 participants). Conversely, in patients receiving PD (Wong 2005: 52 participants), rosiglitazone resulted in more weight gain (mean ± SD) compared to placebo (2.0% ± 5.6% versus control, ‐0.8% ± 4.4%; P = 0.049). |
| eGFR (mL/min/1.73 m2) | ‐ | ‐ | ‐ | ‐ | ‐ | Not reported. |
| Hypoglycaemia | 59 per 1,000 | 56 per 1,000 (9 to 358) | RR 0.95 (0.15 to 6.08) | 70 (2 RCTs) | ⊕⊝⊝⊝ VERY LOW 1 4 | ‐ |
| Discontinuation of medication due to adverse events | 0 per 1,000 | 0 per 1,000 (0 to 0) | not estimable | 63 (1 RCT) | ⊕⊝⊝⊝ VERY LOW 4 5 | ‐ |
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; RR: Risk ratio; MD: mean difference; HbA1c: haemoglobin A1c (glycated); FBG: fasting blood glucose; eGFR: estimated glomerular filtration rate; HD: haemodialysis; CKD: chronic kidney disease | ||||||
| GRADE Working Group grades of evidence High quality: We are very confident that the true effect lies close to that of the estimate of the effect Moderate quality: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low quality: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low quality: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect | ||||||
1 Risk of attrition and funding bias
2 Substantial heterogeneity
3 CI is wide and effect shows appreciable benefit and harm
4 Only 1 study had data for this outcome
5 Risk of selection, performance and detection bias
6 Narrative/qualitative synthesis was conducted. Estimates were not precise