Introduction

Bariatric surgery (BS) is an effective treatment for obesity that is being used increasingly worldwide; however, physical changes before or after surgery can induce significant stress1. People with severe obesity awaiting BS may refer psychiatric symptomatology2 and poor quality of life3. Moreover, personality traits and different degrees of psychopathology may influence outcomes after BS1. Nevertheless, research regarding psychopathology and psychological factors for prediction of successful weight loss have been inconsistent so far. In this sense, some studies4,5,6,7 but not all8,9 have reported a negative relationship between presurgical psychopathological symptoms, especially depressive and anxiety disorders with postsurgical weight loss.

Also, BS has been shown to ameliorate health-related quality of life and eating disorder symptoms. However, the correlation of these changes with weight loss is not uniform, suggesting that additional factors have an impact on postoperative outcomes10. A growing body of evidence indicates that individuals being considered for BS are prone to exhibiting eating disorders and/or problematic eating behaviours, with research further indicating the potential persistence or emergence of these issues following the surgical procedure11. The diversity of eating-related behaviours and psychopathology presentations in BS patients have created inconsistencies in the literature in relation to the assessment methodology, diagnosis and treatment12.

These discrepancies could arise from factors like the duration of the follow-up period and differences in the utilization of assessment tools. Thus, presurgical psychological assessment should include instruments to evaluate psychopathological symptoms thoroughly. In that sense, enhancing comprehension of the psychological attributes exhibited by individuals undergoing BS would hold the potential to empower clinicians in pinpointing and effectively addressing distinct areas of concern13. Furthermore, such an understanding could potentially elucidate the extent to which motivational stages can serve as predictive indicators for behavioural changes14. Given that severe and inadequately managed mental health conditions, such as major depression, anxiety, bipolar disorder, and eating disorders, could potentially be viewed as contraindications for BS15,16,17, presurgical psychological assessment is usually focused on detecting severe mental disorders. However, mild intensity psychopathology may also influence postsurgical outcomes; therefore, multidimensional instruments capable of detecting mild intensity psychopathology and maladaptive eating behaviours in candidates for BS might enable treatment that might improve postsurgical outcomes.

An increasing number of assessment instruments have been validated for use in BS samples. The Personality Assessment Inventory (PAI) appears particularly useful for assessing candidates for BS18. Specifically, in this population, significantly high scores on the Positive Impression Management, Somatization, Depression, Dominance, and Warmth scales of the PAI are relevant7,18. A recent retrospective study found that psychological factors, as assessed by the PAI, affected long-term maintenance of weight loss after BS; anxiety and mania were associated with greater initial weight loss and faster weight regain after about 3 years, and alcohol problems were associated with worse BMI outcome19.

Moreover, binge eating stands out as the disorder most significantly impacted by BS. According to a comprehensive review, the prevalence of presurgical binge eating disorder spans from 2 to 49%, while subclinical binge eating behaviours are reported at rates ranging from 6 to 64% across various studies20. Notably, research indicates that the scores on the Binge Eating Scale (BES) show marked improvement postsurgery. Noteworthy findings also highlight a drastic reduction in the proportion of participants classified under the severe BES category, plummeting from 78% to a mere 5% (p < 0.01)21. Additionally, the severity of problematic eating behaviours, as assessed by the Eating Disorder Examination-Self-Report Questionnaire (EDE‐Q), demonstrates a consistent decline after surgery, maintaining levels lower than baseline throughout the follow-up period22. Given its validity within the Hispanic population, the EDE‐Q can be effectively integrated into preoperative psychological assessment batteries to screen for eating disorder psychopathology23. In light of these findings, both the binge eating scale (BES)24 and the Eating Disorder Examination-Self-Report Questionnaire (EDE‐Q)25 have proven to be robust instruments for the evaluation of these complex disorders.

In this study, we employed the PAI, EDE-Q, and BES to assess both the psychopathological profile and eating-related behaviours among individuals seeking bariatric surgery (BS), while also investigating potential changes in this profile 1-year post-surgery. The proposed mechanisms underlying the psychopathological shifts following BS are intertwined with the notion that individuals with obesity often contend with an adverse body image, thereby giving rise to responsive anxious-depressive manifestations and maladaptive conduct. BS, yielding substantial weight reduction, holds the potential to engender an amelioration in body image perception, subsequently culminating in the alleviation of maladaptive symptoms.

Materials and methods

Subjects

The recruitment of participants was on the basis of adding consecutive patients ready for BS during a period of time from March 2016 to March 2017 in a university hospital. Patients were considered ready after the selection by a multidisciplinary team led by an Endocrinologist and assisted by a Psychologist/Psychiatrist. All were aged ≥ 18 years with moderate-to-severe obesity, body mass index (BMI) ≥ 40 kg/m2 (or BMI ≥ 35 kg/m2, in patients with weight-related comorbidities), at least 6 months of failure to achieve and maintain weight loss in several diet programs, capacity to understand the surgical procedure and its effects, motivation for dietary monitoring (loss of 5% of weight in the 3 months prior to referral to surgery) and informed written consent to participate in the study. Exclusion criteria were: uncontrolled major psychiatric pathology or that prevents postoperative compliance, impaired judgment and/or active toxic habits, severe active eating disorder, unacceptable anesthetic risk, pregnancy or short-term desire, lactation, inflammatory bowel disease and evolutionary diseases of limited vital prognosis (neoplastic or systemic). All procedures involving the participants were performed according to the principles outlined in the Declaration of Helsinki, and the protocol was approved by the local ethics committee (CEIm 2015621).

Design

In this prospective study, a multidimensional assessment inventory called PAI was employed to comprehensively assess psychopathological symptomatology26,27,28, and specific inventories for eating behaviour disorders, namely BES and EDE-Q24,25, were used to evaluate changes before and after BS, in comparison with normative data from the general population.

Procedure

After endocrinological evaluation and a first appointment with the surgeon, candidates were referred to a psychiatrist for routine psychological evaluation comprising (a) a psychosocial interview including questions about personal and educational background, weight and weight loss trajectory, medical history, and adherence to medical recommendations; (b) the Structured Clinical Interview for DSM29; and (c) the Bulimic Investigatory Test, Edinburgh30. Patients were considered eligible for the study after this initial routine screening. Thereafter, they signed informed consent and were assessed with the PAI26, BES24 and EDE‐Q25. Patients were reassessed with these three instruments before BS and 1 year after surgery.

Measures

The PAI26 is a self-administered test of personality and psychopathology features comprising 22 nonoverlapping full scales: 4 validity scales (infrequency, inconsistency, negative impression management, and positive impression management), 5 complementary validity scales (end questionnaire inconsistency, simulation index, Rogers discriminant function, defensiveness index, and Cashel discriminant function), 11 clinical scales (somatic complaints, anxiety, anxiety-related disorders, depression, mania, paranoia, schizophrenia, borderline features, antisocial features, alcohol problems, and drug problems), and 2 interpersonal scales (dominance and warmth). Ten of these scales are divided into 31 conceptually defined subscales. Participants qualify items on a Likert scale ranging from 1 (totally false, not at all true) to 4 (very true). The PAI also yields five treatment indicators (aggression, suicidal ideation, nonsupport, stress, and treatment rejection). To enable comparison with a census-matched community sample, scores on all PAI scales and subscales were transformed to T-scores. Transformed T-scores have a mean of 50 and a standard deviation of 10; thus, T-scores > 50 are above the mean value in the community sample, T-score > 60 is considered clinically relevant and T-scores ≥ 70 (two standard deviations above the mean) signify a marked deviation from the typical response in the community sample. Combining the clinical and personality scales and subscales can identify various psychopathological profiles, 15 clinical syndromes and 10 personality disorders26,27.

In addition, eating disorders were evaluated with these two questionnaires to detect mild altered behaviour pattern and attitudes in front of food. Firstly, the 36‐item EDE‐Q25 contains 14 attitudinal questions that assess the severity of eating pathology over the previous 28 days. Participants qualify items on a Likert scale ranging from 0 (none) to 6 (every day), with higher scores reflecting either greater severity or higher frequency of eating pathology. Item 1–15 and 29–36 generate the global score, which is the average score of four subscales: restraint, eating concern, weight concern, and shape concern. It is particularly valuable for assessing behaviour related to anorexia nervosa, bulimia nervosa, binge eating disorder, and other specified feeding and eating disorders. A score ≥ 5 points on the global scale and ≥ 5 on each of the subscales is considered clinically relevant.

The BES24 is a 16-item instrument widely used as a self-report measure of the severity and frequency of binge eating episodes and related psychological features. Participants respond to each item on a Likert scale, indicating the degree to which each statement applies to them. The scale assesses various aspects related to binge eating behaviour, including: frequency of binge eating, emotional aspects, loss of control, guilt and shame. The BES is useful for screening and diagnosing binge eating disorder, as well as tracking changes in binge eating behaviour over time, such as during treatment. A score of 17 is considered the most appropriate clinical cutoff point based on the sensitivity and specificity values of the test.

Data analysis

Data from the PAI were prepared as recommended by Morey26,27: cases were excluded when (a) > 5% of questions were unanswered, (b) inconsistency T-scores were > 73, or (c) infrequency T-scores were > 75.

To test the normality of variables’ distribution, we used the Shapiro–Wilk test. In direct scores, we summarized normally distributed variables as means with standard deviations and non-normally distributed variables as medians and interquartile ranges. To compare with normative group, we used independent t-tests, the Mann–Whitney U, or chi-square tests, as appropriate. To compare pre-surgical and post-surgical parameters, we used paired t-tests, the Wilcoxon signed-rank test, or the McNemar test, as appropriate. Significance was set at p < 0.05. We utilized interclass correlation coefficients along with their corresponding 95% confidence intervals to assess the agreement or similarity between the test scores. We used Stata Statistical Software, version 13.1 (StataCorp LP; College Station, TX, USA) for all analyses.

Ethical statement

All procedures involving the participants were performed according to the principles outlined in the Declaration of Helsinki, and the protocol was approved by the local ethics committee. This study protocol was reviewed and approved by the CEIm Committee, which is accredited by The Office For Human Research Protections of the United States Department of Health and Human Services, Federalwide Assurance (FWA), FWA00025624, approval number [CEIm 2015621].

Consent to participate

Written informed consent was obtained from study participants. All of them signed a standard document to ensure the understanding of the objective and the signing the agreement to participate.

Results

Participants

A total of 42 individuals with moderate-to-severe obesity sought BS at our university hospital during this period of time. Of these, the routine screening excluded 6 for medical or surgical contraindications to BS and 4 for severe, active eating or mental disorders (1 individual met both exclusion criteria).

Thus, we analyzed data from 32 individuals: 21 (65.6%) women; mean age, 49.2 ± 6.8 years, range, 32‒59 years). At the presurgical evaluation, 22 (68.8%) were employed. BS consisted of sleeve gastrectomy in 15 (45.5%) patients and gastric bypass in 18 (54.5%). Twenty-six individuals were evaluated after surgery (loss to follow-up 18.75%) Mean BMI after surgery was 29.46 (5.1) kg/m2 after a mean follow-up of 14.1 (3.6) months.

Table 1 reports patients’ demographics and clinical history.

Table 1 Participants’ demographics and clinical history.
Full size table

Comparison of PAI scales with the normative group before BS

The PAI was completed by 32 (100%) participants before surgery. None of the questionnaires were invalidated, because no T-scores on the validity scales were outside the acceptance protocol. Mean T-scores on the clinical scales ranged from 43.9 ± 9.7 for mania to 67.1 ± 14.1 for somatic concerns. The overall mean score 51.7 ± 10.5 was virtually identical to that of the standardization sample. The psychometric profile (stress 57 ± 13.2; anxiety 51 ± 11; depression 59.8 ± 13.3; somatic complaints 67 ± 14.1, and obsessive–compulsive disorder 49.7 ± 10.9) suggested that our cohort had mild somatic complaints with mild intensity mixed adjustment disorder. Table 2 shows the clinical scales, with the highest scores for somatic complaints and depression, however, only somatic complaints were high enough to be considered clinically relevant (T > 60). Table 3 shows the subscales’ scores. T-scores > 60 was considered clinically relevant. Among the somatic complaints, health concerns, conversion, and somatization together with the physiological depression subscale exhibited clinically relevant scores.

Table 2 Mean T-scores full-scale scores on the Personality Assessment Inventory before and after bariatric surgery.
Full size table
Table 3 Mean T-scores subscale scores on the Personality Assessment Inventory before and after bariatric surgery.
Full size table

In summary, the presurgical psychometric profile suggested a mild intensity mixed adjustment disorder focused on somatic complaints. No psychopathological personality disorders were found. T-scores for other aspects in the psychometric profile were around 50, corresponding to the mean values in the general population.

PAI full-scale and subscale profiles after BS

Twenty-six patients (81.25%) were assessed. At clinical scales, patients improved significantly after surgery on somatic complaints (SOM p < 0.001) and depression (DEP p = 0.04). At treatment-related scales, treatment rejection was also significantly increased (RTR p = 0.03) (Table 2).

In relation to subscales, after surgery, patients improved significantly in all somatic complaint areas (SOM-C p = 0.01; SOM-S p = 0.05; SOM-H p < 0.001), especially in concern for health (SOM-H: health concerns). Regarding anxiety-related disorders, the obsessive–compulsive symptoms decreased (TRA-O p = 0.04) and regarding psychotic disorders, cognitive distortions also decreased (ESQ-A p = 0.03) (Table 3).

Eating behaviour before and after BS

Before surgery, the absence of eating behaviour disorders is noted in our cohort. In BES n = 32 (100%) the mean score was 5.29 ± 4.12 and in EDE-Q/G n = 32 (100%) the global men score was 2.44 ± 0.88. Restrictive eating behaviour and cognitive symptoms of concern for weight and body shape improved significantly after surgery, becoming in line with the values of the general population. After surgery N = 26 (81.25%), significant improvements were observed in scores for compulsive intake (BES p < 0.001), restrictive behaviour (EDE-Q/R p < 0.001), concern for weight (EDE-Q/W p < 0.001), shape concern (EDE-Q/S p < 0.001), and overall key behaviours of eating disorders and related cognitive symptoms (EDE-Q/G p < 0.001). After surgery, patients reported lower occurrence and frequency of key behavioural traits in eating disorders, including compulsive eating, misuse of laxatives and diuretics, and concern for weight and body shape. However, concern for food did not change significantly (EDE-Q/E p = 0.73) (Table 4).

Table 4 Eating behaviour data.
Full size table

Discussion

BS can be an effective treatment for obesity, although the presence of psychopathology, even of mild intensity or subclinical, eating-related behaviours and certain personality traits may affect postsurgical outcomes19. To our knowledge, this is the first study to prospectively assess patients with the PAI before and after BS. Our findings using the multidimensional PAI add strength and depth to those reported in previous studies using unidimensional psychological tests that reported a higher prevalence of mental health disorders (especially symptoms of depression or anxiety) in people with obesity as well as improvements after BS31,32,33,34,35,36.

In the present study by PAI, mild somatic complaints, especially health concerns, depressive and anxiety symptoms and compulsive personality traits were more common in participants than in the general population. We found that patients’ presurgical profile was compatible with a mild intensity mixed (anxious-depressive) adjustment disorder. In this sense, the only scale with mean clinically relevant scores (T > 60) was somatic complaints, particularly health concerns, characterized by excessive concern about having a serious illness or the likelihood of becoming ill despite a lack of medical evidence for illness. This dimension is closely related to depression and stress, which also involve feelings of worry and anxiety, while generating frustration and helplessness. A comparable psychopathological clinical profile has been previously documented utilizing alternative instruments to the PAI in 15% of obese patients awaiting BS or in hyperphagic obesity up to 30%22. In the same direction as the PAI, the baseline EDE-Q identified symptoms of cognitive anxiety about weight and body shape that were closely related to a somatic complaint, and these symptoms significantly improved 1 year after surgery. Moreover, while the PAI did not diagnose any personality disorders among our patients, it did illuminate certain associated personality traits. These traits were typified by a compulsive and obsessive behavioural pattern, alongside the presence of distorted or maladaptive thoughts concerning their cognitive perceptions of health status. According to the results of the PAI, after surgery, the profile improved by a noteworthy decrease in somatic complaints and anxious-depressive symptoms, enhanced adherence to therapeutic measures, reduced compulsive behaviour, and diminished cognitive distortion. Additionally, there was a substantial increase in patients’ activity. This profile was more linear and closer to that of the general population, although somatic complaints and affective disturbances remained more common in our cohort than in the general population.

Eating disorders were less prevalent and exhibited milder severity within our cohort compared to findings from other studies, as indicated by lower presurgical scores on the EDE-Q global scale4,10,22. This difference can likely be attributed to our standard presurgical evaluation process, which systematically identifies and excludes severe eating disorders that could potentially compromise surgical outcomes37,38,39. We found no symptoms of compulsive intake or diagnostic criteria for binge eating disorder in BES, although restrictive eating behaviours and cognitive symptoms of concern for weight and body shape were common. Compulsive and obsessive behaviour and eating disturbances both entail easy frustration at having to wait for rewards; action in individuals with these traits is often geared toward achieving immediate satisfaction40 and avoiding situations that do not provide immediate reinforcement or that require perseverance to achieve rewards.

Following BS, notable enhancements were observed in various facets of psychopathology, particularly in somatic complaints, the anxious-depressive syndrome, and cognitive distortions. Additionally, there was a marked amelioration in restrictive eating behaviours and disturbances related to body image subsequent to BS. A previous study already reported an improvement in the BES score after surgery and a decline of the percentage of patients in the severe category from 78 to 5% (p < 0.01)21. Similarly, in our cohort, mean BES score was reduced in 2.79 points after surgery, although the score was in the category of not clinically relevant across the duration of the study. According with others22 the severity of problematic eating behaviours measured by EDE‐Q decreased up to 54.9% after surgery and remained lower than baseline throughout follow-up. Importantly, a majority of these symptoms (including compulsive eating, misuse of laxatives and diuretics, concern for weight and body shape) also exhibited improvement post-surgery, with patients’ post-surgical scores aligning closely with those observed in the general population. These findings underscore the potential for BS to effectively mitigate mild eating disorders, as well as alleviate mild psychopathological symptoms and maladaptive thought patterns.

Other findings41 support the notion that a significant portion of the improvements in mental health can be attributed to the weight loss itself, resulting in positive changes in body image, self-esteem, and self-concept. Nevertheless, it is important to acknowledge that other factors also contribute to postoperative mental well-being, including the patient’s perception of regained control over their life and the support received from healthcare professionals. Therefore, substantial weight loss together with the perception of the own personal progress and recovering the command over one’s life has the potential to foster an enhanced perception of body image, subsequently leading to a reduction in maladaptive symptoms, particularly in cases of mild mixed adjustment disorder.

Numerous studies have examined psychopathology, personality traits, and eating disorders within the bariatric population; however, few have delved into the intricate interconnections among these psychological, personality, and eating aspects among individuals with obesity who are pursuing BS42. The utilization of the PAI offers a comprehensive tool to illuminate both psychopathological tendencies and personality disorders, thereby pinpointing the pivotal psychological attributes that define these individuals. Furthermore, it aids in evaluating these attributes as potential predictors of surgical outcomes. Consequently, these psychological and personality variables warrant consideration for inclusion in psychotherapeutic interventions aimed at mitigating psychological distress and maladaptive eating behaviours. These findings underscore the significance of incorporating personality and psychological variables into the pre-surgery assessment process, as well as the potential utility of targeting them for psychotherapeutic interventions either prior to or following the surgical procedure.

Several limitations inherent to our study warrant consideration. Primarily, the modest sample size has likely curtailed the statistical robustness of our analyses, potentially compromising the sensitivity of certain investigations. Furthermore, the limited sample size impeded the execution of a comprehensive prognostic analysis and the incorporation of adjustments for bias and confounding variables. It is noteworthy, however, that a mere 18.75% of patients were lost to follow-up, a retention rate that proves favorable when contrasted with analogous obesity-focused investigations. Secondly, this longitudinal exploratory inquiry abstained from contrasting subjects with a control group (only with normative data), constituting a prospective analysis restricted to a solitary year of follow-up and bereft of supplementary data. As a result, any predictions drawn from the study’s findings must be approached with caution. It is essential for future investigations to encompass control groups and larger cohorts observed over an extended period of follow-up.

In conclusion, in our cohort, candidates for BS have somatic complaints (especially concern for their health), mild affective disorder (depression) and stress, restrictive eating behaviours, and cognitive symptoms of concern for weight and body shape, but they do not have a formal eating or personality disorder. Most of these symptoms improve after BS. Further studies are needed to evaluate whether improvements in mild intensity psychopathology benefits outcomes after BS.