Ghrelin as a Biomarker of Stress: A Systematic Review and Meta-Analysis

Introduction: Ghrelin is an orexigenic hormone which favors food-seeking behavior and has been postulated to be a biomarker of stress. We conducted a systematic review and meta-analysis on the evolution of ghrelin levels following acute stress. Methods: The PubMed, Cochrane Library, Embase, and ScienceDirect databases were searched for studies reporting ghrelin levels before and after acute stress in humans. Results: We included ten studies for a total of 348 patients. Acute stress (intervention) was always in a laboratory. Acute stress was psychological (Trier Social Stress Test), physical, or mixed (cold pressure test). The overall meta-analysis demonstrated an increase in ghrelin after the stress intervention (ES = 0.21, 95CI 0.09 to 0.34) compared with baseline levels. Stratification by time demonstrated an acute increase in ghrelin levels in the five minutes immediately following the initiation of stress (0.29, 0.10 to 0.48) but without any difference after. Obese individuals had a more significant (ES = 0.51, 95CI 0.18 to 0.84) and prolonged increase in ghrelin levels for up to 45 min compared with non-obese individuals who had a significant increase only five minutes after stress. Moreover, the ghrelin levels increased in response to stress with BMI (coefficient 0.028, 0.01 to 0.49; p = 0.013) and decreased with the time after the stress intervention (coefficient -0.007, −0.014 to −0.001; p = 0.025). Conclusion: Ghrelin is a biomarker of stress, with a short-term increase following acute stress. Obese individuals have both a higher and prolonged response, emphasizing the link between obesity and stress.


Introduction
Psychosocial stress has been known as a major public health problem for more than 30 years [1]. Identifying acute stressful events with objective measures is growing in interest in physiology and preventive medicine [2]. Biomarkers of stress may help to assess specific working conditions and develop preventive strategies for stress management that can be evaluated. Despite the most common biomarkers of stress derived from the hypothalamicpituitary-adrenal axis being associated with the physiological stress response [3], other biomarkers have been proposed. Ghrelin is a 28-amino-acid peptide discovered in 1996, most commonly known as one of the main appetite regulation hormones, with an orexigenic role [4]. Increased ghrelin levels influence meal initiation and food-seeking behavior [5]. However, besides its multiple functions and despite conflicting results, ghrelin has been suggested to be a biomarker of stress [4], due to increased levels following acute stress. Under normal physiological conditions, ghrelin exhibits high and immediate variations in response to a meal [6], with a short circulating half-life of less than half an hour [7,8]. Ghrelin has an acute, immediate, and non-prolonged response to a stimulus, suggesting that the increase in ghrelin levels in response to acute stress may have a short duration. However, the results are equivocal concerning the ghrelin response to acute stress. To our knowledge, no meta-analysis to date has examined the effects of acute stress on ghrelin levels. Moreover, stress and obesity are intrinsically linked [9]. Stress has been suggested to lead to obesity via inappropriate eating behaviors [10]. Therefore, we hypothesized that obese individuals will exhibit a larger response in ghrelin levels following acute stress. Understanding the possible impact of ghrelin in the biological relationship between obesity and stress may help to determine an efficient preventive strategy.
Thus, we aimed to conduct a systematic review and meta-analysis to demonstrate that ghrelin could be a biomarker of acute stress, with a short-term increase following acute stress, and that the ghrelin response may be greater in obese individuals compared with normal-weight individuals.

Literature Search
We searched PubMed, Cochrane Library, Embase, and ScienceDirect with the following keywords, "ghrelin" and "stress", on 15 September 2020, for all articles describing our primary outcome variable-i.e., the measurement of ghrelin levels before and after acute stress-with or without a control group. We chose to use broad keywords to retrieve all possible articles. The search was not limited to specific years, languages, or regions. Additionally, reference lists of all publications that met the inclusion criteria were manually searched to identify any further studies that were not found with the electronic search, as well as reference lists from reviews. We excluded studies on animals and patients with psychiatric disorders. Two authors (J.-B.B.-M. and M.T.) conducted all literature searches, collated the abstracts, and separately reviewed the abstracts, and based on the selection criteria, decided the suitability of the articles for inclusion. A third author (F.D.) was asked to review the articles where consensus on suitability was debated. All the authors then examined the eligible articles. The search strategy is presented in Figure 1.

Quality of Articles and Study Designs
Using the NOS criteria, all of the ten included studies had a score of >6 and considered high-quality ( Figure 2). All the studies were cohort studies without blind assessment and mentioned ethical approval.

Quality of Assessment
We used the Newcastle Ottawa Scale (NOS) [11] to evaluate the quality of the included articles. The NOS was developed to assess the quality of studies based on three types of biases: bias of selection, bias of comparability, and bias of exposure. Those three items were evaluated through eight items or sub-items, with one point given when the study fulfilled the criteria. The maximal score was eight, which was then converted into a percentage.

Statistical Considerations
The statistical analysis was conducted using Stata software (version 16, StataCorp, College Station, TX, USA). The ghrelin levels were summarized for each study sample and reported as the mean and the standard-deviation (SD) before and after the acute stress. Random effects meta-analyses (DerSimonian and Laird approach) were conducted when data could be pooled. p values less than 0.05 were considered statistically significant. We described our results by calculating the effect size (ES, standardized mean differences as SMD) of ghrelin levels from before and after acute stress. A positive ES denotes an increase in ghrelin levels. A scale for ES has been suggested with 0.8 reflecting a large effect, 0.5 a moderate effect, and 0.2 a small effect. In particular, we conducted three main meta-analyses stratified by time after the withdrawal of the stress: a global metaanalysis with all participants, a meta-analysis with only normal-weight individuals, and a meta-analysis with only obese individuals. All the articles included reported several measurement times of ghrelin levels following acute stress. All the meta-analyses were stratified by time after stress. The number of stratifications was determined by the number of studies within each stratification. For example, the main meta-analysis was stratified on four time measures: <5 min, 5-15 min, 15-45 min, and >45 min. Heterogeneity in the study results was evaluated by examining forest plots, and confidence intervals (CI), and by using formal tests for homogeneity based on the I 2 statistic, which is the most common metric for measuring the magnitude of between-study heterogeneity and is easily interpretable. I 2 values range between 0% and 100% and are typically considered low at <25%, modest at 25-50%, and high at >50%. For rigor, funnel plots of these meta-analyses were used to search for potential publication biases. In order to verify the strength of the results, further meta-analyses were then conducted, excluding studies that were not evenly distributed around the base of the funnel. We also tried to compute a meta-analysis stratified on the type of sampling (blood, saliva, etc.) or the type of stress.
Meta-regression analyses were conducted to explore the influence of the study or the participants' characteristics on the standardized mean differences. The following characteristics were considered when available: age, sex of the participants (male versus female), body mass index (BMI) and other sociodemographic variables, characteristics of the acute stress (duration, type of stress), and the time of sampling. Results were expressed as regression coefficients and 95%CI.

Results
The initial search retrieved 6631 putative articles (60 in the Cochrane Library, 651 in PubMed, 82 in Embase, and 6723 in ScienceDirect). The removal of duplicates and the use of the selection criteria reduced the number of articles reporting the evaluation of ghrelin level in blood or saliva to ten articles for the systematic review [12][13][14][15][16][17][18][19][20][21] and nine for the meta-analysis-with one study not reporting the number of participants [15] (Figure 1). All all articles were written in English.

Quality of Articles and Study Designs
Using the NOS criteria, all of the ten included studies had a score of >6 and considered high-quality ( Figure 2). All the studies were cohort studies without blind assessment and mentioned ethical approval.

Outcome and Aim of the Studies
The main objective of all of the included studies was to assess the effect of acute stress instigated in a laboratory on ghrelin levels (and other hormones).

Type of Stress
Seven studies used the Trier Social Stress Test [13][14][15][16][17][18][19]. This procedure combines social stress (public speech] with mental stress (arithmetic under time pressure) and is a validated tool to provoke psychobiological stress responses. Two studies used the Cold Pressor Test [20,21], i.e., the immersion of the non-dominant hand up to the wrist in a rectangularshaped container of 0 • C ice-water for 2 min. One study used a physical stressor (maximum oxygen uptake treadmill running test) [12] (Table 1).

Meta-analyses in Normal-weight Population were Based on Eight Studies (28 groups).
Stratification by time demonstrated an increase in ghrelin levels in the 5 minutes following the stress intervention (0.23, 0.01 to 0.45) but without a difference between 5 and 45 minutes (0.10, −0.09 to 0.28) and after 45 minutes (0.02, −0.20 to 0.24), nor in overall results (0.11, −0.01 to 0.23) (Figure 4).

Meta-regressions and Sensitivity Analysis
Levels of ghrelin after stress increased with BMI (ES = 0.026, 95CI 0.004 to 0.48). Ghrelin levels decreased with time after the stress intervention (−0.007, −0.014 to −0.001). Sociodemographic variables (age, sex) did not influence the level of ghrelin, nor did the duration of stress or time of sampling ( Figure 6). A sensitivity analysis demonstrated similar results after the exclusion of studies that were not evenly distributed around the funnel plot (Figure 7). Insufficient data precluded further analyses by type of stress (mental, physical, or both) or by type of sampling (blood, saliva, etc.).

Meta-Regressions and Sensitivity Analysis
Levels of ghrelin after stress increased with BMI (ES = 0.026, 95CI 0.004 to 0.48). Ghrelin levels decreased with time after the stress intervention (−0.007, −0.014 to −0.001). Sociodemographic variables (age, sex) did not influence the level of ghrelin, nor did the duration of stress or time of sampling ( Figure 6). A sensitivity analysis demonstrated similar results after the exclusion of studies that were not evenly distributed around the funnel plot (Figure 7). Insufficient data precluded further analyses by type of stress (mental, physical, or both) or by type of sampling (blood, saliva, etc.).

Discussion
The main findings were that ghrelin is a biomarker of stress, with a short-term increase following an acute stress intervention. Moreover, overweight and obese individuals had an extended response compared to normal-weight individuals, demonstrating the link between obesity and stress.

Ghrelin as a Biomarker of Stress
The active form of ghrelin (acyl-ghrelin) is a 28-amino-acid peptide discovered in 1996 [22], with multiple functions [23] but is mainly an appetite-regulating hormone produced by the stomach [24] whish acts on the central nervous system [25]. Very interestingly, we demonstrated that acyl-ghrelin is also a biomarker of stress that increases following exposure to a stress environment. Acyl-ghrelin is known for two main effects: (1) balancing energy by promoting food intake [26,27] and secreting of insulin [28] and ACTH, and (2) release of growth hormone (GH) [23,29]. The increase in ghrelin in response to stress can be an adaptive mechanism that may help people with stress [30]. Moreover, high ghrelin levels seem to influence the learning process and memory [31,32]. From a biomolecular point of view, ghrelin is the result of two cleavages from the 117-amino-acids pre-pro-ghrelin [33] followed by an octanoylation by the Ghrelin O-Acyltransferase (GOAT) to produce the active form of ghrelin (acyl-ghrelin). Despite demonstrating an increase in active ghrelin levels following acute stress, the mechanisms of the increase are not yet known, i.e., an increase of the production of the pre-pro-ghrelin, an increase of the cleavages, and/or an increase of the GOAT octanylation. Lastly, contrary to common thoughts, ghrelin and GOAT are not only expressed in the stomach [25,34]. They are also expressed in different tissues such as the pancreas, small-intestine, anterior pituitary gland, placenta, lymphocyte, kidney, gonads, lung, brain, and hypothalamus [24,25]. Hypothalamus expression is mediated by the circadian clock and environmental factors such as stress.

Acute Response to Acute Stress Intervention
We demonstrated that ghrelin levels had a short-term increase in response to acute stress, with high levels immediately after the stressful event and then a gradual decrease. This could be explained in part by its short half-life of between 10 and 31 min [7,8]. Therefore, ghrelin can be considered a biomarker of acute stress, such as catecholamines, heart rate variability, or cytokines [35,36]. Moreover, ghrelin is also implicated in the regulation of stress through its action on the hypothalamic-pituitary-adrenal axis [30]. Physiologically, the main role of ghrelin is for meal initiation and food-seeking behavior [6]. Daily variations of plasma ghrelin levels show transient increase in pre-prandial periods followed by a decrease in post-prandial periods [37,38]. The decrease has been proposed to be triggered by absorption of nutrients within the intestine [39]. However, the decrease in plasma ghrelin levels following acute stress cannot be linked with meal stimulus, so other mechanisms of regulation should apply. Ghrelin also has a short-term role in the regulation of mental health; it is implicated in short-term adaptations against depression, and in the control of anxiety after acute stress [30].

Ghrelin Levels and Characteristics of Stress
In our meta-analysis, ghrelin levels globally increased with whichever the type of stress was involved (physical, mental, or both). Even if there was no statistical difference between the types of stress, the literature is scarce regarding the exploration of ghrelin levels during physical stress [12] and both physical and mental stress [20,21]. Most of the studies included in our meta-analysis used the (Trier Social Stress Test) TSST, one of the most common tools to induce acute mental stress. Even though a dose-response relationship has been demonstrated between the duration or the type of stress and some biomarkers of stress, we failed to demonstrate a dose-response relationship between the duration of stress and ghrelin levels. Moreover, none of the nine included studies reported such relations. A limited number of studies with physical stress precluded further analyses. However, studies of other biomarkers suggest a larger increase when both physical and mental stress is induced [20]. In our meta-analyses, as well as in literature, sociodemographic variables (age, sex) did not influence ghrelin levels. However, larger increases in cytokines have been reported in older individuals under stressful conditions [40]. Lastly, even if insufficient data precluded further analyses, blood sampling is commonly reported as being a better way to assess stress compared with saliva or urine sampling [41].

A prolonged Response in Overweight and Obese Individuals
Even though a positive effect size was identified for the overall population, there is a difference between normal-weight and overweight people. Indeed, the normal-weight population's effect size was small and only significant during the first five minutes. This could impact our conclusions on the overall meta-analysis and possibly reduce the generalization of our results. In contrast, we showed a higher and longer increase in ghrelin levels in the overweight patients than in the normal-weight patients, in accordance with the literature [42]. Moreover, we demonstrated that levels of ghrelin after stress increased with BMI. Stress and obesity are two public health issues [43,44] that are intrinsically linked [9]. It has been proposed that stress can lead to obesity via inappropriate eating behaviors [10] through a non-adaptive response to ghrelin levels [9]. In addition, obesity is a major stress factor [45]. Stressed people are those who have the greatest difficulties losing weight [46]. The relationship between obesity and stress is biological via the main stress hormones regulating appetite (leptin, ghrelin) [47,48]. This relationship between obesity and stress facilitated an international recommendation that suggests implementing stress management programs in obesity for a sustainable weight loss [49]. The direction of the relationship between the heightened and prolonged ghrelin response to stress in overweight and obese individuals requires further consideration. It is not yet possible to determine if a higher ghrelin response to stress contributes to the development of obesity, or if obese individuals have developed a heightened ghrelin response as a result of weight gain. The results from our studies therefore have implications to use ghrelin as a screening tool to identify people at risk of obesity. Although associations between ghrelin levels, obesity and stress exist, causal pathways have not been established [50,51]. The biological pathways of ghrelin response to acute stress are yet to be fully understood [52], and therefore the biological explanations of a greater ghrelin response following acute stress in obese individuals are also lacking.

Limitations
Our study has some limitations. Noon of the included studies had a randomized controlled design. Moreover, blinding interventions were deemed infeasible. Though there were similarities between the inclusion criteria, they were not identical. Limiting our meta-analysis to studies sharing precisely the same inclusion and exclusion criteria was not possible due to the limited data. All of the studies were monocentric, limiting the generalizability of our results. However, our selection of articles was rigorous, and few studies were outliers considering meta-funnels. The low I 2 to the attested homogeneity between the studies. Moreover, the homogeneity of the data precluded further sensitivity analysis of the different methods for measuring ghrelin levels (ELISA and RIA), stressors (TSST, sport, and CPT), and sampling (blood and saliva). The high I 2 (69.1%) for metaanalysis in obese individuals might limit the generalization. However, we demonstrated a positive effect of BMI in our meta-regression.

Conclusions
Ghrelin is a biomarker of stress, with a short-term increase following acute stress. Obese individuals have both a higher and prolonged response, emphasizing the link between obesity and stress. Appetite regulation testing warrants further investigation for use as part of weight management strategies, especially in overweight/obese individuals.