Depressive and Anxiety Disorders Showing Robust, but Non-Dynamic, 6-Year Longitudinal Association With Short Leukocyte Telomere Length
Publication: American Journal of Psychiatry
Abstract
Objective:
Several cross-sectional studies have related depressive and anxiety disorders to shorter leukocyte telomere length (LTL) as an indicator of cellular aging. However, these studies have left many unresolved questions about underlying causality and ordering of associations. The objective of the present large, longitudinal study was to examine the relationship between depressive and anxiety disorders and LTL over a 6-year time period.
Method:
Data are from the Netherlands Study of Depression and Anxiety, including 2,292 patients with remitted and current diagnoses of depressive or anxiety disorders and 644 healthy control subjects. LTL was assessed using quantitative PCR and measured at baseline and after 6 years; depressive and anxiety disorder diagnoses and characteristics (course, duration, and severity) were determined at baseline and after 2, 4, and 6 years.
Results:
Results showed that persons with remitted (B=−52.6) and current (B=−60.8) depressive or anxiety disorder had consistently shorter LTL compared with healthy control subjects across baseline and at the 6-year follow-up, remaining significant when controlling for lifestyle and somatic health variables. Changes in the course of depressive or anxiety disorder characteristics over 6 years, however, were not associated with different LTL attrition rates.
Conclusions:
This study confirmed robust associations of depressive and anxiety disorders with shorter telomeres, but interestingly, it did not demonstrate that depressive and anxiety disorders and LTL change together over time, suggesting the absence of a direct within-person relationship. Short LTL is suggested to be either a long-term consequence or an underlying vulnerability factor for depressive or anxiety disorders.
Depressive and anxiety disorders are associated with shorter leukocyte telomere length (LTL), an indicator of cellular aging (1). This might help explain why depressed or anxious persons have increased onset risks of several aging-related somatic conditions, such as cardiovascular disease, diabetes type II, and obesity (2, 3). One of the proposed pathways is that various biological abnormalities in depressive and anxiety disorders, such as a dysregulated immune system or increased oxidative stress (4, 5), may lead to cellular damage, including shortened telomere length. Another less-explored explanation is that short telomere length might be a risk factor for persons to develop a depressive or anxiety disorder. Alternatively, an underlying third factor, such as genotype or lifestyle, may make persons vulnerable for both short telomeres and depressive or anxiety disorders (6). As human life expectancy increases, research into aforementioned pathways becomes increasingly important. A better understanding of biological mechanisms may yield novel interventions that prevent aging to be inevitably accompanied by aging-related somatic diseases and disability.
LTL has emerged as an indicator of biological or cellular, rather than chronological, age, since 1) it decreases progressively with every cell division (unless counteracted by sufficient activity of the ribonucleo-protein enzyme, telomerase) and thus with age (7); 2) it reflects damage accumulated over the years by lifestyle, psychological stress, and cytotoxic environments (8); and 3) it correlates and predicts the incidence of numerous serious medical illnesses (9). Several cross-sectional studies, with a few exceptions (10–12), found shorter LTL in depressed patients compared with control subjects (13–16), including investigations in the present sample (17). This was further confirmed by a recent meta-analysis that found significant effect sizes for the association between depression and shorter LTL (1). Also, although less often studied, shorter LTL was found in anxiety disorder patients compared with control subjects (18, 19). A major shortcoming of most studies published until now is their cross-sectional nature, limiting the ability to draw any inferences regarding cause and effect. Only two studies employed longitudinal designs: Hoen et al. (13) found that depression status in 608 heart disease patients did not predict 5-year change in LTL, while Shalev et al. (20) found that 193 men with a depression or anxiety disorder between the ages of 26 and 38 showed greater LTL decrease in that time frame compared with 226 healthy men, but this association was not present in women. In summary, it remains unclear to what extent possible telomere shortening is reversible when persons recover, whether chronicity of a disorder leads to accelerated shortening of LTL over time, whether there are common (genetic) risk factors for both, or, alternatively, whether short LTL is a vulnerability factor that may antecede the development of depressive and anxiety disorders. Longitudinal data are needed to explore these questions.
The present study examines the 6-year longitudinal relationship between LTL and depressive and anxiety disorders using a large sample (N=2,936). Objectives were to examine whether presence of diagnoses and symptom severity were consistently associated with LTL over time and whether changes in depressive and anxiety disorder characteristics (course, duration, and severity) corresponded with changes in LTL.
Method
Study Sample
Data were from the Netherlands Study of Depression and Anxiety (NESDA), an ongoing longitudinal cohort study examining the course and consequences of depressive and anxiety disorders. Study sample and methods have been described in detail elsewhere (21). In short, the NESDA baseline sample consisted of 2,981 persons between 18 and 65 years old, including persons with a current or remitted depressive and/or anxiety disorder (74%) and healthy control subjects (26%). To represent various settings and stages of psychopathology, depressed and anxious participants were recruited at three different locations in the Netherlands in different settings: the community, primary care, and specialized mental health care settings. In order to maintain representativity, there were only two exclusion criteria: 1) insufficient command of the Dutch language and 2) a primary clinical diagnosis of bipolar disorder, obsessive-compulsive disorder, posttraumatic stress disorder (PTSD), severe substance use disorder, or a psychotic disorder. The study was approved by the ethical review board of participating centers, and all participants signed informed consent. Participants were assessed during a 4-hour clinic visit. Every 2 years after the baseline assessment, face-to-face follow-up assessments were conducted. Follow-up assessments had a response of 87.1% (N=2,596) at the 2-year follow-up, 80.6% (N=2,402) at the 4-year follow-up, and 75.7% (N=2,256) at the 6-year follow-up.
Measurements
Depressive and/or anxiety disorder status.
At each assessment, persons were classified as 1) a control subject, 2) having a remitted depressive or anxiety disorder, or 3) having a current diagnosis. Control subjects were defined as having no lifetime history of depressive or anxiety disorders at all as assessed by the DSM-IV Composite International Diagnostic Interview (CIDI), version 2.1. Persons in the remitted group had a lifetime history of depression or anxiety disorder but no diagnosis in the past 6 months as diagnosed with the CIDI, and current patients had CIDI-diagnosed depressive disorders (major depressive disorder, dysthymia) and/or an anxiety disorder (social phobia, panic disorder with or without agoraphobia, generalized anxiety disorder, or agoraphobia) in the past 6 months. At baseline, 45 participants were excluded because of missing LTL data, leaving 2,936 individuals (control subjects, N=644; remitted patients, N=620; current patients, N=1,672). At the 6-year follow-up, blood samples were available for 2,003 participants, of whom 120 had unreliable LTL measurement, leaving 1,883 persons (control subjects, N=440; remitted patients, N=915; current patients, N=528 [see Table 1]). Persons who did not participate at the 6-year follow-up were younger and less educated, had longer LTL at baseline and higher depression and anxiety severity scores, and had more often a lifetime depression or anxiety diagnosis (all p values <0.01).
| Characteristic | Baseline (N=2,936) | 6-Year Follow-Up (N=1,883) | ||
|---|---|---|---|---|
| Mean | SD | Mean | SD | |
| Demographic | ||||
| Age (years) | 41.8 | 13.1 | 48.6 | 12.9 |
| Years of education | 12.2 | 3.3 | 12.9 | 3.3 |
| N | % | N | % | |
| Female sex | 1,950 | 66.4 | 1,231 | 65.4 |
| Mean | SD | Mean | SD | |
| Lifestyle and health | ||||
| Physical activity (in 1,000 MET-minutes per week)a | 3.7 | 3.0 | 4.0 | 3.4 |
| Number of somatic diseases | 0.6 | 0.9 | 0.6 | 0.9 |
| N | % | N | % | |
| Body mass index | ||||
| Underweight | 64 | 2.2 | 30 | 1.6 |
| Normal | 1,489 | 50.7 | 847 | 45.0 |
| Overweight | 893 | 30.4 | 633 | 33.6 |
| Obese | 490 | 16.7 | 373 | 19.8 |
| Smoking status (%) | ||||
| Never | 825 | 28.1 | 555 | 29.5 |
| Former | 975 | 33.2 | 797 | 42.3 |
| Current | 1,136 | 38.7 | 531 | 28.2 |
| Alcohol Status (%) | ||||
| Nondrinker | 499 | 17.0 | 330 | 17.5 |
| Mild-moderate drinker | 2,064 | 70.3 | 1,367 | 72.6 |
| Heavy drinker | 373 | 12.7 | 186 | 9.9 |
| Mean | SD | Mean | SD | |
| Psychiatric | ||||
| Depressive symptoms (Inventory of Depressive Symptoms-Self Report) | 21.4 | 14.1 | 15.2 | 11.9 |
| Anxiety symptoms (Beck Anxiety Inventory) | 12.1 | 10.6 | 8.3 | 8.4 |
| Percent time with depressive symptoms | 19.1 | 26.9 | 12.8 | 22.2 |
| Percent time with anxiety symptoms | 25.5 | 32.8 | 16.3 | 24.9 |
| N | % | N | % | |
| Diagnosis status | ||||
| Control | 644 | 21.9 | 440 | 23.4 |
| Remitted | 620 | 21.1 | 915 | 48.6 |
| Current | 1,672 | 56.9 | 528 | 28.0 |
| Within current diagnosis | ||||
| Depressive disorder | 389 | 23.3 | 159 | 30.1 |
| Anxiety disorder | 535 | 32.0 | 190 | 36.0 |
| Comorbid disorders | 748 | 44.7 | 179 | 33.9 |
| Antidepressant use | ||||
| Tricyclic antidepressants | 79 | 2.7 | 56 | 3.0 |
| Selective serotonin reuptake inhibitor | 502 | 17.1 | 224 | 11.9 |
| Other antidepressants | 164 | 5.6 | 104 | 5.5 |
| Leukocyte telomere length | ||||
| Mean | SD | Mean | SD | |
| Base pairs | 5,467 | 617 | 5,387 | 433 |
a
MET-minutes=metabolic equivalent of number of calories spent per minute.
Psychiatric characteristics.
Severity of depressive symptoms in the past week was assessed in all subjects with the 30-item Inventory of Depressive Symptoms-Self Report (22). Anxiety severity was assessed with the 21-item Beck Anxiety Inventory (23). For both severity measures, a change score was calculated by subtracting the score at baseline from the score at the 6-year follow-up. Duration was determined as the percentage of time with depressive or anxiety symptoms during each of the 2-year intervals between the follow-up assessments, as assessed by the calendar method of the Life Chart interview (24). Time with symptoms between assessments was averaged to calculate the average percentage of time with symptoms over the 6-year period. As another measure of chronicity, the number of assessments (baseline, 2-year, 4-year, and 6-year follow-ups) with a CIDI-diagnosed depressive and/or anxiety disorder was summed, ranging from 0 to 4. Finally, to investigate the associations of the course of depressive or anxiety disorders with LTL attrition, we created the following six groups based on diagnosis status at baseline and at the 2-, 4-, and 6-year follow-ups (persons were allowed to have one missing): 1) control group, 2) new onset, 3) persistent remitted, 4) relapse, 5) remission, and 6) chronic (for details, see Table 2).
| Group | N | Baseline | 2-Year | 4-Year | 6-Year |
|---|---|---|---|---|---|
| Control | 368 | No lifetime diagnosis | No diagnosis | No diagnosis | No diagnosis |
| New onset | 76 | No lifetime diagnosis | At least one current diagnosis at one follow-up assessment | At least one current diagnosis at one follow-up assessment | At least one current diagnosis at one follow-up assessment |
| Persistent remitted | 199 | Remitted diagnosis | Remitted diagnosis | Remitted diagnosis | Remitted diagnosis |
| Relapse | 150 | Remitted diagnosis | At least one current diagnosis at one follow-up assessment | At least one current diagnosis at one follow-up assessment | At least one current diagnosis at one follow-up assessment |
| Remission | 465 | Current diagnosis | Remitted/current diagnosis | Remitted/current diagnosis | Remitted diagnosis |
| Chronic | 501 | Current diagnosis | Remitted/current diagnosis | Remitted/current diagnosis | Current diagnosis |
a
The total N is less than 1,860 because of missing data at the 2-year or 4-year assessments.
Leukocyte telomere length.
LTL was assessed at baseline and the 6-year follow-up. Fasting blood was drawn from participants in the morning and stored in a –20°C freezer. Baseline and 6-year LTL were determined at the laboratories of Telomere Diagnostics (Menlo Park, Calif.) and the University of California, San Francisco, in 2012 and 2014, respectively, using quantitative polymerase chain reaction (qPCR) adapted from the published original method by Cawthon (25). Telomere sequence copy number in each patient’s sample (T) was compared with a single-copy gene copy number (S), relative to a reference sample. The resulting T/S ratio is proportional to mean LTL. The detailed method is described elsewhere (17). As previously described, T/S ratios were converted into base pairs (bp) with the following formula: bp=3274+2413×([T/S-0.0545]/1.16).
The reliability of the assays was adequate: the included quality control DNA samples on each PCR run illustrated that the interassay coefficient of variation (CV) was sufficiently low (baseline: CV=4.6%; 6-year: CV=3.0%), as well as for the telomere (baseline: CV=2.0%; 6-year: CV=5.4%) and the single-gene assays (baseline: CV=1.6%; 6-year: CV=4.8%) separately. Variances caused by these CVs are negligible on a group level. The 6-year follow-up T/S ratios were adjusted relative to the baseline samples for systematic differences caused by different reference samples, by rerunning and comparing samples from baseline sample plates (N=226, up to eight samples from each of the baseline plates), together with 6-year follow-up samples. On average, the T/S ratios of the 6-year follow-up runs were at 76% of the T/S ratios of baseline; consequently, the follow-up T/S ratios were divided by 0.76. DNA samples were de-identified, and the laboratories that performed the assays were blind to all other measurements, and thus samples for case patients and control subjects were randomly distributed over the plates. Overall, 1,860 persons had complete LTL at both time points. For those persons, an LTL change score was calculated by subtracting baseline values from 6-year values. Also, percent change was calculated, and this variable was categorized into 1) shorteners (persons who showed >5% telomere shortening); 2) lengtheners (persons who showed >5% telomere lengthening); and a stable group (persons whose LTL did not substantially change [<5% change] over 6 years). Additionally, a categorization variable with a cutoff of 10% was presented.
Covariates.
Sex and age were documented during the baseline interview, and years of education was assessed at baseline and the 6-year assessments. Measured body mass index (BMI, weight/height2) was divided into underweight (<18.5), normal (18.5–24.9), overweight (25.0–30.0), and obese (>30.0). Alcohol consumption was categorized as nondrinker, mild-moderate drinker (female <14 drinks per week, male <21 drinks per week), or heavy drinker (female ≥14 drinks per week, male ≥21 drinks per week). Smoking status was categorized as current, former, or never. Physical activity was assessed using the International Physical Activity Questionnaire (26) and expressed as overall energy expenditure in metabolic equivalent total minutes per week. The number of self-reported somatic diseases for which participants received medical treatment (i.e., diabetes, osteoarthritis, stroke, cancer, or heart, chronic lung, intestinal, or thyroid diseases) was counted. Finally, current regularly used antidepressant medication was included as a binary covariate (yes/no): tricyclic antidepressants, selective serotonin reuptake inhibitors (SSRIs), and other antidepressants were defined using World Health Organization (WHO) classifications (27). Furthermore, we calculated a derived daily dose for all participants, which consists of the daily dose used by participants divided by the defined daily dose (the mean advised dose) assigned by the WHO for their specific medication. While no baseline associations between antidepressants and LTL were found in this sample (17), and although correcting for antidepressant use might be over-adjusting because they are more likely to be used in the most severe cases, we checked whether associations between LTL and psychiatric characteristics were influenced by antidepressant use.
Data Analyses
Sample characteristics for those who had complete LTL and psychiatric status data on both time points were calculated. Associations between LTL and covariates from both time points (baseline and 6-year follow-up) were analyzed using generalized estimated equations analyses. Generalized estimated equations analyses were performed with an exchangeable correlation structure, which takes within-person correlations due to multiple observations per participant into account (28). Time was coded as 1 (baseline) and 2 (6-year follow-up).
To test whether diagnosis status and symptom severity were consistently related to LTL, generalized estimated equations analyses were conducted with diagnosis status (control, remitted, or current) or symptom severity (Inventory of Depressive Symptoms-Self Report and Beck Anxiety Inventory) at both time points as predictors and LTL at both time points as outcome variable. All participants with available LTL and diagnosis status data on at least one time point were included because generalized estimated equations analyses tolerate missing observations. Analyses included covariates at both time points in different models: 1) sociodemographic characteristics (age, sex, education); 2) addition of health and lifestyle variables (smoking, alcohol use, BMI, chronic disease, and activity); and 3) addition of current medication use (tricyclic antidepressants, SSRIs, other antidepressants). We tested whether sex was an effect modifier by adding a group-by-sex interaction to the model. Similar generalized estimated equations analyses were performed to test whether baseline diagnosis status was associated with LTL over 6 years.
Six-year changes in depression/anxiety status and 6-year changes in LTL were first examined using generalized estimated equations analyses with the 6-year course variable (see Table 2) as predictor and LTL across both time points as outcome. A group-by-time interaction was added to test whether the six course groups predicted differential 6-year LTL attrition over time. Next, linear regression models were used with changes in 1) symptom severity (Inventory of Depressive Symptoms-Self Report and Beck Anxiety Inventory); 2) symptom duration over the 6-year time period; and 3) the number of assessments with diagnoses as predictors, and LTL change score as the outcome, corrected for LTL at baseline.
Results
The mean age of the study sample was 41.8 years (SD=13.1) at baseline, and 66.4% were female (see Table 1). Overall, LTL decreased significantly over the 6-year follow-up period (p<0.001), which corresponded to an average shortening rate of 13.3 bp per year. Baseline LTL predicted LTL at the 6-year follow-up (β=0.478; p<0.001) and explained 22.9% of the variance. There was a strong inverse correlation between baseline LTL and LTL change (β=–0.718; p<0.001), confirming earlier findings that persons with long LTL at baseline have a higher chance of shortening and vice versa (29, 30). At the 6-year follow-up, 27% of the persons showed telomere shortening (27% of control subjects compared with 26% of current patients), 26% showed telomere lengthening (52% compared with 46%), and 47% showed no substantial change in their LTL (21% compared with 28%). Additionally, 13.2% of the total sample showed >10% shortening, and 11.7% showed >10% lengthening. Generalized estimated equations analyses, using LTL and covariates from both baseline and the 6-year follow-up, showed that LTL was negatively associated with age (B=−13.3; p<0.001), male sex (B=−95.5; p<0.001), smoking (B=−80.1; p<0.001), being overweight (B=−48.4; p=0.008) or obese (B=−49.6; p=0.027), and heavy alcohol use (B=−59.5; p=0.013), but not with education, physical activity, the number of chronic somatic diseases, or antidepressant medication and their associated derived daily doses.
Are Diagnosis Status and Symptom Severity Consistently Associated With LTL Over Two Time Points?
Generalized estimated equations analyses including LTL and diagnosis status from both baseline and the 6-year follow-up showed that LTL was shorter in persons with a remitted and a current depressive and/or anxiety disorder compared with control subjects (see Table 3 and Figure 1), replicating our earlier cross-sectional findings (17, 18). LTL of persons with a current or remitted diagnosis did not differ from each other (p=0.645). These associations remained significant after adjustment for health and lifestyle variables (remitted: B=−46.7; p=0.046; current disorder: B=−54.3; p=0.017), as well as current medication use variables. Associations were not different for men and women (diagnosis-by-sex interactions >0.05).
| Variable | B | SE | pa |
|---|---|---|---|
| Control group (reference) | |||
| Remitted diagnosisb | –52.6 | 22.8 | 0.021 |
| Current diagnosisb | –60.8 | 22.4 | 0.007 |
| Time | –66.2 | 12.4 | <0.001 |
| Inventory of Depressive Symptoms-Self-Report | –1.7 | 0.6 | 0.007 |
| Time | –71.9 | 12.1 | <0.001 |
| Beck Anxiety Inventory | –2.3 | 0.9 | 0.009 |
| Time | –69.4 | 11.9 | <0.001 |
a
Generalized estimated equations analyses were adjusted for age at baseline, sex, and education.
b
Diagnosis refers to a depressive and/or anxiety disorder.
FIGURE 1. Leukocyte Telomere Length at Two Time Points by Depressive and Anxiety Disorder Diagnosis Status (N=2,936)a
a The p values are based on comparison of leukocyte telomere length at both time points with the control group in the analyses adjusted for age at baseline, sex, education, and time.
*p<0.05.
Associations were rather similar for depressive and anxiety disorders (depression: remitted [B=−56.1; p=0.015], current [B=−51.8; p=0.033]; anxiety: remitted [B=−34.8; p=0.144], current [B=−71.8; p=0.002]). This, in combination with the high comorbidity between disorders (63% [31]), led us to conduct further analyses with depressive and anxiety disorders combined.
Generalized estimated equations analyses examining the association between LTL and symptom severity across baseline and the 6-year follow-up showed that LTL was negatively associated with scores on the Inventory of Depressive Symptoms-Self Report and Beck Anxiety Inventory (Table 3), also in adjusted models. Results of unadjusted analyses of LTL with diagnosis status and symptom severity (data not shown) closely resembled sociodemographic-adjusted analyses.
Does Baseline Diagnosis Status Predict LTL Attrition Over 6 Years?
Persons with a current depressive or anxiety disorder at baseline did not show different LTL attrition rates compared with control subjects, as shown by nonsignificant time interaction terms (remitted diagnosis-by-time: p=0.796; current diagnosis-by-time: p=0.410) in a sociodemographic adjusted generalized estimated equations model.
Is the 6-Year Course of Depressive and Anxiety Disorders Related to 6-Year LTL Change?
Generalized estimated equations analyses tested whether six groups based on diagnosis status at baseline and at 2-, 4-, and 6-year follow-up (see Table 2) predicted average LTL across baseline and 6-year follow-up. As shown in Figure 2, the remitted group (B=−72.9; p=0.011) and the chronic patients (B=−58.1; p=0.041) had on average shorter LTL than control subjects, adjusted for sociodemographic characteristics. Adjustment for health, lifestyle, and current antidepressant medication use showed similar LTL differences for the remitted group (p=0.015) and the chronic group (p=0.048). To investigate the associations of incidence, remission, or chronicity of depressive or anxiety disorders with LTL attrition rate, a group-by-time interaction was added to the sociodemographic-adjusted model. The absence of a significant interaction (p=0.926) showed that there was no difference in slope between the six groups (i.e., LTL attrition rates were not dependent on the incidence or remission of depression and anxiety disorders). Thus, the pre-existing LTL differences at baseline did not become more pronounced or diminished over 6 years.
FIGURE 2. Leukocyte Telomere Length at Baseline and 6-Year Follow-Up Based on the Course of Depressive and Anxiety Disordersa
a The p values are based on comparison with the control group in the analyses adjusted for age at baseline, sex, education, and time.
*p<0.05.
In addition, changes in depression and anxiety characteristics were analyzed against changes in LTL, adjusted for sociodemographic characteristics (Table 4). This showed that 6-year change in depression (Inventory of Depressive Symptoms-Self Report score) or anxiety (Beck Anxiety Inventory score) severity, percent of time with symptoms over 6 years, and the number of time points with a diagnosis over 6 years were not significantly associated with 6-year change in LTL.
| Variable | β | p |
|---|---|---|
| Change in Inventory of Depressive Symptoms-Self-Report scoreb | 0.021 | 0.271 |
| Change in Beck Anxiety Inventory scorec | 0.005 | 0.795 |
| Percent of time with depressive symptoms | –0.017 | 0.311 |
| Percent of time with anxiety symptoms | –0.007 | 0.649 |
| Number of waves with diagnosis | –0.017 | 0.288 |
a
All analyses are adjusted for age, sex, education, and leukocyte telomere length at baseline.
b
Analyses are additionally adjusted for baseline Inventory of Depressive Symptoms-Self-Report scores.
c
Analyses are additionally adjusted for baseline Beck Anxiety Inventory scores.
Discussion
This large study with 6-year longitudinal data demonstrated that persons with a lifetime depressive or anxiety disorder had consistently shorter LTL than nonpsychiatric control subjects at two time points, irrespective of the current diagnosis status. In line with a dose-response association, we found negative associations between LTL and the severity of depressive and anxiety symptoms, indicating that the most severe patients had the shortest telomeres. In different analytical approaches, diagnosis status or changes in diagnosis status did not correspond with changes in LTL. First, persons with depression or anxiety disorder at baseline did not show accelerated 6-year LTL attrition compared with control subjects. Second, six groups based on the course of depressive and anxiety disorders over four assessments (i.e., control subjects, persistently remitted persons, persons with new onset, persons with relapse/remission, chronic persons) showed no difference in telomere attrition rate. Third, changes in symptom severity, duration of symptoms, and number of study waves with a diagnosis did not correlate with change in LTL. Overall, these findings do not suggest a dynamic relationship: existing differences in LTL did not become larger or reduced over time depending on, for example, whether a person developed more severe, chronic symptoms or recovered.
Our results point toward a between-person rather than a within-person relation of depressive and anxiety disorders and LTL, since a person’s LTL did not change along with changes in that person’s psychopathology. In other words, persons with a lifetime diagnosis were found to have shorter LTL, which represented a difference in “cellular age” of 4 years (remitted diagnosis) and 4.5 years (current diagnosis), and this was irrespective of whether they relapsed or recovered from a disorder. This suggests that either 1) shorter LTL is indeed a consequence of depressive and anxiety disorder-related physiological disturbances, leaving a long-term (cellular) scar, or 2) persons with short LTL, possibly due to genetic heritability (6, 32), are more prone to developing a depressive or anxiety disorder, and LTL attrition rate is not necessarily accelerated when those persons meet a disorder diagnosis. Our results resemble those of Hoen et al. (13) in 206 heart disease patients with depression. Shalev et al. (20) did find greater LTL decrease in 193 men between the ages of 26 and 38 with one or more depressive or anxiety disorders (depression and generalized anxiety disorder) in a 12-year time frame compared with 226 men without a diagnosis, but this association was not reported for women. Substantial study differences might account for the conflicting outcomes regarding LTL change: Shalev et al. had a younger (mean age: 26 years compared with 42 years at baseline) and smaller (N=758 compared with N=1,860) study sample and a longer follow-up period (e.g., 12 years compared with 6 years). Moreover, unlike Shalev et al., our study assessed LTL at baseline and the 6-year follow-up in different batches, which might have caused noise between the two time points. Furthermore, Shalev et al. did not report whether the number of phases with an internalizing disorder increasingly affected LTL change and whether the effect was still significant after adjustment for lifestyle variables.
It is important to note that we found that LTL at baseline was negatively associated with LTL change, indicating that those with short LTL at baseline had a higher chance of lengthening, and vice versa (29). However, despite this compensatory effect that is possibly due to an internal telomere homeostasis system (33), we still found that persons with a current and remitted depressive or anxiety disorder had consistently shorter LTL over the two time points. Something else to consider is that LTL shortening might not be limited to diagnostic categories, and it may be possible that LTL reflects underlying pathophysiological processes that surpass traditional diagnoses (8). Evidence for this comes from studies that confirm shorter LTL among other psychiatric cases, such as those with schizophrenia (34) or PTSD (35).
The major strengths of this study are its large sample size, including well-characterized patients with current and remitted depressive and anxiety disorders, as well as healthy control subjects. Moreover, the sample had a wide age range, and important covariates such as health and lifestyle variables were assessed at multiple time points. Also, LTL was measured reliably with qPCR, and interassay coefficients of variation were sufficiently low. These strengths allowed us to thoroughly examine the longitudinal relation between LTL and depressive and anxiety disorders. However, some limitations of this study should also be noted. First, it should be noted that results of this observational study did not take into account whether persons recovered spontaneously or as a result of psychological or pharmacological treatment, whereas this may actually have an impact on LTL or its regulatory mechanisms (30); possible short-term changes due to treatment may thus not have been captured. Hence, although results of this study are not suggestive of reversibility of LTL shortening after recovery from a depressive or anxiety disorder, multiple in vitro (36, 37) and in vivo (38) studies have systematically shown that such reversibility is indeed possible, possibly even as a result of psychotropic treatment (37, 39, 40). Furthermore, LTL from baseline and the 6-year follow-up was measured 2 years apart, which could have caused noise between the time points. To adjust for possible systematic differences, samples from both time points were rerun together and LTL at follow-up was converted accordingly. Next, as in most studies, we used leukocytes for TL measurement, which is a validated and often-used indicator for cellular aging. A limitation of using average leukocyte TL is that it consists of different cell types. A recent study found different TL change rates for T-cells, B-cells, and monocytes, which makes it difficult to distinguish whether LTL differences are due to actual shortening/lengthening or rather to a redistribution of cell types (41). However, LTL has been found to have a rather high consistency and rather similar shortening rates across various tissues (i.e., buccal cells, skeletal muscle, skin, and subcutaneous fat) (42), but not necessarily all tissues. This suggests that the results of studies conducted in leukocytes are to some extent generalizable to other cell types. Another limitation is that telomerase activity has not been measured, which would be interesting in future research, since telomerase activity may play an important role in the maintenance of LTL, as suggested by several intervention studies (30), and may even mediate beneficial effects of psychotropic medication (43).
In conclusion, this 6-year longitudinal study showed that LTL was consistently shorter for those with a lifetime depressive or anxiety disorder diagnosis, especially among those with the most severe symptoms (dose-response). Importantly, we found that LTL differences over 6 years did not become more pronounced or diminished as a consequence of incidence, remission, or chronicity of depressive or anxiety disorders. While it should be noted that possible short-term treatment effects might not have been captured in this observational study, absolute LTL changes were not related to changes in diagnosis status, symptom severity, or duration. This indicates a static, nondynamic relationship between depressive or anxiety disorders and short LTL and suggests the absence of a direct within-person relationship, since changes in psychiatric characteristics did not correspond with changes in LTL. Future research should elucidate whether short leukocyte telomere length is a long-term consequence or a pre-existing risk factor for the development of depressive or anxiety disorders.
References
1.
Schutte NS, Malouff JM: The association between depression and leukocyte telomere length: a meta-analysis. Depress Anxiety 2015; 32:229–238
2.
Penninx BWJH, Milaneschi Y, Lamers F, et al: Understanding the somatic consequences of depression: biological mechanisms and the role of depression symptom profile. BMC Med 2013; 11:129
3.
Roy-Byrne PP, Davidson KW, Kessler RC, et al: Anxiety disorders and comorbid medical illness. Gen Hosp Psychiatry 2008; 30:208–225
4.
Dowlati Y, Herrmann N, Swardfager W, et al: A meta-analysis of cytokines in major depression. Biol Psychiatry 2010; 67:446–457
5.
Black CN, Bot M, Scheffer PG, et al: Is depression associated with increased oxidative stress? a systematic review and meta-analysis. Psychoneuroendocrinology 2015; 51:164–175
6.
Gotlib IH, LeMoult J, Colich NL, et al: Telomere length and cortisol reactivity in children of depressed mothers. Mol Psychiatry 2015; 20:615–620
7.
Blackburn EH: Switching and signaling at the telomere. Cell 2001; 106:661–673
8.
Lindqvist D, Epel ES, Mellon SH, et al: Psychiatric disorders and leukocyte telomere length: underlying mechanisms linking mental illness with cellular aging. Neurosci Biobehav Rev 2015; 55:333–364
9.
Cawthon RM, Smith KR, O’Brien E, et al: Association between telomere length in blood and mortality in people aged 60 years or older. Lancet 2003; 361:393–395
10.
Schaakxs R, Verhoeven JE, Oude Voshaar RC, et al: Leukocyte telomere length and late-life depression. Am J Geriatr Psychiatry 2015; 23:423–432
11.
Teyssier JR, Chauvet-Gelinier JC, Ragot S, et al: Up-regulation of leucocytes genes implicated in telomere dysfunction and cellular senescence correlates with depression and anxiety severity scores. PLoS One 2012; 7:e49677
12.
Simon NM, Walton ZE, Bui E, et al: Telomere length and telomerase in a well-characterized sample of individuals with major depressive disorder compared to controls. Psychoneuroendocrinology 2015; 58:9–22
13.
Hoen PW, de Jonge P, Na BY, et al: Depression and leukocyte telomere length in patients with coronary heart disease: data from the Heart and Soul Study. Psychosom Med 2011; 73:541–547
14.
Simon NM, Smoller JW, McNamara KL, et al: Telomere shortening and mood disorders: preliminary support for a chronic stress model of accelerated aging. Biol Psychiatry 2006; 60:432–435
15.
Wikgren M, Maripuu M, Karlsson T, et al: Short telomeres in depression and the general population are associated with a hypocortisolemic state. Biol Psychiatry 2012; 71:294–300
16.
Cai N, Chang S, Li Y, et al: Molecular signatures of major depression. Curr Biol 2015; 25:1146–1156
17.
Verhoeven JE, Révész D, Epel ES, et al: Major depressive disorder and accelerated cellular aging: results from a large psychiatric cohort study. Mol Psychiatry 2014; 19:895–901
18.
Verhoeven JE, Révész D, van Oppen P, et al: Anxiety disorders and accelerated cellular ageing. Br J Psychiatry 2015; 206:371–378
19.
Kananen L, Surakka I, Pirkola S, et al: Childhood adversities are associated with shorter telomere length at adult age both in individuals with an anxiety disorder and controls. PLoS One 2010; 5:e10826
20.
Shalev I, Moffitt TE, Braithwaite AW, et al: Internalizing disorders and leukocyte telomere erosion: a prospective study of depression, generalized anxiety disorder and post-traumatic stress disorder. Mol Psychiatry 2014; 19:1163–1170
21.
Penninx BWJH, Beekman ATF, Smit JH, et al: The Netherlands Study of Depression and Anxiety (NESDA): rationale, objectives and methods. Int J Methods Psychiatr Res 2008; 17:121–140
22.
Rush AJ, Gullion CM, Basco MR, et al: The Inventory of Depressive Symptomatology (IDS): psychometric properties. Psychol Med 1996; 26:477–486
23.
Beck AT, Epstein N, Brown G, et al: An inventory for measuring clinical anxiety: psychometric properties. J Consult Clin Psychol 1988; 56:893–897
24.
Lyketsos CG, Nestadt G, Cwi J, et al: The Life-Chart interview: a standardized method to describe the course of psychopathology. Int J Methods Psychiatr Res 1994; 4:143–155
25.
Cawthon RM: Telomere measurement by quantitative PCR. Nucleic Acids Res 2002; 30:e47
26.
Craig CL, Marshall AL, Sjöström M, et al: International physical activity questionnaire: 12-country reliability and validity. Med Sci Sports Exerc 2003; 35:1381–1395
27.
World Health Organization Collaboration Centre for Drug Statistics Methodology: Anatomical Therapeutic Chemical (ATC) Classification System. Olso, Norway, World Health Organization Collaboration Centre for Drug Statistics Methodology, 2007
28.
Twisk JWR: Longitudinal data analysis: a comparison between generalized estimating equations and random coefficient analysis. Eur J Epidemiol 2004; 19:769–776
29.
Epel E: How “reversible” is telomeric aging? Cancer Prev Res (Phila) 2012; 5:1163–1168
30.
Verhoeven JE, Révész D, Wolkowitz OM, et al: Cellular aging in depression: permanent imprint or reversible process? an overview of the current evidence, mechanistic pathways, and targets for interventions. BioEssays 2014; 36:968–978
31.
Lamers F, van Oppen P, Comijs HC, et al: Comorbidity patterns of anxiety and depressive disorders in a large cohort study: the Netherlands Study of Depression and Anxiety (NESDA). J Clin Psychiatry 2011; 72:341–348
32.
Broer L, Codd V, Nyholt DR, et al: Meta-analysis of telomere length in 19,713 subjects reveals high heritability, stronger maternal inheritance and a paternal age effect. Eur J Hum Genet 2013; 21:1163–1168
33.
Liu Y, Kha H, Ungrin M, et al: Preferential maintenance of critically short telomeres in mammalian cells heterozygous for mTert. Proc Natl Acad Sci USA 2002; 99:3597–3602
34.
Lin PY: Shortened leukocyte telomere length in patients with schizophrenia is related to disease status. Schizophr Res 2015; 168:597–598
35.
Jergović M, Tomičević M, Vidović A, et al: Telomere shortening and immune activity in war veterans with posttraumatic stress disorder. Prog Neuropsychopharmacol Biol Psychiatry 2014; 54:275–283
36.
Jaskelioff M, Muller FL, Paik JH, et al: Telomerase reactivation reverses tissue degeneration in aged telomerase-deficient mice. Nature 2011; 469:102–106
37.
Wei YB, Backlund L, Wegener G, et al: Telomerase dysregulation in the hippocampus of a rat model of depression: normalization by lithium. Int J Neuropsychopharmacol 2015; 18:pyv002
38.
Batista LFZ, Pech MF, Zhong FL, et al: Telomere shortening and loss of self-renewal in dyskeratosis congenita induced pluripotent stem cells. Nature 2011; 474:399–402
39.
Wolkowitz OM, Mellon SH, Epel ES, et al: Resting leukocyte telomerase activity is elevated in major depression and predicts treatment response. Mol Psychiatry 2012; 17:164–172
40.
Zhou QG, Hu Y, Wu DL, et al: Hippocampal telomerase is involved in the modulation of depressive behaviors. J Neurosci 2011; 31:12258–12269
41.
Lin Y, Damjanovic A, Metter EJ, et al: Age-associated telomere attrition of lymphocytes in vivo is co-ordinated with changes in telomerase activity, composition of lymphocyte subsets and health conditions. Clin Sci (Lond) 2015; 128:367–377
42.
Daniali L, Benetos A, Susser E, et al: Telomeres shorten at equivalent rates in somatic tissues of adults. Nat Commun 2013; 4:1597
43.
Bersani FS, Lindqvist D, Mellon SH, et al: Telomerase activation as a possible mechanism of action for psychopharmacological interventions. Drug Discov Today 2015; 20:1305–1309
Information & Authors
Information
Published In
History
Received: 13 July 2015
Revision received: 7 September 2015
Revision received: 5 November 2015
Accepted: 30 November 2015
Published online: 4 March 2016
Published in print: June 01, 2016
Authors
Competing Interests
Dr. Penninx has received research funding (not related to this study) from Jansen Research. All other authors report no financial relationships with commercial interests.
Funding Information
NWO-VICI grant: 91811602
Supported by an NWO-VICI grant (number 91811602) to Dr. Penninx, Josine Verhoeven, and Dóra Révesz for telomere length assaying. The infrastructure for the Netherlands Study of Depression and Anxiety (www.nesda.nl) is funded through the Geestkracht program of the Netherlands Organization for Health Research and Development (Zon-Mw, grant number 10-000-1002) and is supported by participating universities and mental health care organizations (VU University Medical Center, GGZ inGeest, Arkin, Leiden University Medical Center, GGZ Rivierduinen, University Medical Center Groningen, Lentis, GGZ Friesland, GGZ Drenthe, Institute for Quality of Health Care [IQ Healthcare], Netherlands Institute for Health Services Research [NIVEL], and Netherlands Institute of Mental Health and Addiction [Trimbos]).Metrics & Citations
Metrics
Citations
Export Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.
For more information or tips please see 'Downloading to a citation manager' in the Help menu.
View Options
View options
PDF/EPUB
View PDF/EPUBGet Access
Login options
Already a subscriber? Access your subscription through your login credentials or your institution for full access to this article.
Personal login Institutional Login Open Athens loginNot a subscriber?
PsychiatryOnline subscription options offer access to the DSM-5-TR® library, books, journals, CME, and patient resources. This all-in-one virtual library provides psychiatrists and mental health professionals with key resources for diagnosis, treatment, research, and professional development.
Need more help? PsychiatryOnline Customer Service may be reached by emailing [email protected] or by calling 800-368-5777 (in the U.S.) or 703-907-7322 (outside the U.S.).