Gastroesophageal reflux disease, functional dyspepsia and irritable bowel syndrome: common overlapping gastrointestinal disorders

GERD and FH

Typical symptoms of GERD, such as heartburn and regurgitation, are widespread in the community. The prevalence of GERD based on symptom perception in individual cross-sectional surveys varies from 2.5% to more than 25% [5,30]. GERD occurs less frequently in Asia than in Western countries [31,32].

The worldwide prevalence of GERD has increased during the past 15 years. GERD may be classified into two different forms: erosive reflux disease (ERD) and NERD; the latter represents the majority of reflux patients (up to 70%), who have typical symptoms without any esophageal mucosal lesion visible on endoscopy [33]. The severity of reflux esophagitis is usually classified according to the Los Angeles criteria (from A to D denoting increasing severity and extension of inflammation) [34].

The diagnosis of NERD should be supported by evidence that symptoms are due to acid reflux, on the basis of an excess of acid, or a positive correlation between symptoms and acid reflux using functional pH testing, or evidence of a symptomatic response to acid suppression [35,36]. Moreover, the use of 24-h MII-pH has shown that weak acidic reflux is also able to induce typical reflux symptoms, which, however, do not respond to PPIs [37-39].

Recently, the Rome IV criteria for functional esophageal disorders defined FH as retrosternal burning discomfort or pain refractory to optimal antisecretory therapy in the absence of GERD (i.e., normal acid exposure and no evidence of a symptom–reflux correlation), histopathologic mucosal abnormalities, major motor disorders or organic disease [40]. Notably, Savarino et al [16] suggested that, to be diagnosed with FH, patients should display a normal upper endoscopy, a normal AET in the distal esophagus and a negative symptom association with both acid and non-acid reflux by means of MII-pH.

IBS

IBS is a functional intestinal disorder characterized by recurrent abdominal pain, on average at least 1 day per week over a 3-month period, associated with two or more of the following criteria: the pain is related to defecation; the pain is associated with a change in frequency of stool and/or with a change in form (appearance) of stool [41]. IBS has traditionally been subcategorized into four subtypes based on predominant stool pattern: with constipation (IBS-C); with diarrhea (IBS-D); mixed (IBS-M) or unsubtyped (IBS-U) [42].

An estimated 5-25% of the people worldwide suffer from IBS. It is now recognized as the most common chronic bowel disorder [42,43]. Overlap of GERD and IBS symptoms ranges between 5% and 30% in community individuals [26,44,45], whereas the underlying mechanisms for this overlap and the cardinal symptoms and characteristics of GERD patients with IBS remain unclear. A detailed identification of prevalent symptoms is determinant of GERD.

FD

According to Rome IV criteria, the broad term FD comprises patients from the diagnostic categories of postprandial distress syndrome (PDS), characterized by meal-induced dyspeptic symptoms; epigastric pain syndrome (EPS), which refers to epigastric pain or epigastric burning that does not occur exclusively postprandially, can occur during fasting, and can even be improved by meal ingestion; and overlapping PDS and EPS, characterized by meal-induced dyspeptic symptoms and epigastric pain or burning. Moreover, the diagnostic criteria for FD specify that there should be no evidence of structural disease likely to explain the symptoms: i.e., upper endoscopy is normal [46]. Large-scale studies reported a 10-30% prevalence of FD worldwide, depending on the definition used and the geographical location [47].

IBS and GERD overlap: questionnaire-based diagnosis

In a recent review article, our group appraised and critically discussed studies supporting a possible overlap of GERD and FH with IBS. Notably, in studies where heartburn was investigated through only a symptom questionnaire, we found remarkable variability regarding the prevalence of GERD in patients with a diagnosis of IBS (Manning, Rome I, Rome II, Rome III criteria), with values ranging from 3-79% [48]. In 2012, Ford et al [45] conducted a systematic review and meta-analysis to estimate the prevalence of GERD-type symptoms in individuals with IBS and to quantify the overlap between the disorders. The authors selected population-based studies reporting the prevalence of IBS and GERD-type symptoms defined using symptom-based criteria or questionnaires. Overall, they selected 13 studies, in which the prevalence of GERD-type symptoms in IBS was 42%; the overall odds ratios of GERD-type symptoms in individuals with IBS was fourfold that of individuals without IBS. In particular, the positive association between GERD-type symptoms and IBS persisted in all geographical regions examined, and was consistent with each definition of IBS and gastroesophageal reflux symptoms, although the degree of overlap varied between 12% and 27%, depending on the diagnostic criteria used to define each condition [45].

In addition to the aforementioned meta-analysis, an Arab cross-sectional survey study found that 74.7% of IBS patients (198 individuals, Rome III criteria) had GERD symptoms at the same time (i.e., heartburn and/or acid regurgitation at least once per week for the last 3 months) [49]. More recently, Rasmussen et al [50] investigated the prevalence and overlap of these two conditions in a general Western population. In this nationwide study, 49,706 randomly selected individuals were eligible and underwent a web-based questionnaire survey, which showed the prevalences of GERD (Montreal definition) and IBS (Rome III criteria) to be 11.2% and 10.5%, respectively. Notably, the overlap between GERD and IBS was 56.9%.

IBS and GERD overlap: endoscopic diagnosis

Few data are available regarding the prevalence of IBS in patients with erosive esophagitis. The overall prevalence of IBS symptoms in the GERD population ranges from 10-74%, and it seems to be more frequent in patients with NERD than those with ERD. We focused our attention on some population-based studies in which GERD was detected using endoscopy. Few studies have reported pathophysiological evaluations to exclude patients with FH from the NERD population.

First, Smart et al [51] described an overlap with IBS symptoms (Manning criteria) in 9/22 (40.9%) patients with erosive esophagitis and/or esophageal complications. Neumann and coworkers [52] evaluated the prevalence of IBS symptoms (Rome III criteria) in patients with GERD classified by subgroup into NERD, ERD and Barrett's esophagus (BE). They observed that IBS was common in the entire spectrum of GERD: 63% in NERD, 44% in ERD and 21% in BE patients. The authors concluded that the presence of these disorders might explain why many patients with GERD have no complete symptom relief with PPI treatment. In 118 Mexican patients with IBS (Rome I criteria), Camacho et al [53] showed a general prevalence of IBS symptoms (48%) in patients with ERD and NERD. In a group of ERD (286) and NERD (74) patients from South Korea, Noh et al [54] described a prevalence of IBS (Rome III criteria) of 10.1%. However, the proportion of IBS-related symptoms was higher in NERD than in ERD patients (74.3% vs. 10.5%). In a different and more recent Korean study by Nam et al [55], who aimed to evaluate the association among IBS (Rome III criteria), NERD and ERD, a total of 2769 participants completed questionnaires and underwent upper and lower endoscopy. Psychological distress was also evaluated and was associated with both IBS and NERD, but not with ERD. Overall, the authors concluded that IBS was positively associated with NERD, but found no association with ERD [55].

In a case-control study from Taiwan, Hsu et al [56] evaluated the prevalence of IBS (Rome III criteria) in patients with and without esophageal-related symptoms (including ERD and NERD patients). The prevalence of IBS was at least 10% in patients with GERD and 5.5% in those without. The authors found that patients with GERD and IBS reported more frequent healthcare-seeking behavior, more severe GERD symptoms, poorer sleep quality, and greater depression than did those with IBS or GERD alone [56].

Monnikes et al [57] evaluated and treated a large number of patients with ERD (626 intention-to-treat; 500 per protocol analysis) with pantoprazole (up to 16 weeks). They described an overlap with IBS symptoms (Rome II/III criteria) in 13.6% of patients. At the end of the treatment period, they observed that all patients showed healed esophageal erosions but, more interestingly, that only 4.7% continued to report IBS symptoms. This percentage decreased to 2.8% after 6 months. They concluded that pantoprazole was effective in the treatment of patients suffering from signs and symptoms that suggested an overlap of GERD and IBS by providing a sustained response in the post-treatment period. After adjusting for many confounding factors, including psychological stress and Helicobacter pylori infection, Namm et al showed that IBS symptoms had a strong positive association with NERD, but no association with ERD [55].

IBS and GERD overlap: pathophysiological examination

The diagnosis of IBS is clinically based on the presence of characteristic symptoms, such as abdominal pain or discomfort and altered bowel habits. Lower endoscopy or hematological examinations are indicated only to exclude organic diseases, but these tests are not compulsory to correctly diagnose IBS [42]. Likewise, a GERD diagnosis could be clinically suspected and characterized by the presence of “troublesome” heartburn or regurgitation [4].

Within the spectrum of GERD, the pathophysiological relationship between ERD and NERD remains a subject of debate [58]. The classical and perhaps intuitive view, in which NERD is a mild form of GERD that might progress with time to ERD, is supported by some physiological, anatomical and histopathological findings [59]. However, NERD is a large umbrella that includes patients with reflux symptoms and negative upper endoscopy and this diagnosis is often made without any effort to distinguish FH (defined by the Rome III criteria) from NERD using pathophysiological investigations [60].

Currently, FH is a disorder characterized by symptoms of heartburn not related to GERD, diagnosed using 24-h impedance-pH testing, and it must be distinguished from NERD [16,33,61-63]. We previously described that the prevalence of GERD in the IBS population was 37.5%, with a large variability ranging from 11-79% [48]. First, Smart et al described abnormal reflux parameters using pH-metry in 11 (50%) of 22 patients with an IBS diagnosis (Manning criteria) [51]. Hershcovici et al [64] studied a high prevalence (49%) of IBS-related symptoms in a large cohort of patients with NERD (defined as abnormal 24-h pH-metry). Zimmerman et al described 256 patients with GERD, 50% of whom met the criteria for IBS. The authors concluded that in patients with NERD, the relationship between AET and symptoms was affected significantly, and in opposite directions, by smoking and IBS [65].

Using 24-h MII-pH, Martinucci et al [66] observed that IBS overlaps more frequently with FH (66%) than with ERD (17%) or NERD (39.5%). Recently, Garros et al [67] evaluated 168 patients with GERD symptoms classified into subgroups of responders (26) and non-responders (126) to PPI treatment. All patients underwent pathophysiological examinations. The authors concluded that IBS was more frequent in patients who did not respond to PPI treatment (27.2%) than in the responder group (7.7%). However, the authors did not report the prevalence of various subcategories of GERD in patients who showed the overlap with functional symptoms.

A recent meta-analysis showed that the odds ratio for GERD-related symptoms in individuals with IBS was fourfold that of individuals without IBS [45]. Accordingly, we can speculate that patients with GERD symptoms include subjects with either abnormal (NERD) or normal acid exposure, as well as FH patients. Savarino et al [33] found a greater prevalence of dyspeptic symptoms in patients with FH than in patients with NERD. The authors suggested that the data reinforce the concept that FGIDs extend beyond the boundaries suggested by the anatomical location of symptoms. This conclusion should be considered a further argument for testing patients with symptoms of GERD to separate patients with FH from patients with NERD [33]. In Fig. 1 we report the estimated overlap between GERD and IBS according to the Rome III and Rome IV definitions. More in-depth pathophysiological examinations are needed to detect patients with NERD and those with FH to better estimate the overlap between patients with heartburn and IBS.

Figure 1 (A) Different overlap between GERD and IBS according to Rome III (B) and Rome IV

GERD, gastroesophageal reflux disease; NERD, non-erosive reflux disease; HE, hypersensitive esophagus; IBS, irritable bowel syndrome.

FD and GERD overlap: questionnaire-based diagnosis

In 2011, a systematic literature review found that dyspeptic symptoms of epigastric pain, early satiety, bloating, nausea and vomiting were more common in patients with frequent GERD symptoms (range 21-63%) than in those with intermittent or no GERD symptoms [68]. A Korean population-based study, using the Rome III criteria, found an overlap between GERD and uninvestigated dyspepsia in 50% of GERD patients [69]. However, it should be noted that uninvestigated dyspepsia includes patients with an organic, systemic, or metabolic cause for the symptoms that could have been identified by traditional diagnostic procedures, which were not carried out.

A nationwide study of 100,000 individuals investigated the prevalence and overlap of GERD, FD and IBS in a Western general population by means of a web-based questionnaire survey. Among individuals fulfilling the criteria for GERD, FD and IBS, 69.3% fulfilled the criteria for only one of the conditions, 22.9% for two, while 7.7% fulfilled the criteria for all three. Among individuals meeting the criteria for one or more of the conditions, 30.7% fulfilled the criteria for two or all three conditions [28].

Interestingly, a study by Pleyer et al [70] found evidence supporting a systematic bias away from diagnosing FD, favoring a GERD diagnosis. Indeed, the authors assessed survey-based upper gastrointestinal symptom reporting between 1988 and 2009 and, though the majority of subjects (63%) reporting GERD symptoms received a GERD diagnosis, only a minority (13%) of those reporting FD symptoms received an appropriate FD diagnosis [70]. Such observer bias may be an important factor that hinders a correct assessment of the prevalence and overlap between these syndromes.

FD and GERD overlap: endoscopic diagnosis

In a study of 171 GERD patients, cases with symptoms compatible with FD, based on the Rome III diagnostic criteria and negative endoscopic findings, were assigned to the overlapping GERD-FD group (28% of enrolled patients). Patients with concomitant GERD and FD were more likely to be younger and female. Moreover, overlapping GERD and FD had the worst impact on the quality of life of the affected individuals, compared with GERD only and overlapping GERD and peptic ulcer disease [71].

FD and GERD overlap: pathophysiological examination

Abnormal pH monitoring was found in 23% of 247 patients with FD [72]. Subsequently, in a study by Savarino et al [33], 24-h ambulatory MII-pH data and FD questionnaires collected prospectively from a large group of NERD patients showed that clinically relevant dyspeptic symptoms were present in 44% of the NERD population. In particular, this study showed that when NERD patients were classified into three different subgroups on the basis of MII-pH results, dyspeptic symptoms were reported more frequently in patients with FH (63%), compared to those who had NERD with abnormal acid exposure time (41%) or hypersensitive esophagus (37%), suggesting a significantly different degree of overlap of NERD subgroups with FD [33]. Fig. 2 presents the estimated overlap between GERD and FD according to the Rome III and Rome IV definitions.

Figure 2 (A) Different overlap between GERD and FD according to Rome III (B) and Rome IV

GERD, gastroesophageal reflux disease; NERD, non-erosive reflux disease; HE, hypersensitive esophagus; FD, functional dyspepsia.

Pathophysiological similarities in GERD, FD and IBS

From a pathophysiological point of view, FGIDs are defined as gut–brain interaction disorders, characterized by visceral hypersensitivity, motility disturbance, altered mucosal and immune function, altered gut microbiota and altered central nervous system processing [41]. Notably, previous studies suggested that visceral hypersensitivity, minimal abnormalities of motility, and peripheral and central neural mechanisms could be considered the main common pathophysiological mechanisms among GERD and FGIDs [73,74].

It has been shown that patients with FGIDs exhibit a reduced pain/discomfort threshold during visceral stimulation [75]. Visceral hypersensitivity to various stimuli (chemical, distension, physical stimuli) has been demonstrated in patients with both IBS-related [76] and GERD-related [77] symptoms. Costantini et al [78] performed esophageal provocative testing (balloon distension and bethanechol) in IBS patients and healthy controls. They observed that esophageal symptoms appeared more frequently in patients with IBS, without any evidences of abnormal esophageal motility. Likewise, FD patients are also sensitive to acid, whether mediated through gastric acid “sensitivity” or impaired duodenal acid clearance [79,80].

Motor abnormalities are frequently reported as a common pathophysiological mechanism in both GERD and IBS [81]. Jones et al [82] speculated that the overlap between IBS and GERD could be explained by an unspecified dysfunction of smooth muscle across the gastrointestinal tract. Motor dysfunction of the upper stomach has also been described in patients with overlapping GERD and FD. Gonlachanvit et al [83] observed that gastric emptying was delayed in about 50% of patients with GERD, FD, and overlapping GERD and FD; proximal retention was greatest in patients with GERD and distal retention was a feature of both FD and the overlap groups. Moreover, van Lelyveld et al [84] showed that proximal gastric volume was smaller in FD and larger in GERD than in healthy control patients.

Hsu et al showed that GERD patients with more severe GERD symptoms, poor sleep quality, and greater depression are at higher risk of comorbid IBS. Likewise, IBS patients who reported poor sleep quality had a higher risk for GERD-related symptoms. The authors speculated that behavioral and psychological symptoms such as sleep disturbance and depression could be considered as risk factors to explain the overlap between GERD and IBS [85].

Psychosocial factors may also predispose individuals to the development of IBS [86]. Rubenstein et al [87] observed that patients with heartburn who reported increased levels of psychological distress and IBS-related symptoms recorded abnormal esophageal perception of both acid perfusion and balloon distension.

Effect of GERD treatment on FD and IBS

The majority of GERD patients are directly managed by general practitioners and receive PPI empirically, based on the fact that PPIs have a very low toxicity profile and good efficacy for suppressing gastric acid secretion and symptom relief. This approach is named “PPI trial” and it commonly involves a high dose of PPIs for at least 4 weeks [88]. A major limitation to note is its moderate sensitivity and specificity (78% and 54%, respectively) as a test to discriminate reflux-related manifestations [89]. However, the safety, efficacy and relatively low cost of these medications justify the use of PPIs as a first tool in a “step-down” approach for patients with GERD symptoms. Nevertheless, a “step-up” strategy is commonly adopted in patients with mild and intermittent symptoms, with the therapy strengthened every 2-4 weeks in case of lack of benefit. Regardless of the strategy undertaken, the management of GERD must include the avoidance of risk factors through lifestyle and dietary modifications [88,90]. Subsequently, when medical therapies are considered, acid secretion suppressors are considered the most effective in determining adequate symptom relief. There are also some reports that suggest an improvement in GERD-related symptoms after treatment with alginates [91,92].

Although there has been consistent progress in understanding the pathophysiology of GERD [11,21,23,38,62,93] and the several therapeutic options available, a considerable proportion of patients do not achieve a satisfactory therapeutic benefit. Therefore, the management of refractory GERD remains a disorder to be treated in specialized settings.

PPI therapy is undoubtedly associated with a superior healing rate and symptom control, as well as a lower rate of relapse compared with placebo, histamine-2 receptor antagonists, sucralfate, and prokinetics [94-97]. Although the acid content of the refluxate plays the main pathogenetic role in GERD, there are several factors that contribute to the disease, including alkaline reflux, pepsin, visceral hypersensitivity and psychological stress. The presence of concomitant functional disorders (IBS and FD) is, indeed, strictly related to a minor response to medical treatment. In fact, functional disorders, especially IBS, are more frequent in NERD patients than in ERD [33,66,98]. Translated into clinical terms, PPIs have greater benefit in ERD (70-80%) patients than in those with NERD (50-60%) [99-101]. However, the response rates vary according to the criteria employed to classify patients, the dosage administered and the duration of therapy. In fact, better discrimination among functional disorders, higher dosage and prolonged therapy are associated with a higher response rate to the treatment. The influence of functional disorders on treatment response seems to have been confirmed by Zerbib et al [102], who showed that the absence of esophagitis, the presence of FGIDs and a body mass index ≤25 are strongly associated with PPI failure in patients with GERD-related symptoms. Likewise, Garros et al [67] showed that the percentage of patients with FD and IBS was higher in non-responders to PPI than in responders (65.6% and 27.2%, respectively, vs. 38.5% and 7.7%). In this regard, Mönnikes et al showed that the presence of IBS-like symptoms reduced the response to PPIs in GERD patients, regardless of whether they had ERD or NERD [103].

It is worth noting that NERD is a term that includes three different types of patients characterized by various responses to the medical therapy: “true NERD”, hypersensitive esophagus and FH [12]. As proven in the meta-analysis by Weijenborg, if patients are diagnosed with NERD based on MII-pH (true NERD), the response to PPI treatment is as successful as in those affected by ERD [104]. In patients with NERD, there is no evidence to recommend maintenance treatment, even though approximately 70-75% of NERD patients relapse off therapy [105,106]. In this regard, some authors have suggested that an “on-demand” PPI treatment may be an alternative option [107,108]. Wu et al [109] showed, in a group of GERD patients (163 NERD and 102 ERD), that NERD patients have a higher failure rate of on-demand PPI therapy than do patients with ERD, especially if the former reported concomitant IBS symptoms.

A different subcategory of GERD that would benefit from PPI treatment includes patients with normal AET but a positive symptom–acid reflux relation (i.e., hypersensitive esophagus) [110]. The Rome IV consensus reconsidered patients with hypersensitive esophagus as functional patients.

Greater difficulties are encountered with NERD patients who do not have excessive AET and any relationship between symptoms and acid or weakly acidic refluxes. This subgroup of patients with FH is no longer classified under the term of GERD. Since these patients are typically poor responders to PPIs [35,110], other drugs aimed at reducing visceral sensitivity should be considered. To date, the medications that have been better investigated are tricyclic antidepressant agents, selective serotonin reuptake inhibitors and serotonin–norepinephrine reuptake inhibitors. Unfortunately, there are few placebo-controlled trials aimed at demonstrating the effectiveness of such compounds. Evidence from studies conducted on patients with IBS and functional dyspepsia have shown that there might be an overlap between FH and IBS [61,111,112]. Moreover, recent data obtained by Viazis et al [113] seem to support this argument. Their study demonstrated that citalopram, a selective serotonin reuptake inhibitor, provides greater benefit than placebo in controlling heartburn in hypersensitive patients. In contrast to FH, HE patients belong to the GERD population. However, IBS symptoms are often coexistent and the efficacy of PPI therapy is lower than in NERD and EE patients.

The relationship between IBS and GERD is under debate. One hypothesis suggests that IBS and GERD are two different pathologies that possess overlapping pathophysiologic processes [114,115]. Therefore, patients should receive two distinct treatments, each of which may affect the symptoms related to the other condition only minimally or not at all. Another hypothesis proposes that IBS is an extra-esophageal manifestation of GERD [116,117]. Several trials support this argument, reporting IBS symptom resolution in 20-40% of patients after PPI therapy [116,118]. Likewise, in a large-scale clinical study, lansoprazole effectively relieved dyspepsia in addition to reflux symptoms in patients with GERD [119]. Recently, another study showed that, after 4 weeks of treatment with lansoprazole, >60% of both patients with NERD and FD reported an improvement in reflux symptoms and in most dyspepsia symptoms [120]. In line with these results, Mönnikes et al observed improvement of FD and IBS symptoms after PPI therapy in GERD patients [57]. Surprisingly, there was no recrudescence of symptomatology after the treatment was suspended, and a sustained response with pantoprazole 40 mg was maintained up to 16 weeks after treatment [57].

Nevertheless, it should be kept in mind when prescribing PPIs that these medications might be directly involved in the genesis, or may at least favor (“two-hit” hypothesis) the onset, of IBS. Indeed, various mechanisms have been hypothesized: alteration in intestinal permeability [121,122] and alteration of the intestinal barrier function [123] and the composition of intestinal microbiota [124], which in turn may induce secondary changes in epithelial integrity [125].

Recently, a long period of treatment with PPIs was associated with increased risk of developing microscopic colitis [126], while one investigator speculated about a possible association between microscopic colitis and changes in intestinal permeability [121]. PPIs have been associated with some described alterations in intestinal barrier function [123] and subsequently with secondary important changes in intestinal microbiota composition [124]. It might be possible to speculate that these changes play a determining role in IBS pathogenesis.

Currently, scant data are available on the complex topic of overlapping FGIDs. MII-pH certainly contributes to tailoring reflux treatment by discriminating among different types of patients in the NERD realm, characterized by various responses to treatment [12,33,63]. Furthermore, a recent study showed that MII-pH could potentially be a helpful tool for predicting the response to PPIs in patients with heartburn [110]. However, further case-control trials, including studies with a placebo arm and conducted with accurate identification of upper and lower FGIDs, are needed.

Therapeutic options in IBS

The goal of IBS treatment is to provide relief of symptoms and improve overall quality of life. Because the symptoms of IBS are so wide-ranging, each patient must receive therapy individualized to achieve these goals. Currently, the ideal pharmacological therapy in patients with IBS would be one that completely addresses the entire IBS symptom complex, with minimal or no adverse side effects. Medications that modulate visceral hypersensitivity should do so without adversely affecting cognitive functions, somatic perception or gastrointestinal motility [127].

In the largest study to date of selective serotonin reuptake inhibitor therapy for IBS, 257 patients with severe IBS were randomized to paroxetine (20 mg/day), psychotherapy or usual care. After 12 weeks of treatment, the paroxetine group experienced a small but significant reduction in the number of days with abdominal pain compared to baseline (P=0.014) as well as improvements in their quality of life [128]. Another recent cross-over trial found that citalopram was more effective than placebo in improving abdominal pain, bloating and overall wellbeing in 23 non-depressed IBS patients referred to a tertiary care center in Belgium [129]. The mechanisms via which antidepressants have a beneficial effect on IBS are not clear. It is possible that these effects may reflect an improvement in the psychiatric profile of these patients, as many have anxiety and depression as a comorbidity.