Statins and new‐onset diabetes mellitus – a risk lacking in familial hypercholesterolaemia

Statins are the cornerstone of treatment to safely and effectively lower serum low-density lipoprotein cholesterol (LDL-C) levels, and they reduce atherosclerotic cardiovascular disease events and mortality 1. Statins are effective in all relevant subgroups of patients, including those with diabetes 2. Therefore, it may seem paradoxical that statins have recently been shown to increase the risk of new-onset diabetes (NOD), a condition commonly associated with atherosclerotic cardiovascular disease. Is this a real clinical concern and what is the underlying mechanism? Although it has previously been reported that statins may affect glucose metabolism, for example by Jula et al. in 2002 3, the relationship between statin use and NOD only became well recognized after the analysis of 17 802 apparently healthy men and women participating in the placebo-controlled Justification for the Use of Statins in Primary Prevention: An Intervention Trial Evaluating Rosuvastatin (JUPITER) trial, in which a minimal, but statistically significant, increase in glycated haemoglobin values amongst the statin-treated patients was observed 4. Soon after the results of the JUPITER trial had been reported, a meta-analysis of 13 statin trials with 91 140 participants revealed that statin therapy was associated with a 9% increase in the likelihood of NOD during follow-up [odds ratio 1.09; 95% confidence interval (CI) 1.02–1.17] and that it took an average of 4 years to develop NOD after having started statin treatment 5. An exception was the West of Scotland Coronary Prevention Study (WOSCOPS) which showed that pravastatin decreased the frequency of NOD 6. A subsequent meta-analysis of five statin trials with 32 752 participants also showed that intensive-dose statin therapy was associated with an increased risk of NOD 7. Furthermore, increased NOD risk due to statin treatment has been shown in several observational studies. Despite the association between statins and NOD, the results of all studies investigating this association have confirmed the overriding cardiovascular therapeutic benefit of statin treatment. Both population-based studies and meta-analyses have revealed interesting differences in the effects of different statins on NOD risk/incidence. The relation between the use of statins and the incidence of NOD amongst patients 66 years of age or older was analysed in a Canadian population-based cohort study 8. A total of 471 250 patients were included, 227 994 of whom were receiving statin therapy for primary prevention of atherosclerotic cardiovascular disease. The analysis was limited to patients who had started statin treatment during the preceding year or earlier and excluded those with diabetes. A significant limitation of this study was that important risk factors for NOD, including weight, ethnicity and family history, were not identified. It was shown that, when compared with pravastatin, the use of atorvastatin, rosuvastatin or simvastatin was associated with an increased risk of NOD, with adjusted hazard ratios of 1.22 (95% CI 1.15–1.29), 1.18 (95% CI 1.10–1.26) and 1.10 (95% CI 1.04–1.17), respectively. However, compared to patients treated with pravastatin, the risk of NOD was not increased amongst those receiving fluvastatin or lovastatin. These findings applied to patients who received statins for either primary or secondary prevention of atherosclerotic cardiovascular disease. Another recent comprehensive meta-analysis of randomized controlled trials involving 113 394 persons has also revealed an increased risk of NOD amongst statin users 9. In this analysis, different types and doses of statins were compared with placebo. Pravastatin at 40 mg day−1 was associated with the lowest (7%), atorvastatin 80 mg day−1 with an intermediate (15%) and rosuvastatin 20 mg day−1 with the highest (25%) risk of NOD. The results with simvastatin 40 mg day−1 (21% increased risk) were comparable with those for rosuvastatin 20 mg day−1. Whilst the effects of metabolically different statins (i.e. the two subclasses hydrophilic and hydrophobic statins; see below) and the risk of NOD were apparent in the above-mentioned cohort study and meta-analysis 8, 9, they have not been observed in any other similar analyses. An important example of an analysis in which statin treatment was associated with higher risk of NOD, but a subclass effect of statins could not be detected, is the meta-analysis by Naci et al. 10 of 246 955 patients from 135 randomized and controlled studies. Type 2 diabetes (T2D) is a complex and multifaceted disease which develops when insulin secretion is not able to compensate for insulin resistance 11. Statins may directly or indirectly decrease insulin synthesis or disturb insulin secretion 12, and they may also impair insulin sensitivity of the target cells, thereby causing insulin resistance. Furthermore, the insulin resistance-induced effect of statins may be indirect, as statins have been shown to predispose towards increased weight. As stated above, the statins can be divided into lipophilic and hydrophilic ones. Pravastatin and rosuvastatin have predominantly hydrophilic properties, which means that their cellular uptake depends on carrier-mediated mechanisms, whilst atorvastatin, fluvastatin, lovastatin, pitavastatin and simvastatin belong to the lipophilic subclass of statins. The lipophilic statins are less hepatoselective than the hydrophilic statins 13, an interesting exception being rosuvastatin, which is hydrophilic but acts in the same manner as a lipophilic statin. The difference between hydrophilic and lipophilic statins may be important when investigating differences in insulin sensitivity between the two subclasses of these drugs. Thus, using isolated rat islet ß-cells, it could be shown that lipophilic statins can inhibit glucose-induced signalling 14. Of note, in the cited study, any clinically relevant concentrations of pravastatin failed to have an inhibitory effect on glucose-induced signalling. Pravastatin was compared to atorvastatin in a very small, Japanese, open-label, cross-over study of patients with early-stage T2D 15. It was found that pravastatin had favourable effects on pancreatic ß-cell function but, as the authors pointed out, an effect on ß-cell function in the clinical setting was very modest, if present at all. In a very recent follow-up study, 8749 nondiabetic subjects aged 45–73 years were followed for an average of 5.9 years 16. Statin treatment increased the risk of T2DM by 46%; this large increase could be attributed to sensitive detection of dysglycaemia (including use of the oral glucose tolerance test), that is the presence of a decrease in insulin secretion and/or sensitivity could be determined. In addition to the probable effects on insulin secretion and sensitivity, several other possible mechanisms underlying an association between statins and NOD have been suggested. It has been proposed that these mechanisms are related to insulin signal transduction 17, adipocyte maturation 18 or a secondary decrease in various downstream products of mevalonate along the cholesterol synthesis pathway and its branches, such as dolichol, farnesyl pyrophosphate, geranylgeranyl pyrophosphate and coenzyme Q10, in pancreatic islet cells 19. In heterozygous familial hypercholesterolaemia (FH), statin treatment is initiated in childhood 20. This recommendation is based on coronary angiographic studies in middle-aged FH patients, which have allowed to calculate roughly that in untreated individuals with heterozygous FH, coronary stenosis starts to develop in male patients as young as 17 years of age and in female patients as young as 25 years of age 21. Additionally, increased carotid intima–media thickness is detectable in untreated children with FH as early as the second decade of life 22-24. Clearly, however, epidemiological studies have shown that FH patients, if treated effectively during adult life, have a life expectancy similar to that of the general population 25, 26. Recent follow-up studies have suggested that long-term statin therapy is not associated with a risk of NOD in patients with FH 27, 28. A total of 194 children with FH (age at study initiation 8–18 years) and their 83 non-FH siblings were followed for 10 years by Kusters et al. 27. All 194 FH children used statins throughout the follow-up period, and there was no difference in diabetes incidence between the FH and non-FH groups. Skoumas et al. 28 followed 212 adult FH patients (mean age 44 years) for 10 years and found that statin treatment was not associated with an increased risk of NOD in these patients. In an earlier study of 51 homozygous and 20 heterozygous FH patients, high-dose statin therapy was not associated with impaired insulin resistance 29. Thus, the results regarding FH patients are positively contrasting those in the above-mentioned recent follow-up of hypercholesterolaemic patients without FH 16. Taken together, the evidence to date suggests that the risk of developing NOD is not increased in statin-treated children or adults with FH. Because statin treatment is not related to insulin resistance in FH 29, it is possible that the apparent protective effect of FH against developing statin-associated NOD is rather related to insulin secretion by the pancreatic beta cells. Interestingly, in a very recent study by Besseling et al. 30 from the Netherlands, it was shown that the prevalence of T2D amongst FH patients was lower than amongst their unaffected relatives (relative difference of 50%). Although the possible cellular mechanisms underlying this difference remained unclear, the authors speculated that the common pathway in FH and statin therapy relates to altered intracellular cholesterol metabolism in pancreatic beta cells. In this state of uncertainty, we need to consider, besides the endothelial cells and pancreatic beta cell, also the hepatocyte as a potential cellular site in which the genetic lack of half of LDL receptors somehow confers protection against statin-induced metabolic changes contributing to the development of NOD. In this regard, there is a growing understanding of how sterol regulatory element-binding protein-1 (SREBP-1) takes part in the regulation of free fatty acid metabolism in the liver, and how insulin is involved in this complex regulatory process 31. Interestingly, gene expression studies have shown that fibroblast SREBP-1 expression appears to be higher in FH patients compared to control subjects 32. However, because T2D is a complex disease, several yet unknown molecular mechanisms may explain the apparent resistance of FH patients to statin-induced NOD. Accordingly, we have to admit that there is insufficient evidence in support of any specific mechanism that would explain the lack of elevated risk of NOD in FH patients, and only hope that future studies will be able to clarify why FH patients are resistant to NOD. Ideally, such understanding could lead to better ways to combat the diabetes-inducing ability of statins in hypercholesterolaemic non-FH patient populations. Perhaps the biggest concern with regard to statin-related NOD is that it may lead to discontinuation of a highly beneficial statin therapy. Such discontinuation would be particularly problematic in patients with FH, in whom the benefits of statin therapy are much greater than in the general population, and, as discussed above, the risk of NOD is not increased. Explaining the unexpected apparent protective effect of LDL receptor mutations against T2D is challenging, and, once mechanistically understood, may also clarify the intriguing link between statin treatment and plasma glucose elevation in the large sector of the population using these potent drugs. Alpo Vuorio has received lecture honoraria from Aegerion Ltd. Timo Strandberg has received consultancy/lecture fees related to cholesterol treatment from Amgen, AstraZeneca, Merck, Orion Pharma and Pfizer; he owns a minor amount of stock in Orion Pharma (listed company). He is currently in the committee creating and updating the Finnish National Guidelines for Dyslipidemia (nonprofit position). Wolfgang Schneider. None. Petri Kovanen has received consultancy fees from Amgen and Aegerion Ltd. Petri Kovanen has received payment for lectures from Raisio and Unilever. Petri Kovanen holds stock/stock options with the Orion Pharma (listed company). Petri Kovanen is a member of the committee creating and updating the Finnish National Guidelines for Dyslipidemia (nonprofit position).

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.