فاکتور 1 رشد سرم شبه انسولین و افسردگی در اواخر عمر: یک مطالعه مبتنی بر جمعیت

Objective Serum insulin-like growth factor 1 (IGF-1) concentration decreases, while the prevalence of depressive symptoms increases with advancing age. Although basic research indicates a link between low IGF-1 concentration and depression, this has scarcely been investigated in humans. This study investigates whether lower IGF-1 concentrations are associated with prevalent and incident late-life depression over a 3-year period. Methods The study included 1188 participants, aged ≥65 years, from the Longitudinal Aging Study Amsterdam (LASA), an ongoing, population-based cohort study. Depression was assessed at baseline and after three years using the Center for Epidemiological Studies-Depression Scale (CES-D) and the Diagnostic Interview Schedule (DIS), and categorized into minor depression and major depression (MDD). Serum IGF-1 concentration was determined at baseline. Associations were adjusted for relevant confounders. Results Serum IGF-1 concentrations were within the normal range (mean 13.9 nmol/l, standard deviation 5.3 nmol/l). At baseline, in men, as compared to high concentrations, mid concentrations decreased the probability of prevalent minor depression (odds ratio [OR] = 0.35, 95% confidence interval [CI] = 0.15–0.82). In women, as compared to high concentrations, low concentrations tended to increase the probability of prevalent MDD (OR = 2.66, 95% CI = 0.89–7.89). At three-year follow-up, in men, no significant prospective associations were detected. In women, as compared to high concentrations, mid concentrations decreased the probability of incident minor depression (OR = 0.43, 95% CI = 0.19–0.95). Conclusions Several associations, which differed across the genders, were observed between IGF-1 and depression. Cross-sectional findings were not supported by longitudinal findings, which suggest that IGF-1 may not play an important predictive role in the development of depression in older persons over time. However, a more acute role of IGF-1 in current depression, as indicated by the cross-sectional results, may be possible. Further studies are needed to elucidate the complex relation between IGF-1 and late-life depression.

Depression is a common and burdensome disorder in older individuals (Beekman et al., 1999 and Copeland et al., 2004). While the prevalence of major depression (MDD) is relatively low (1.8%), the prevalence of all clinically relevant depressive syndromes, some of which do not fully fulfill criteria for MDD, is high (13.5%) in community dwelling elderly (Beekman et al., 1999). Both may have detrimental consequences for the wellbeing and functioning of older persons, and are associated with increased mortality (Alexopoulos, 2005, Cuijpers et al., 2013 and Scafato et al., 2012). To date, the exact etiology and pathophysiology of depression are still unclear, but are presumed to be multifactorial (Alexopoulos, 2005). It is hypothesized that biological factors, such as insulin-like growth factor 1 (IGF-1) may play a role (Duman, 2004). In healthy adults, the secretion of growth hormone (GH) and IGF-1 decreases with advancing age, a process also known as the ‘somatopause’ (Bartke, 2008 and Corpas et al., 1993). Decreased functioning of the GH-IGF-1 axis is also found in adult patients with growth hormone deficiency (GHD). Adult GHD is, amongst others, associated with decreased cognitive functioning, impaired quality of life (QoL) and psychiatric symptoms (de Boer et al., 1995, Giusti et al., 1998 and Mahajan et al., 2004). GH supplementation has been shown to improve these symptoms (Kokshoorn et al., 2011 and Mahajan et al., 2004). Since features of normal aging resemble those of GHD, it has been suggested that the ‘somatopause’ contributes to age-associated changes and disorders (Bartke, 2008, Corpas et al., 1993, Mahajan et al., 2004 and Mitschelen et al., 2011). For instance, lower IGF-1 concentrations are related to impaired cognitive functioning, which frequently coincides with late-life depression, in healthy elderly (Arwert et al., 2005). In addition, IGF-1 receptors have been localized within many brain structures that are associated with mood such as the hippocampus and the amygdala (Adem et al., 1989). Also, there is evidence that IGF-1 has direct effects on the central nervous system (CNS) through blood-brain barrier passage and local autocrine and paracrine mechanisms (Schneider et al., 2003). Moreover, basic research has shown that the expression of neurotrophic and growth factors, such as IGF-1, is decreased by chronic stress and depression in brain regions associated with depression, and increased by antidepressant treatment (Duman, 2004 and Mitschelen et al., 2011). Recently, an animal study demonstrated that adult-onset long-term IGF-1 deficiency is able to induce depressive behavior in normally developed mice (Mitschelen et al., 2011). Nevertheless, only few studies with limited sample sizes and no or short follow-up time have addressed the relation between IGF-1 and depression in humans (Deuschle et al., 1997, Franz et al., 1999, Lesch et al., 1988a, Lesch et al., 1988b, Rupprecht et al., 1989, Schilling et al., 2011 and Weber-Hamann et al., 2009). So far, only one epidemiological study has been reported, in which lower IGF-1 concentrations were associated with any depressive disorder in women, while higher IGF-1 concentrations were associated with any depressive disorder in men (Sievers et al., 2014). However, in this study only a screening self-report questionnaire was used for the assessment of any depressive disorder, which complicates interpretation of the results. Moreover, the relation between IGF-1 and depression has not yet been specifically studied in older individuals. Therefore, we aimed to investigate the association between IGF-1 and depression cross-sectionally as well as prospectively in a large, representative cohort of community dwelling elderly, using both a depression symptoms rating scale and a diagnostic psychiatric evaluation. We hypothesized that persons with lower IGF-1 concentrations would be more likely to have current minor depression or MDD and would be more likely to develop minor depression or MDD in the future than persons with higher IGF-1 concentrations.

The study cohort included 1188 participants, mean age 75.4 (6.5) years, of whom 161 (13.6%) had minor depression and 32 (2.7%) had MDD at baseline. The mean serum IGF-1 concentration was 13.9 (5.3) nmol/l. Table 1 describes the baseline characteristics according to tertiles of IGF-1 concentration. Table 1. Baseline characteristics of the study population according to tertiles of IGF-1 concentration. Tertile I ≤11.5 nmol/l Tertile II >11.5 to ≤15.5 nmol/l Tertile III >15.5 nmol/l p-Value f No. of subjects 393 393 402 IGF-1 (nmol/l)a 8.6 (2.1) 13.5 (1.1) 19.5 (4.1) <0.001 Depression statusc 0.02 no depression 314 (79.9) 338 (86.0) 343 (85.3) minor depression 60 (15.3) 49 (12.5) 52 (12.9) major depression 19 (4.8) 6 (1.5) 7 (1.7) Age at baseline (yr)a 77.5 (6.5) 75.5 (6.6) 73.4 (5.8) <0.001 Female sexc 233 (59.3) 188 (47.8) 177 (44.0) <0.001 Smokingc 0.001 never 172 (43.8) 131 (33.3) 119 (29.6) former 159 (40.5) 191 (48.6) 212 (52.7) current 62 (15.8) 71 (18.1) 71 (17.7) Alcohol consumptionc 0.22 none 103 (26.2) 94 (23.9) 92 (22.9) light 201 (51.1) 183 (46.6) 203 (50.5) moderate 71 (18.1) 92 (23.4) 76 (18.9) (very) excessive 18 (4.6) 24 (6.1) 31 (7.7) BMI (kg/m2)a 26.8 (4.5) 26.8 (4.1) 27.0 (3.7) 0.69 Physical activity (min/day)b 137.3 (0–624) 129.3 (0–625) 128.9 (0–600) 0.12 Albumin (g/l)a 44.1 (2.8) 44.5 (2.8) 44.8 (2.5) <0.001 No. of chronic diseasesb 1 (0–5) 1 (0–5) 1 (0–5) 0.23 MMSE score and 28 (12–30) 28 (12–30) 28 (10–30) 0.01 Education and 0.54 low 248 (63.1) 230 (58.7) 239 (59.5) middle 99 (25.2) 110 (28.1) 119 (29.6) high 46 (11.7) 52 (13.3) 44 (10.9) Systemic corticosteroids usec 7 (1.8) 8 (2.0) 7 (1.7) 0.95 Antidepressant usec 12 (3.1) 6 (1.5) 9 (2.2) 0.36 Benzodiazepine usec 37 (9.4) 34 (8.7) 26 (6.5) 0.29 IGF-1, insulin-like growth factor 1; BMI, body mass index; MMSE, Mini-Mental State Examination. a Normally distributed continuous variables are expressed as mean (standard deviation). b Skewed continuous variables are expressed as median (range). c Categorical variables are presented as number (percentage). d Missing n = 3. e Missing n = 1. f Continuous variables were tested with either one-way analysis of variance (normally distributed variables) or the Kruskal–Wallis test (skewed variables). Categorical variables were examined with the χ2-test. Table options 3.2. Cross-sectional associations The Spearman correlation coefficient between baseline IGF-1 concentration and baseline continuous CES-D score, which measures depressive symptoms, was −0.11 (p < 0.001). Multinomial logistic regression analysis was used to further examine the cross-sectional association of IGF-1 concentration with prevalent minor depression and MDD (Table 2). The analyses were stratified for gender, since a significant interaction between IGF-1 and gender was observed (p = 0.02). Table 2. Results of multinomial logistic regression analyses for the association of IGF-1 concentration with prevalent minor and major depression at baseline. Unadjusted model Model 1 Model 2 OR (95% CI) p-Value OR (95% CI) p-Value OR (95% CI) p-Value IGF-1 Minor depression Males First tertile 0.85 (0.42–1.7) 0.64 0.60 (0.29–1.27) 0.18 0.55 (0.25–1.22) 0.14 Second tertile 0.41 (0.18–0.90) 0.03 0.34 (0.15–0.77) 0.01 0.35 (0.15–0.82) 0.02 Third tertile Reference Reference Reference Continuous level 1.03 (0.83–1.28) 0.8 1.12 (0.89–1.40) 0.33 1.14 (0.91–1.45) 0.26 Females First tertile 1.33 (0.80–2.23) 0.27 1.27 (0.75–2.13) 0.38 1.38 (0.80–2.39) 0.25 Second tertile 1.35 (0.80–2.30) 0.27 1.31 (0.77–2.23) 0.32 1.13 (0.77–2.33) 0.3 Third tertile Reference Reference Reference Continuous level 0.92 (0.79–1.06) 0.24 0.93 (0.80–1.08) 0.33 0.91 (0.78–1.06) 0.21 Major depression Males First tertile 1.39 (0.19–9.98) 0.71 0.99 (0.13–7.41) 0.99 0.82 (0.10–6.62) 0.85 Second tertile 2.08 (0.38–11.51) 0.4 1.74 (0.31–9.80) 0.53 1.57 (0.27–9.31) 0.62 Third tertile Reference Reference Reference Continuous level 0.86 (0.52–1.41) 0.54 0.94 (0.56–1.58) 0.82 1.01 (0.59–1.72) 0.98 Females First tertile 2.86 (1.03–7.94) 0.04 2.53 (0.90–7.12) 0.08 2.66 (0.89–7.89) 0.08 Second tertile 0.39 (0.075–2.05) 0.27 0.37 (0.07–1.97) 0.25 0.36 (0.07–1.95) 0.23 Third tertile Reference Reference Reference Continuous level 0.70 (0.51–0.96) 0.03 0.73 (0.53–0.01) 0.05 0.72 (0.52–1.02) 0.06 IGF-1, insulin-like growth factor 1; OR, odds ratio; CI, confidence interval. Distribution of IGF-1 concentration per tertile was as follows: first tertile ≤11.5 nmol/l, second tertile >11.5 to ≤15.5 nmol/l, third tertile >15.5 nmol/l. Referent depression outcome group was no depression. Analyses were stratified for gender (lowest p-value interaction terms for gender 0.02). Model 1 was adjusted for age. Model 2 was fully adjusted for age, alcohol consumption, albumin, smoking status, number of chronic diseases, BMI and physical activity. Table options In men, both the unadjusted and the adjusted models (age, alcohol consumption, albumin, smoking status, number of chronic diseases, BMI and physical activity) demonstrated significantly lower odds for minor depression for IGF-1 concentrations in the middle tertile compared to IGF-1 concentrations in the highest tertile. In women, the odds for MDD were higher for IGF-1 concentrations in the lowest tertile compared to IGF-1 concentrations in the highest tertile in the unadjusted model. However, after adjustment for relevant confounders this association did not remain statistically significant. No other statistically significant associations were observed. Separate exclusion of specific subgroups (users of anti-depressants, systemic corticosteroids, benzodiazepines and subjects with a decreased renal function) from the analyses, did not substantially influence the results.