Introduction

Chronic fatigue syndrome (CFS) is a chronic and disabling condition that closely resembles Beard’s description of neurasthenia (or nervous exhaustion) in the nineteenth century (Wessely, Hotopf, & Sharpe, 1998). Today, CFS is considered part of a larger group of functional somatic syndromes that are characterized by persistent somatic symptoms (e.g., musculoskeletal pain, fatigue, gastrointestinal problems), which cannot be fully explained by a medical or psychiatric condition (Wessely et al., 1998; Wessely, Nimnuan, & Sharpe, 1999). The syndrome is most commonly diagnosed in women between the ages of 40 and 49 (Boneva et al., 2011), and is usually associated with high personal costs (e.g., job loss, social isolation; Assefi, Coy, Uslan, Smith, & Buchwald, 2003). Prevalence rates of the illness remain unclear, but according to Reeves et al. (2007) are estimated at 2.5% of the adult population. The diagnosis of CFS is by exclusion: it is based on consensus criteria developed by experts and involves an extensive medical and psychiatric examination to exclude other disorders that could explain the symptoms. According to current diagnostic criteria, the core symptoms of CFS are unexplained and prolonged fatigue, and effort intolerance (i.e., post-exertional malaise or extreme exhaustion following physical or mental efforts; Carruthers et al., 2011; Fukuda et al., 1994). Additional symptoms include muscle pain, headaches, sore throat, tender lymph nodes, substantial impairment of short-term memory or concentration, and unrefreshing sleep (Fukuda et al., 1994). In most cases, the symptoms are triggered by acute physical and/or emotional stressors (e.g., after a viral infection or after a psychological loss experience; Van Houdenhove & Egle, 2004; Van Houdenhove, Luyten, & Kempke, 2013; see also below) and the illness onset gradually develops over months or years (Boneva et al., 2011; Van Houdenhove, Van Hoof, et al., 2009).

Several studies in CFS have shown that comorbidity is the rule rather than the exception (Van Houdenhove, Kempke, & Luyten, 2010). For instance, there is a high rate of comorbidity between CFS and other functional somatic syndromes, most notably fibromyalgia (FM) syndrome (i.e., chronic widespread pain in muscles and joints), leading researchers to propose that CFS and FM are part of one spectrum of pain and fatigue disorders that are interrelated through common etiopathogenic mechanisms (Ablin et al., 2012; Van Houdenhove et al., 2010, 2013). Moreover, CFS has been associated with elevated rates of both lifetime and current psychiatric disorders – particularly depression (Van Houdenhove et al., 2010; Van Houdenhove & Luyten, 2006). According to a population-based study by Nater et al. (2009), for instance, almost 60% of patients with CFS had at least one current psychiatric diagnosis.

Although the etiology of CFS is considered to be multifactorial (Van Houdenhove & Luyten, 2008), there is accumulating evidence that life stress – probably in combination with genetic factors – may play a role in the illness via disturbances in the stress response system (Nater, Maloney, Heim, & Reeves, 2011; Van Houdenhove et al., 2013). Specifically, recent evidence strongly suggests that at least in a subgroup of patients, CFS may reflect a fundamental dysregulation – or, more specifically, a loss of resilience – of the hypothalamic–pituitary–adrenal (HPA) axis, the main human stress response system (Van Houdenhove & Luyten, 2010; Van Houdenhove et al., 2013; Van Houdenhove, Van Den Eede, & Luyten, 2009). The HPA axis is responsible for the production of the stress hormone cortisol, which enables an individual to adequately adapt to various stressors (Claes & Nemeroff, 2005). In terms of McEwen’s allostatic load concept (i.e., the cumulative wear and tear on the body resulting from chronic stress; McEwen, 1998), CFS seems to be associated with an “allostatic crash” after a prolonged period of hyperactivity of the HPA axis due to chronic stress or overexertion, as evidenced by, for instance, lower cortisol activity, enhanced negative feedback, and blunted cortisol reactivity to stress, compared to non-fatigued controls (Cleare, 2003; Papadopoulos & Cleare, 2011; Tak et al., 2011; Tomas, Newton, & Watson, 2013; Van Houdenhove, Van Den Eede, et al., 2009). This may explain why patients with CFS do not adequately respond to and recover from mental and/or physical stressors (“effort intolerance”), which is one of the most important diagnostic characteristics of the illness (Van Houdenhove & Luyten, 2010; Van Houdenhove et al., 2013). Importantly, dysregulation of the HPA axis has been shown to affect various other biological systems, such as neurotransmitter systems, the immune system as well as pain regulatory systems (Van Houdenhove & Egle, 2004). For instance, CFS has been associated with increased immune activity, generating typical post-exertional flu-like malaise and “sickness-behavior” (Silverman, Heim, Nater, Marques, & Sternberg, 2010).

From a psychological point of view, there is increasing evidence that in at least a subgroup of patients with CFS, a strong achievement orientation, a self-critical attitude, and a perfectionistic lifestyle are key factors contributing to chronic stress/overload and the ensuing neurobiological alterations that underlie the illness (Kempke et al., 2013a; Kempke, Luyten, Mayes, Van Houdenhove, & Claes, in press; Luyten et al., 2011; Van Houdenhove et al., 2013). Moreover, we and others have argued that excessive achievement strivings and self-critical perfectionism may reflect disruptions in normal personality development, i.e., be compensatory strategies to defend against negative self-feelings and to prove one’s self-worth (Kempke et al., 2013a; Van Houdenhove, Neerinckx, Onghena, Lysens, & Vertommen, 2001). Indeed, behind the image of self-sufficiency and resilience (whereby patients present themselves in relation to others as self-reliant and strong) often lies extreme vulnerability, particularly due to unresolved trauma. In this chapter, we discuss findings showing that a lack of integration of (opposing) self-experiences or mental representations may contribute to a dysregulation of the central stress mechanisms in CFS. We take a dialectic perspective on the emergence and experience of the self in health and disease, rooted in Sidney Blatt’s two-configurations theory of normal and pathological development (e.g., Blatt, 2004). First, we give an overview of theory and research concerning the role of self-critical perfectionism (SCP) and its correlates in CFS. Next, we discuss the role of attachment insecurities in CFS within the context of empirical research on early adverse experiences. Finally, we consider the implications of the above-mentioned theory and research for treatment programs for CFS.

Case example

Catherine, a 41-year-old secretary, has suffered for three years from severe mental and physical fatigue combined with widespread muscle pain. Lately, she feels weak and tired after minimal effort (e.g., reading a book or going for a walk), and it seems she will not be able to continue her full-time job. She has been through a very stressful period – a difficult and painful divorce from her husband. After a comprehensive examination at the local hospital, she was diagnosed with CFS because of unexplained fatigue lasting for more than six months. She has always been very critical and demanding of herself, pushing herself beyond her limits and ignoring bodily signals (e.g., recurrent throat infections, unrefreshing sleep). She wanted to do everything perfectly, both in her job as a secretary but also at home. Last year she was awarded as the best secretary of the company; now she cannot even do simple household activities. She felt she had failed both as a wife and as a secretary. She says she does not want any help, especially not from her mother. The relationship with her mother has always been problematic; Catherine described her mother as very critical, pragmatic, and rather unemotional. Catherine was named after her mother’s first daughter who passed away after birth. She always had the feeling of being unwanted.

Self-critical perfectionism and overload in CFS

As illustrated by the story of Catherine, many patients with CFS are characterized by high achievement orientation and perfectionistic personality traits (Kempke et al., 2013a). Following Blatt’s theory, maladaptive personality development in CFS may reflect a disruption of the normal dialectical interaction between two fundamental developmental lines, i.e., an anaclitic, relatedness or attachment line that normally leads to increasingly mature, complex, and mutually satisfying interpersonal relations, on the one hand, and an introjective or self-definitional line that normally leads to the development of a stable, realistic, and essentially positive self and identity on the other hand (Luyten & Blatt, 2013; Luyten et al., 2011). Disruption in this dialectic, however, leads to an overemphasis on or exaggeration of one developmental line to the neglect of the other. Self-critical perfectionism (SCP), which shows similarities with Beck’s description of “autonomy” and the concept of attachment avoidance (Blatt & Levy, 2003; Luyten, Blatt, & Corveleyn, 2005; Shahar, 2001; Sibley, 2007), thus includes an overemphasis on the self-definition line and a defense against dependency needs – also termed “counterdependency” (Gregory, Manring, & Wade, 2005). This personality configuration is characterized by a harsh ego ideal leading to overly critical self-evaluations and feelings of guilt when not meeting standards (Blatt, 2004; Luyten et al., 2005). Hence, self-representations in patients with CFS are often fragmented and extremely dependent on the approval of others. In order to protect the self from “negative introjects” – often experienced as a threat to the self – and to prove one’s self-worth, self-critical patients with CFS primarily use higher order (counteractive) defense mechanisms, in particular reaction formation and overcompensation in terms of mental and/or physical overexertion, leading to a “false” self that is self-reliant and strong (Cuykx, Van Houdenhove, & Neerinckx, 1998; Van Houdenhove, Neerinckx, Onghena, et al., 2001). In the same vein, CFS has been associated with excessive self-sacrificing tendencies (e.g., compulsive caregiving), reflecting the underlying search for acceptance and recognition (Van Houdenhove, Neerinckx, Lysens, et al., 2001). This is reminiscent of Blumer and Heilbronn’s (1982, p. 387) classical description of the core issues of the pain-prone disorder in terms of “strong needs to be accepted and to depend on others, as well as marked needs to receive affection and to be cared for.”

Empirical research has supported the important role of overcompensation in the development and maintenance of CFS. In an interesting qualitative study, Wentz and colleagues (Wentz, 2005; Wentz, Lindberg, & Hallberg, 2004) identified an unprotected self in fibromyalgia (FM) patients, reflecting high levels of stress during childhood (due to lack of protection in relation to stimuli and affects) and emotion regulation difficulties in adulthood, promoting the use of so-called compensating strategies, such as “hypomanic repair” (i.e., meeting the needs of others to an extreme extent), conflict handling through action, and demonstrating strength/being in control. According to Wentz (2005), extreme helpfulness to other people and intense activity has become a way for these individuals to compensate for psychological vulnerability. In the same vein, Pemberton and Cox (2014), using semi-structured interviews, reported a pre-morbid state of constant motion among patients with CFS and a tendency towards negative feelings and beliefs about “doing nothing.” Similar findings have been reported in a population-based study consisting of 4779 participants who were followed for the first 53 years of their life (Harvey, Wadsworth, Wessely, & Hotopf, 2008). Results of this study showed that increased pre-morbid levels of exercise were associated with an increased risk of developing CFS.

Moreover, both cross-section and longitudinal studies suggest that SCP is implicated in CFS (Dittner, Rimes, & Thorpe, 2011; Kempke et al., 2013a; Kempke, Van Houdenhove, et al., 2011; Magnusson, Nias, & White, 1996; Moss-Morris, Spence, & Hou, 2011). For instance, Kempke, Van Houdenhove, et al. (2011), in an attempt to replicate previous findings on the nature of perfectionism (Dunkley, Blankstein, Masheb, & Grilo, 2006), demonstrated a distinction between adaptive and maladaptive (i.e., concern over mistakes, doubts about actions) perfectionism in CFS in that only the maladaptive (or self-critical) component was significantly related to fatigue. More recently, we showed that SCP predicted increased levels of fatigue and pain over the 14-day period and this was independent of the severity of depression, and was probably mediated by its impact on stress and activity levels, as discussed in greater detail below (Kempke et al., 2013a). Using a prospective design, Moss-Morris et al. (2011) showed that maladaptive perfectionism predicted the onset of CFS in patients suffering from glandular fever. Recently, we investigated whether SCP in CFS may reflect disruptions in normal personality development, i.e., be a compensatory strategy to defend against negative self-feelings, or be associated with broader personality disturbances. In a controlled study, we demonstrated elevated levels of maladaptive personality features (notably obsessive–compulsive personality disorder features) that are related to negative perfectionism, but found no evidence for increased prevalence rates of personality disorders (16.3%) in patients with CFS (Kempke, Van Den Eede, et al., 2013). Finally, a number of studies have demonstrated higher levels of maladaptive or unhealthy perfectionism in CFS as compared to normal controls and non-fatigued patients (Deary & Chalder, 2010; Sirois & Molnar, 2014; White & Schweitzer, 2000). Of particular interest is a study by Luyten, Van Houdenhove, Cosyns, and Van den Broeck (2006) showing that patients with CFS retrospectively reported both higher pre- and post-morbid levels of perfectionism compared to normal controls. Most recently, Sirois and Molnar (2014) found higher levels of maladaptive perfectionism among patients with CFS compared to healthy controls and patients with irritable bowel syndrome or fibromyalgia/arthritis. It is important to note here that although SCP is more closely related to exhaustion and chronic fatigue, studies have also demonstrated a link between high self-criticism and chronic pain (Lerman, Rudich, & Shahar, 2010; Lerman, Shahar, & Rudich, 2011; Rudich, Lerman, Gurevich, Weksler, & Shahar, 2008).

Several mechanisms may explain the observed relationship between SCP and chronic fatigue. SCP may increase stress sensitivity and lead to the “active” generation of stress (i.e., experiences of failure because of having unrealistic high expectations) as well as the “degeneration” of protective factors (e.g., positive affect, social support), resulting in an exacerbation of symptoms (Blatt & Shahar, 2005; Luyten & Blatt, 2013; Luyten, Blatt, Van Houdenhove, & Corveleyn, 2006; Shahar, 2001). Recently, our group replicated these findings in patients with CFS by showing that SCP was associated with increased stress sensitivity and more daily hassles, which in turn predicted increased levels of depression (Luyten et al., 2011). This fits with our previous findings showing that maladaptive perfectionism, because of high self-criticism, may give rise to negative self-perceptions (i.e., low self-esteem), which, in turn, increase the chance of depression and even more self-criticism (Kempke, Luyten, et al., 2011; Kempke, Van Houdenhove, et al., 2011).

As noted above, we and others have hypothesized that SCP may be associated with hypofunctioning of the HPA axis in CFS because of the “wear and tear” on the body due to chronic or repeated high stress levels (Kempke, Luyten, De Coninck et al., 2015; Kempke, Luyten, Mayes et al., in press; Luyten et al., 2011). In line with this reasoning, we recently demonstrated that SCP was related to increased levels of subjective (self-reported) stress, but with decreased (blunted) HPA axis reactivity (i.e., lower cortisol responses) following acute experimental stress induced by the Trier Social Stress Test (Kempke, Luyten, Mayes et al., in press). These findings provide preliminary evidence that SCP in CFS may be associated with loss of “adaptability” of the neurobiological stress system (Van Houdenhove et al., 2013).

Further, SCP has been related to specific maladaptive cognitions and behaviors – or dominant coping strategies – to compensate for feelings of inferiority and low self-esteem, including high “action-proneness” (i.e., a tendency toward direct action leading to periodic overactivity; Van Houdenhove, Bruyninckx, & Luyten, 2006), high “persistence” (i.e., a tendency to persevere despite frustration and fatigue; Fukuda et al., 2010; Van Campen et al., 2009), and “all-or-nothing” behavior (Moss-Morris et al., 2010), leading to physical and/or mental overexertion and lack of acceptance of functional limitations (Brooks, Rimes, & Chalder, 2011; Van Houdenhove, Neerinckx, Onghena, et al., 2001). Importantly, the tendency towards overactivity often persists after illness onset, as illustrated by periodic “outbursts of activity” when feeling better followed by complete rest/inactivity (a “boom-and-bust” activity pattern), which may perpetuate symptoms and impede recovery (Kempke et al., 2013a; Luyten et al., 2011; Moss-Morris et al., 2010).

Driver (2005), in her analysis of a young woman suffering from chronic fatigue, argued that the swings between overactivity and underactivity in CFS may reflect an underlying “fight–flight” response demonstrating a fundamental inability to regulate internal states. Indeed, patients with CFS, and highly self-critical patients in particular, have been shown to have impaired emotion regulation strategies, such as with respect to the suppression of emotions, emotional avoidance, and the tendency to express negative beliefs about emotions (e.g., emotions are a sign of “weakness,” “others will negatively react to displays of emotions”; Hambrook et al., 2011; Oldershaw et al., 2011; Rimes & Chalder, 2010; see also Van Middendorp et al., 2008), thereby increasing stress sensitivity and overload. Interestingly, Nemiah and Sifneos, in the early 1970s, coined the term “alexithymia” to describe a similar personality style in psychosomatic patients characterized by difficulty describing feelings and distinguishing them from bodily sensations (Sifneos, 1973; Taylor, Bagby, & Parker, 1991). A few years earlier, theorists from the Paris School of Psychosomatics introduced the notion of operational or concrete thinking (pensée opératoire), which is closely related to the concept of alexithymia (Aisenstein, 2006). Taerk and Gnam (1994, p. 320), in turn, from an object-relations perspective, stated that CFS vulnerability results from “poorly developed capacity for regulating internal states in response to certain types of stressors, namely disturbances in object relations.”

More recently, Luyten and colleagues (Luyten, van Houdenhove, Lemma, Target, & Fonagy, 2012; Luyten, Van Houdenhove, Lemma, Target, & Fonagy, 2013) have argued that CFS is related to impaired mentalization (i.e., the metacognitive ability to reflect on internal processes) and, in particular, dysfunctions in “embodied mentalization” – referring to difficulties in identifying and transforming bodily sensations into mental states (i.e., the “embodied” self). This may explain patients’ tendencies to ignore affective signals of being stressed or overburdened. Of particular interest in this regard is a study by Oldershaw et al. (2011), showing that CFS is associated with difficulties in emotion recognition and inferring own emotions. Comparable findings have been reported in somatoform disorders. Subic-Wrana and colleagues (Subic-Wrana, Beutel, Knebel, & Lane, 2010; Subic-Wrana, Bruder, Thomas, Lane, & Kohle, 2005) showed decreased emotional awareness and deficits in emotion recognition in these patients.

Developmentally, impairments in mentalization have been linked to a history of early adversity and insecure attachment (Bateman & Fonagy, 2006), as discussed later in this chapter. Indeed, impaired mentalization has been implicated in trauma-related psychopathology (e.g., borderline disorders) (Bateman & Fonagy, 2006). Yet, impairments in mentalization in CFS might also be exacerbated by the illness itself, because of being chronically invalidated and feeling dismissed or not understood. Moreover, the overwhelming fatigue and pain may be felt like an attack on the self and is often experienced as a “bad” internal object threatening one’s identity, leading to a dissociation between the (suffering) body and the self (Osborn & Smith, 2006; Schattner, Shahar, & Abu-Shakra, 2008; Smith & Osborn, 2007; Van Damme & Kindermans, 2015). In this regard, Sachs (2001) has argued that the (medical) diagnosis of CFS may transform the patient’s lived (subjective) body into a medicalized (real) body.

Childhood adversity and attachment insecurity in CFS

In this section, we will show that the premorbid overactive lifestyle and “counterdependency” observed in many patients with CFS may reflect attempts to compensate for low self-esteem and narcissistic injuries associated with early negative experiences and disruptive attachment experiences (Van Houdenhove, 2005; Van Houdenhove, Neerinckx, Lysens, et al., 2001). As Van Houdenhove and colleagues (2001, p. 574) have noticed, “particularly those with childhood victimisation experiences often show a tendency to exceed physical limits (in work or sports) as a way of coping, i.e., to maintain self-esteem, stabilise the affective equilibrium and prevent anxiety and depression.”

Indeed, a number of studies have demonstrated elevated rates of adverse childhood experiences in patients with CFS, with a considerable subgroup of patients reporting multiple trauma (for an overview, see Kempke et al., 2013b), congruent with the notion of an “unprotected” self in chronic pain and fatigue conditions (Wentz et al., 2004, 2005). These findings are also in keeping with more basic research showing that a history of abuse and maltreatment, especially during critical periods of development, may lead to increased stress sensitivity and impairments in social cognition, thereby increasing the risk for stress-related disorders in genetically susceptible individuals (Heim, Ehlert, & Hellhammer, 2000; Lupien, McEwen, Gunnar, & Heim, 2009).

More than a decade ago, Van Houdenhove et al. (2001) reported for the first time elevated levels of early adversity in a mixed sample of CFS and FM patients (i.e., 64.1% experienced adversity) as compared to controls. Interestingly, emotional neglect (48.4%) appeared to be one of the most important trauma domains among CFS/FM patients. Moreover, patients in this study were more likely to experience re-victimization in later life (e.g., living in an abusive relationship), which might be related to impaired mentalization capacities within the context of attachment relationships. These findings have been replicated in a number of other studies in CFS. For instance, Heim et al. (2006, 2009), in a population-based study in the USA, demonstrated that early childhood trauma was associated with a three- to eightfold increased risk for CFS. In keeping with these findings, in a sample of 90 well-screened patients with CFS, Kempke et al. (2013b) showed that more than half of the patients (54.4%) had experienced at least one type of early childhood trauma. Importantly, prevalence rates were particularly high for emotional trauma (i.e., emotional abuse and/or emotional neglect: 46.7%). Furthermore, we found that early adversity directly influenced the course of CFS in terms of levels of pain and fatigue over a period of 14 days, probably due to its impact on stress sensitivity, stress generation, and interpersonal distress in particular, as well as on maladaptive cognitive-perceptual factors such as catastrophizing and somatic hypervigilance (Luyten et al., 2011; Van Houdenhove & Luyten, 2008; Van Houdenhove et al., 2013; Van Houdenhove, Neerinckx, Lysens, et al., 2001).

Although early adversity has been implicated in a wide range of functional somatic syndromes, including FM and irritable bowel syndrome (Afari et al., 2014), the findings seem to converge that CFS is more closely linked to “emotional” adverse experiences rather than more “objective” or physical types of trauma. This fits with the trauma studies reviewed above, and suggests that emotional disturbances and dysfunctions in embodied mentalization may be particularly prominent in patients with CFS (Luyten et al., 2012, 2013). In fact, emotional childhood adversities in CFS can include a wide range of negative experiences, such as early parentification, being abandoned, or growing up in an emotionally unresponsive or highly critical environment (Van Houdenhove, Neerinckx, Lysens, et al., 2001; Van Houdenhove, Neerinckx, Onghena, et al., 2001). According to Wentz et al. (2004; Wentz, 2005), many patients prone to CFS/FM were overstrained as a child; that is, unsupported or overexposed to mental load.

Turning to attachment insecurity, childhood adversity may have a profound effect on the development of mental representations or internal working models (IWM) of the self in relation to significant others (Mikulincer & Shaver, 2007). Specifically, CFS has been associated with insecure attachment or the use of so-called “secondary” attachment strategies to seek an inner sense of security, most notably dismissive–avoidant attachment strategies; this is further reinforced by the current scientific controversy and even overt societal disbelief surrounding CFS (Luyten et al., 2012, 2013). As a result, these patients often lack a basic sense of trust in others and tend to deactivate the attachment system (i.e., denial of attachment needs, self-reliance and independence) to regulate affective states by distancing themselves from others, leading to isolation and loneliness, thereby decreasing the stress-buffering effects of social support (Luyten et al., 2013). Moreover, it is recognized that patients who rely on dismissive attachment strategies are prone to alexithymia, less emotional awareness, and, most importantly, “deactivating” emotion regulation strategies (e.g., suppression of negative emotions), leading to increased levels of stress and “allostatic load” in the long run (Mikulincer & Shaver, 2007). Also, these patients tend to deny the impact of psychosocial factors (e.g., life stress, early attachment relationships) on the course of their illness and typically do not seek help for their problems unless there is a “medical” cure (Luyten et al., 2013). Indeed, as described by Muller (2009, 2010), deactivation reflects “minimization” of painful attachment-related experiences. It should be noted, however, that not all CFS (or FM) patients are characterized by deactivating attachment strategies. Some patients may have a predominant hyperactivating (preoccupied) attachment pattern (e.g., these patients try to elicit support through clinging and claiming behavior, leading to hostility in others and subsequent rejection), while others may show a combination of hyperactivating and deactivating attachment strategies (Luyten et al., 2013). Moreover, not all behavior is “deactivated” in avoidant attached patients (Muller, 2010). In fact, only intimacy is avoided, because the attachment figure was often the source of distress and the patient has learnt that it is safer to be self-reliant (Muller, 2010). Although empirical research on attachment in patients with CFS is largely lacking, findings in somatoform and chronic pain patients have shown a higher (relative) incidence of insecure-dismissive attachment, unresolved states of mind, and narrative incoherence – or the inability to construct a coherent self-narrative (Maunder & Hunter, 2008; Neumann, Nowacki, Roland, & Kruse, 2011; Pedrosa Gil, Scheidt, Hoeger, & Nickel, 2008; Subic-Wrana et al., 2005; Waller & Scheidt, 2006; Waller, Scheidt, & Hartmann, 2004).