The QT interval on a 12-lead echocardiogram (ECG) represents the time taken by ventricular cardiomyocytes to depolarize and repolarize. A prolonged QT interval, and especially the T-wave onset to T-peak (1), is thought to result from alterations in sympathetic and parasympathetic activity as well as several other risk factors. Since QTc is a measure of ventricular depolarization and repolarization, having a longer than normal QTc interval risks inducing early afterdepolarizations, and possibly also re-entrant excitation and, torsade de pointes (2–4) ultimately leading to ventricular arrhythmias and ventricular fibrillation (5). Prolongation of the heart rate corrected QT (QTc) is a well-established risk factor for increased cardiovascular mortality (5, 6), all-cause mortality and morbidity (7), even in apparently healthy people (8). Although many of the factors associated with QTc prolongation have been identified, including female sex (9, 10), hypokalemia (11), left ventricular hypertrophy (12), hypertension (13), drug side effects (2) and genetics (14, 15), there are still several unknowns. There is now a need to incorporate metabolic biomarkers in our research to understand the pathophysiology of QTc prolongation.

IGF-I regulates somatic growth, reaching its highest levels during teenage years, with levels decreasing with age (16, 17) and it is involved in cell proliferation, protein synthesis, nutrient homeostasis and nervous system, liver, kidney and cardiac development (18, 19). The decline in IGF-I seems to be greater with higher fat mass (16). IGF-I is also influenced by other hormones, age, sex, diet and nutrition. Previous studies in older age cohorts showed that reduced levels of IGF-I increase the risk of ischemic heart disease and cardiovascular mortality (18, 20). Animal work has shown that insulin like growth factor-I (IGF-I) may influence cardiac repolarization, via the phosphatidyl inositol-3 kinase/protein kinase B (PI3-K/Akt) pathway in cardiomyocytes (21). The PI3-K pathway directly regulates most of the heart's ion channels, including the rapid delayed rectifier potassium channel that specifically influences cardiac repolarization (22) and in some animal models, consequently the QTc duration (23). PI3-K was shown to affect many of the channels involved in the action potential duration, late sodium current, calcium current and slow delayed rectifier potassium channels, through its downstream signaling (22). Yet, little is known about the relationship between IGF-I and QTc duration in humans, particularly in older persons, in whom IGF-I levels are known to decline (24). The insulin growth factor-family (IGFs), including IGF-I and IGF-II, has a wide range of physiological functions including the regulation of cellular proliferation, apoptosis, protein synthesis and metabolism (25). IGF-I in the circulation is bound to insulin like growth factor binding protein-3 (IGFBP-3) and therefore the molar ratio between IGF-I and IGFBP-3 indicates IGF-I bioavailability (26).

We sought to investigate the association between circulating blood levels of IGFs and IGFBP-3 with cardiac repolarization represented by the QTc interval in older age participants of a population-based longitudinal cohort.

In a UK-based sample of older persons aged 60–64 years, declining serum levels and bioavailability of IGF-I (previously known as Somatomedin C) associated with QTcprolongation, a well-established risk factor for sudden death, independent of sex, SEP, BMI, LV mass, heart disease and hypertension.

In several studies it has been shown that reduced levels of IGF-I were associated with increased risk of cardiovascular disease (44, 45). IGF-I is thought to protect cardiac myocytes from arrhythmogenesis and apoptosis by activating the PI3-K/Akt cell survival intracellular signaling (21), although the exact mechanism remains to be fully elucidated, with a possibility that channel transcription might be affected. The PI3-K pathway activates the serine/threonine protein kinase Akt, and Protein Kinase C, promoting cardiovascular homeostasis, neuroprotection, survival, gene expression, and insulin activity (19, 46). The rapid delayed rectifier potassium channels that influence cardiac repolarization in cardiomyocytes are regulated by PI3-K (21, 22, 47). These channels (I_Ks, I_Kr and the atrial specific I_Kur) conduct outward potassium currents during the plateau phase of the action potential (48). Mutations in the genes encoding delayed rectifiers disrupt normal cardiac repolarization and lead to various cardiac rhythm disorders, including congenital long QT syndrome. Late sodium currents, I_NaL, were shown to be activated by another a downward pathway of PI3-K, the serum- and glucocorticoid-regulated kinase (SGK), which phosphorylates the sodium channels on the cardiomyocyte surface. GSK was shown to phosphorylates neural precursor cell expressed developmentally down-regulated protein 4 (NEED4), blocking the ubiquitination of the sodium channel (49, 50). These two channels, I_Kr and I_Na, were thought to finely interact to maintain a correct ventricular repolarization (51). Reduced IGF-I binding to its receptors on cardiomyocytes, has been shown to decrease the activation of the PI3-K/Akt pathway in animal models (22) thus prolonging the action potential duration and also the QTc. One factor that is thought to mediate the electrophysiological effect of IGF-I is nuclear factor erythroid 2-related factor 2 (Nrf2), which normally induces the transcription of cytoprotective enzymes involved in antioxidative pathways (52, 53) that have the ability to suppress ventricular arrhythmias. Evidence suggests that IGF-I regulates the levels of Nrf2 expression and modulates its transcriptional activity via the PI3K/Akt pathway (54). Based on the results from our current study, it is plausible therefore, that in older persons, declining levels and bioavailability of IGF-I reduce expression of Nrf2, prolonging QTc, and increasing the propensity for malignant arrhythmias.

IGF-I and IGF-II are mainly produced by the liver under the influence of growth hormone and nutrition (19, 55, 56). They have a similar structure to insulin with a similar direct effect on the body's glucose metabolism (57, 58), but are found in much higher concentrations in the blood where they are bound to binding proteins (mainly to isoform IGFBP-3), also produced by the liver. Binding to IGFBP-3 increases its half-life and modulates receptor binding. IGF-I fulfills several important functions in the human body by reaching several targets, where both insulin and IGF-1 receptors are found. An imbalance in serum IGF-I levels has been associated with a variety of negative effects, in several body systems, including obesity, diabetes and atherosclerosis. Reduced levels of IGF-I increase risk of hypertension, inflammation and endothelial dysfunction as normal IGF-I levels were shown to be protective, stimulating the release of nitric oxide, a vasodilator, and promoting cell proliferation and differentiation (46, 59). Quantitatively, IGF-II is the predominant circulating IGF, present in adults at a concentration up to three times that of IGF-I (as noted in our cohort). In mammals, IGF-I mediates the growth promoting effects of growth hormone during postnatal life and throughout adulthood. It influences cardiomyocytes and the cardiac action potential as described above, but IGF-II is more involved in placental and fetal growth (including cardiac development) in utero (60), is less growth hormone dependent than IGF-I (61), and is not known to influence the cardiac action potential in mature cardiomyocytes. Indeed, we found no statistically significant association between circulating levels of IGF-II and QTc interval at either time point. As IGF-I fulfills important somatic growth function it reaches its highest levels during teenage years, with levels subsequently decreasing with age (16, 17) in a highly variable and individual process that is related to fat mass (16), sex, diet and hormonal status. Among its various functions, IGF-I protects against inflammation, hypertension, endothelial and β-cell dysfunction (46, 59, 62), inhibits growth hormone hypersecretion and suppresses insulin secretion whilst enhancing insulin's action (63). An imbalance in serum IGF-I levels has been associated with obesity, diabetes and atherosclerosis (46, 59).

The six IGF-binding proteins (IGFBP-1-6) have both IGF-dependent and independent functions (64, 65). IGFBP-3 is the most abundant of the six IGFBPs in the circulation. Low IGFBP-3 levels have been associated with adverse cardiovascular effects, increased vascular disease and higher risk of coronary events (66). IGFBPs serve not only to transport IGFs in the circulation but also to prolong their half-lives, modulate their tissue specificity, and to either potentiate or neutralize their biological actions at tissue level (65). Measuring free (unbound) IGF-I remains a challenge (52) and total measured serum IGF-I is not tantamount to bioavailable IGF-I (33): almost all circulating IGF-I is bound to IGFBPs leaving <1% of IGF-I in a free form bioavailable for receptor-binding (47). The molar ratio of total IGF-I to IGFBP-3 is widely used as the proxy for bioavailable IGF-I (67).

Our study examined whether the differential decline in IGFs, IGFBP-3 or IGF-I/IGFBP-3 ratio over a decade (53 to 60–64 years) put some older persons at higher risk of QTc prolongation. Results showed that adults who experienced a steeper decline in IGF-I over a decade of later life were at higher risk of QTc prolongation. With age, levels of IGFBP-3 decline more steeply than IGF-I which is why the IGF-I/IGFBP-3 ratio (and therefore IGF-I bioavailability) appears to increase (32) (Figure 2). We show that older persons whose molar ratios increased least over a decade—implying less bioavailable IGF-I overall—had prolonged QTc compared to those who had higher IGF-I bioavailability. This aligns with other recently published data showing how low levels of IGFBP-3 increased the risk of cardiovascular disease and mortality (18) and how low IGF-I/IGFBP-3 ratios increased the risk of metabolic syndrome and insulin resistance (68). The fact that after removing persons with cardiovascular disease in the sensitivity analysis, only the 60–64y IGF-I/IGFBP-3 molar ratio and ΔIGF-I/IGFBP-3 ratio retained significant association with QTc, adds credence to the notion that it is the free, bioavailable IGF-I which most strongly determines the electrophysiological effect observed. Therefore, the IGF-I/IGFBP-3 ratio has to maintained at safe levels to avoid adverse metabolic, cardiovascular and neoplastic effects (69).

Our finding, that low levels of IGF-I relate to QTc prolongation and therefore higher sudden cardiac death risk, fits with several other known adverse cardiovascular effects of IGF-I deficiency, that include accelerated cardiovascular aging, reduced cardiac contractility, hypertrophy, hypertension, coronary disease and even atrial fibrillation (19, 53, 70). Conversely, normal IGF-I levels appear to be cardioprotective, by stimulating the release of vasodilatory nitric oxide, and by promoting cell proliferation and differentiation (46, 59). In mice IGF-I receptor deficiency has been associated with cardiomyopathy and heart failure (53).

Previous works that failed to account for confounders, have observed an association between diabetes or insulin resistance with QTc prolongation (71–75), but we found no association between fasting blood glucose levels or HbA1c with QTc at age 60-64. The mechanism of QTc prolongation in insulin resistance is still unclear but it is known that insulin may increase the transmembrane potential of cardiomyocytes by activating the electrogenic Na⁺/K⁺-ATPase leading to hyperpolarization and therefore QT prolongation (76).

We found an association between low blood potassium levels and QTc prolongation, as anticipated. Important cardiomyocyte-specific mechanisms are the main determinants of hypokalemia-induced QT prolongation (77) in addition to altering autonomic nervous system activity (78, 79). Potassium deficiency, as shown experimentally in a legacy study on rats (80), also reduces IGF-I, so it potentially confounds the relationship between IGF-I and QTc. Several previous studies (81–83) explored an association between increased LV mass and prolonged QTc, supporting our evidence, that LV mass is also an important determinant of QTc.

In a large older-age population-based cohort, declining levels and bioavailability of IGF-I associate with prolonged QTc interval. As QTc prolongation is known to be associated with increased risk for sudden death even in apparently healthy people, further work is needed to understand and preserve the potentially anti-arrhythmic effects of IGF-I in older age.

NSHD data is available through the Medical Research Council Skylark website (https://skylark.ucl.ac.uk/Skylark) and full details on the archived data is available at: https://www.nshd.mrc.ac.uk/data.

The studies involving human participants were reviewed and approved by Greater Manchester Local Research Ethics Committee and the Scotland Research Ethics Committee. The patients participants provided their written informed consent to participate in this study.

CC and GC conceived of the study. CC wrote the manuscript and analyzed the data. JM, JH, NC, and AH provided expert review of the manuscript. All authors contributed to the article and approved the submitted version.

GC and JM are supported by the Barts Charity HeartOME grant (MGU0427). GC was supported by British Heart Foundation (MyoFit46 Special Programme Grant SP/20/2/34841) and by the NIHR UCL Hospitals Biomedical Research Center. The NSHD cohort was funded by the UK MRC (program codes MC_UU_12019/1; MC_UU_12019/4; MC_UU_12019/5). JM is directly and indirectly supported by the UCL Hospitals NIHR BRC and Biomedical Research Unit at Barts Hospital, respectively. AH receives support from the British Heart Foundation, the Economic and Social Research Council (ESRC), the Horizon 2020 Framework Programme of the European Union, the National Institute on Aging, the National Institute for Health Research University College London Hospitals Biomedical Research Center and the UK MRC.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.