The term “gastroparesis” implies a partial or complete paralysis of the stomach as a result of muscle or nerve dysfunction. As this book will attest, gastroparesis is far more complex than this simple definition. Gastric emptying time may be normal and yet the patient may respond to the same treatments that work for other patients with study proven delayed gastric emptying. Is gastroparesis caused by dysbiosis, is it an autonomic neuropathic disorder, an autoimmune disorder, a problem of behavioral conditioning, or a central nervous system disorder? Patients may experience overlapping symptoms of nausea, vomiting, distension, abdominal pain, dysmotility, weight loss, early satiety, reflux, and others. This may generate a complex differential diagnosis that includes other conditions such as rumination, functional dyspepsia, somatization, eating disorder, and even Munchausen’s syndrome, to name but a few. Psychiatric conditions can occur as a result of gastroparesis, may be a causative or exacerbating factor in gastroparesis, and may confound the diagnosis and treatment of gastroparesis ( Fig. 27.1 ).
The most common co-morbidities with gastroparesis described in the literature include depression, anxiety, somatization, hypochondriasis, eating disorders, opiate dependence, and personality disorders. We will briefly review the major psychiatric issues we have encountered in collaboratively managing these patients, discuss treatment of those conditions, and describe the amplifying loop of the interaction between psychiatric conditions and gastroparesis.
Psychiatric disorders: A framework for discussion
Psychiatric diagnosis has undergone faddish unifying doctrines since the first descriptions of mental illness in medicine. From imbalances in humors to conflicted unconscious drives, each has sought to explain all disorders of mental life during the time it ruled the psychiatric community. Despite this, a small group of physician scholars worked out some fundamental understandings of the psychiatric disorders they studied. These were well summarized in the book “Perspectives of Psychiatry” by Paul McHugh and Philip Slavney published in 1983 but based on earlier work by Karl Jaspers and Sir Aubrey Lewis. These authors organize psychiatric conditions from the perspectives of disease, behavioral problems, problems of dimensions, and problems of life experience. Psychiatric diseases are simply diseases that affect mental life through a pathological mechanism, or in McHugh’s words, a broken part. We will discuss major depression and bipolar disorder as two of these conditions. Behavioral disorders are based on learned responses to environmental exposures, such as alcoholism and eating disorders. These problems involve “conditioning” or automatic triggering of behavioral cascades. It is important to note that conditioning occurs in animals with no “psyche” and is a biological process. Conditioned anaphylaxis can be a fatal problem and has been demonstrated in animal models. Problems of dimensions are problems associated with excessive endowments of traits that are adaptive in one set of circumstances but are maladaptive in another circumstances, such as a rich endowment of emotions that may overwhelm a logical approach in favor of an emotional one. Patients who are excessively endowed with emotions often frustrate clinicians because of their capricious or unpredictable behavior. Lastly, we will discuss the experiences one encounters in life and how these experiences may shape the individual to see meaning in things or events. These interpretations by patients may not be helpful in understanding the illness they face. All of these disordering types of conditions influence our patients and must be considered in the evaluation of patients with gastroparesis and should be understood as part of routine medical care.
Major depression and gastroparesis
Affective disorders (major depression and bipolar disorder) are disorders of mood, motivation, drive, energy, focus, and especially reward. The bulk of our discussion will be on Major Depression, a disorder with constantly changing appellations. The DSM 5 emphasizes depressed mood or loss of interest or loss of pleasure as well as the episodic nature of the condition. The criteria currently used are:
Depressed mood most of the day, nearly every day.
Markedly diminished interest or pleasure in all, or almost all, activities most of the day, nearly every day.
Significant weight loss when not dieting or weight gain, or decrease or increase in appetite nearly every day.
A slowing down of thought and a reduction of physical movement (observable by others, not merely subjective feelings of restlessness or being slowed down).
Fatigue or loss of energy nearly every day.
Feelings of worthlessness or excessive or inappropriate guilt nearly every day.
Diminished ability to think or concentrate, or indecisiveness, nearly every day.
Recurrent thoughts of death, recurrent suicidal ideation without a specific plan, or a suicide attempt or a specific plan for committing suicide.
We subscribe to the view that the fundamental characteristic of Major Depression is loss of the ability to experience reward, a condition that many of us in the field believe involves dysregulation of dopamine neurotransmission. We will use the term depression even though many patients are not “depressed”. They may be depressed, irritable, flat or anxious or any combination of these, depending on the kind of person that has the disorder. The unifying element is that patients endorse a loss of reward, pleasure, or interest in activities that they characteristically found rewarding. The condition is usually episodic with a return to baseline between episodes.
Major depression must be differentiated from demoralization, a term that describes the understandable psychological reaction to loss that is a nearly universal experience. The now abandoned terms ‘endogenous’ and ‘reactive’ depression emphasized that these states were both associated with depressed mood and sadness. Major depression is most likely a disease state of dopaminergic and other neurochemical derangement, while demoralization is an understandable psychological state of adaptation to loss or adversity. Some of the distinguishing features are shown in Fig. 27.2 . As noted, the distinction between these conditions is further confounded by the fact that patients can have both conditions at the same time, and in fact episodes of Major Depression are often triggered by psychological loss and stress as well as by metabolic stresses associated with disease states.
Another confound is that the term depression is used to describe the symptom as well as the disorder of Major Depression. Measuring symptoms of depression in disease states and then using this as a surrogate for the diagnosis of Major Depression using symptom checklists and severity cutoffs is fraught with problems. A person can have Major Depression of the disease type in the absence of any medical or psychological stressor. They can have mild major depression, following the characteristics described above, and they can appear depressed when they are devastated by grief that is severe and long lasting (usually in proportion to the loss). They can also have severe symptoms of depression due to demoralization in the context of an illness, or even a disrupted response to dopamine neurotransmission caused by medication or metabolic derangement. The relationships of these states to the idiopathic, substantially genetic, episodic disease of Major Depression which occurs in otherwise entirely medically healthy people is still unclear. Major Depression is more common and more chronic in conditions associated with inflammation, such as autoimmune disorders, chronic infections, or neurodegenerative conditions. Inflammation within the CNS is particularly associated (for example Major Depression is more common in MS than in ALS) and is seen at much higher rates in patients with HIV and Hepatitis C. It is clearly associated with stress, possibly generated by the chronic activation of the hypothalamic-pituitary-adrenal stress axis. Conditions such as heart disease and cerebrovascular disease are shown to have poorer outcomes when comorbid with depression, a finding that may be due to the association of these conditions with chronic inflammation as well.
There is a clear association between Major Depression and gastroparesis. Although the studies used rating scales for depressive symptoms to estimate prevalence of depression, there is an elevated prevalence across all studies with rates varying from 20% to 40% of gastroparesis patients . One study looking at a small series of post-surgical gastroparesis patients, found more than 50% had depression , but some of these patients had undergone surgery for pancreatic and gastric cancer, and pancreatic cancer is associated with very high rates of depression. There is no good longitudinal data to parse the temporal relationships between the development of depression and gastroparesis, but both conditions are associated with dysmotility, dysautonomia, and chronic inflammation. In other conditions the arrow seem to go both ways. HIV infection is associated with the development of depression as it progresses, but depression is also a risk factor for acquisition of the virus.
Depression has been associated with the severity of gastroparesis symptoms , however severity of gastroparesis symptoms often does not correlate with the severity of delay in gastric emptying . When looking at specific conditions such as Type II diabetes, the severity of diabetes often does not correlate with the severity of delay in gastric emptying . Depression is associated with more severe symptoms of gastroparesis, such as abdominal pain , but interestingly depression was also associated with the degree of delay of emptying . The severity of symptoms correlates with depression in the related conditions of reflux and functional dyspepsia . In one study, the severity of depression was a factor that correlated with the willingness of survey participants to take high-risk medications, suggesting that this was the most distressing symptom . In a longitudinal study of gastroparesis, only a modest number of patients had improvement (28%), and depression had a negative correlation with the likelihood of improvement, while use of antidepressants correlated with improvement. Use of analgesic medications did not correlate with improvement, and anxiolytics actually correlated with decreased likelihood of improvement .
While a comprehensive discussion of the treatment of depression is beyond the scope of this discussion, there are several issues regarding antidepressants that are important to bear in mind while treating gastroparesis patients for depression. The first and most important is to use adequate doses. Higher doses are often associated with better responses. SSRI’s (fluoxetine, sertraline, citalopram, escitalopram, and paroxetine) are the mainstay of treatment for depression in primary care settings. These drugs are useful for anxiety, which is a symptom experienced by many patients. While all of the pro-serotonergic agents are somewhat GI activating, Sertraline, in our experience, has had the most positive impact on GI emptying. Having said this, it is important to note that many patients do not improve when gastric emptying is the main target, and indeed, many have no demonstrable delays in emptying. In a study of POTS, (Postural orthostatic tachycardia syndrome), in patients with prominent GI symptoms, the most common finding on gastric emptying studies was rapid emptying, which was associated with deconditioning and depression . We have had considerable success with using aggressive medical trials with antidepressants that have widely varying properties.
SNRI antidepressants have noradrenergic activity that can be useful. They are also useful for anxiety. They are more effective for neuropathic pain, and have autonomic effects that may account for some of their utility in sympathetically maintained pain. They are useful for migraine headaches. Of these agents, we have found venlafaxine to be the most GI activating (as noted above this may be helpful or problematic depending on the patient). Venlafaxine has a short half-life and should be tapered as it can produce an unpleasant discontinuation syndrome.
Mirtazapine is an atypical antidepressant that has uses for chronic pain, is relatively less GI stimulating, is great for sleep and is useful for anxiety. It is also the most appetite stimulating antidepressant, a useful property in some patients. It is well tolerated if weight gain is desired.
Tricyclic antidepressants are often considered expert-level drugs, but almost every patient we see has been prescribed one during their disease course. These drugs are most effective when blood levels are measured and used to guide dosing decisions. We often use nortriptyline, which has a tight window of therapeutic effect between 100 ng/dL and 150 ng/dL. This is not a toxicity issue, (until at higher levels), it is an unusual property of this drug that rather than reaching a plateau at 150 mg/dL, effectiveness actually declines at higher blood levels. This does not appear to be the case for other tricyclic antidepressants. There is great heterogeneity in metabolism of this drug and therefore it must be used with caution, but it is often effective when other medications have failed.
There are numerous other medications and combination treatments that may be helpful, and the important issue is to continue to try if the first efforts fail. It is also worth noting that at this time the patients will accuse you of thinking that their illness is “all in their head” rather than “in their stomach where it belongs”. This is a good time to discuss the “gut brain” and the concept of a control system for stomachs. We often refer to the drugs we use as neuromodulators (which is accurate).
The last caveat for this section is the mention of the possibility that patients may have occult bipolar disorder and thus experience a paradoxical effect from antidepressants. These states can be manic, but can also have mixed elements of depression and mania, with an activated misery that is very unpleasant. The atypical presentations of bipolar disorder are more common in women, the majority of our gastroparesis patients. When the medications don’t seem to be working, it may be useful to get a consult from someone who is good at the overlap between medical disorders and psychiatric conditions. The diagnosis and treatment of depression is one of the most important elements of the care of patients with gastroparesis.
Behavioral issues in gastroparesis
William Fordyce, the great leader of Physical Medicine and Rehabilitation, may be seen as the father of behaviorist approaches to chronic pain and rehabilitation. After reading the work of B.F. Skinner, he applied a behaviorist approach to rehabilitation. He used the term “pain behavior” and his work revealed that getting patients to change behavior to increase function in rehabilitation resulted in better outcomes . He focused on behavior rather than on symptoms and used positive and negative reinforcement to change the behaviors he saw associated with chronic pain. He did not think his patients were faking their difficulties nor malingering, but by pushing for improvements he saw many patients make progress that they had not made until these methods were used.
At around the same time, Issy Pilowski, described ‘abnormal illness behaviors’ as behaviors that patients exhibited as part of seeking the “sick role” despite a lack of physiological findings to support the degree of dysfunction they manifest. He pointed out that many patients did not share the goal of rehabilitation and improving function with their doctor, but were committed to continuing in the sick role. They believe that they “can’t” do things that they do not feel emotionally inclined to do. As a result, they often say that they “can’t” attend physical therapy, engage in psychological treatments, or tolerate medications that do not immediately relieve their discomfort. They usually end up on treatments that have no efficacy for pain relief, (such as benzodiazepines), or have lost effectiveness, (opiates), but do not continue treatments (even those that have been shown to be helpful to them), that provide chronic diminution of the sensory complaints that underlie their disorder of chronic pain . He also did not think his patients were malingering, but emphasized how their illness behavior had been ‘conditioned’ to continue.
Gastroparesis patients have intense symptoms including pain, nausea and eating aversion that frequently do not correspond to the degree of delay in emptying nor to other objective findings. We do not think these symptoms are fake or feigned but we have noted that there are a variety of mechanisms by which pain, tinnitus, dizziness, and other sensory conditions are amplified through central and autonomic mechanisms as well as due to neuropathy. While there are a variety of neuromodulators that can reduce these symptoms, we also have noted that symptom amplification can also be conditioned in people and animals by using operant and classical conditioning.
A behavioral approach to patients with chronic gastroparesis pain may also help one understand how some patients develop and continue maladaptive behaviors, and can be the basis for analyzing factors that delay recovery and amplify dysfunction. The behavioral approach also provides a framework for a treatment plan that focuses on rehabilitation, function, quality of life, and healthy behavior that does not imply that the patient is “feigning” their illness.
Behavior can be described as a goal directed activity that either increases with reinforcement, or decreases with a lack of reinforcement. In the early 1900s, Pavlov described ‘classical’ conditioning as the pairing of unrelated stimuli (such as the ringing of a tuning fork) with the presence of stimuli usually associated with a particular behavior (the presence of food is the stimuli for the behavior of salivation). Clinical examples of “classical” or “Pavlovian” conditioning include the gradual development of nausea in cancer patients when arriving at the cancer center even before the administration of chemotherapy . Many patients will spontaneously vomit on arrival to the clinic, even for visits that take place after chemotherapy has concluded. A similar phenomenon has been described by opiate-dependent patients who experienced “cold turkey” opiate withdrawal in a particular environment. They can later experience withdrawal symptoms when exposed to that environment, even after complete discontinuation of opiates. Conditioned withdrawal can easily be produced in experimental animals using this paradigm .
B.F. Skinner described operant conditioning as the shaping of behavior using positive or negative responses to the behavior . He described four types of operant reinforcement. Positive reinforcement, where a behavior results in the delivery of something that is rewarding; negative reinforcement, where the behavior results in the removal of something unpleasant; punishment, where the behavior results in the delivery of something unpleasant; and extinction, where the behavior results in the removal or lack of delivery of something rewarding.
It is common to see medical applications of operant conditioning at work in patients. Opiates have powerful rewarding effects in humans, and therefore behaviors associated with the administration of opiates will increase in frequency and intensity if they consistently result in opiate delivery. Clinicians have been shown to prescribe opiates in response to non-verbal pain behaviors and opiates are often given in response to these behaviors in hospital settings . Physicians also are more likely to prescribe opiates in response to the emotional elements of pain, so that the patient’s display of distress is reinforced and encouraged to increase over time. Nociceptive transmission is enhanced by opiates . Not only are the behaviors related to pain increased by opiate rewards; the pain itself can probably be increased by contingent administration of opiates. Pain, nausea, and opiate withdrawal are aversive experiences, and the administration of opiates will relieve the adverse experience, leading to negative reinforcement. The illness behaviors are reinforced by numerous elements of their everyday existence. Patients are often unaware of the factors that condition them to behave in particular ways, and feel that they “can’t help it” as the behavior has become reflexive and feels automatic.
A behavioral model of addiction
Addiction to drugs is a problem for clinicians who treat gastroparesis. Often, medications that are initially used to relieve pain and distress gradually become a focus for patients who feel like they need these drugs and cannot live without them. The problematic effects of narcotics on the gut are well known, but benzodiazepines, stimulants, and opiates all affect behavior in ways that are negative over the long term and so these must only be used in a time-limited way. Some patients are merely physically dependent on the drugs, and with coaching and slow tapers can stop taking reinforcing drugs that have become ineffective. The doctor needs to take extra time to explain to the patient that the drugs are no longer improving function and are, in fact, diminishing it. Some patients, however, will develop drug-seeking behaviors that are new and look more like those seen in addicted people than in those with illness. The difficulty in defining when the patient has developed an addiction is that it is a process that evolves rather than a discrete change. Opiates are dependence producing and reinforcing and yet many patients are not “disordered” by them. Physical dependence where withdrawal occurs is not the same as addiction. We define addiction as the disordered behavior produced by the increased seeking and use of a substance despite mounting negative consequences. This is a simple behavioral definition, and leaves something to be desired by those who want a clear “category” for when a patient is an addict. Addictive behaviors, such as deception, intoxication, personality deterioration, and self-destructive actions all develop over time with continued drug administration and are reinforced by the drug being used. When describing addictions, it is important to note that these same behaviors can be conditioned to occur in animals, and that it is the drugs that are addictive, and not the patient who is somehow a latent addict. This is not to deny that patients (and animals) clearly vary in the vulnerability to addictive behavior as many elements of vulnerability (genetic, temperamental, social, environmental, and psychiatric co-morbidity) have been demonstrated with valid research.
Psychiatric and psychological factors render people more vulnerable to addiction, including risk-seeking and reward sensitivity. Depression “turns off” the reward system so that ordinary rewards of life are less reinforcing, and people become more sensitive to the rewarding effects of drugs. Life experiences that expose people to drugs and social acceptance of drug use also increase the risk of addiction. Finally, biology is involved in several ways. In the case of alcoholism, genetic make-up affects the degree to which alcohol is rewarding. Some patients tell you that their first drink was so rewarding they began a lifetime of heavy drinking immediately. Others will say they never really liked drinking all that much, and therefore are surprised as they become dependent on alcohol to control the emotional discomforts of their lives. Cocaine addiction is more affected by the genetics of risk-taking and reward-sensitivity that shape personality . Finally, medical conditions such as chronic pain, a variety of disease states, and surgical procedures, may result in exposure to addictive drugs and may amplify their reinforcing effects to the degree that patients may develop iatrogenic addiction.
A behavioral approach to the treatment of gastroparesis
Gastroparesis patients that we see usually have both conditioned pain and other maladaptive behaviors and addictive behaviors. Previous treatment of these patients has often been hampered by medical complexity of the gut nervous system as well as complex psychiatric co-morbidity. All patients need a careful expert evaluation of their gastroparesis as well as a diagnostic formulation that includes the whole person in the context of their life as well as the burdens of their pathologies.
The dysfunction seen in patients with gastroparesis is often a multifactorial dysregulation of the many systems involved in sensing and reacting to pain and gut activity. There is cross-talk between the sensory and autonomic apparatus at every level of the nervous system. Sensory elements of pain can be altered at the nocioceptor apparatus and at every synapse all the way to the thalamus and cortex. The emotional elements of pain and GI dysfunction are likewise complex and open to modification. After a complete work-up for underlying pathologies we use a behaviorally based interdisciplinary approach to our patients.
This process begins with identification of physical and psychiatric conditions that contribute to the problem. Key co-morbidities include depression, personality disorder, and family and social factors that all play a role in the disabling chronic condition. Intoxication with benzodiazepines and opiates (often others), opiate mediated hyperalgesia, opiate induced gut dysmotility, deconditioning, chronic constipation, poor nutrition (or even cachexia), vitamin deficiency, endocrine dysregulation, hypothyroidism, and steroid dependence are commonly seen co-morbidities. These all must be described to the patient and incorporated into the treatment plan.
Physical elements of the disorder such as gut motility, pain, constipation, nausea, food avoidance, and other issues improve only with aggressive pharmacological management when possible, and rehabilitative interventions directed at function. The behavioral elements of treatment are similar to those laid out by William Fordyce in the 1970s. First, behaviors are selected that need to be changed and reinforcers that will be salient to the patient are determined. Often family members must change behaviors that act as reinforcers for the illness behaviors the patient exhibits.
Analyze behavior and reinforcers
Extinguish illness behaviors
Reinforce healthy behaviors
Add reinforcers and expand healthy behaviors
While the reason for pain behavior, distorted eating behaviors, and odd bowel habits may be physiologic, we use physical therapy, occupational therapy and rehabilitation directed at correcting these to the degree physiologically possible.
A variety of behavioral reinforcers have been particularly useful to us in our work. Patient are differentially responsive to reinforcement, some being more consequence avoidant and others more reward seeking. Each patient needs ongoing monitoring of results and ongoing adjustment of the treatment plan. Patients may play an active role in selecting reinforcers as they get better and the transparency of what we are doing does not seem to make it ineffective. We are extremely sympathetic with the discomforts patients must tolerate but push them to engage in treatments such as physical therapy, broadening narrow food choices, accepting therapeutic trials of medications, using distraction, mindfulness and relaxation techniques, and learning food and bowel hygiene. Many of these treatments help with the balance between what we assume to be sympathetic-parasympathetic dysregulation that may play a role in GI dysmotility. We focus on behaviors rather than feelings and progress rather than limitations.
Gastroparesis is a complex disorder involving both sensory and emotional experiences. Symptoms may change over time and may have elements of conditioned behavior in many cases. Chronic opiate treatment is frequently problematic and may result in addictive behaviors, which then exacerbate the conditioning elements of the disorder and its disability. Cannabinoids, benzodiazepines and stimulants can contribute to disability, and a trial of elimination of all these classes should be a part of the treatment plan. A comprehensive approach to rehabilitation of patients disordered by gastroparesis includes some behaviorist methodology. Often function can be improved with a comprehensive program including pharmacotherapy and behavioral change. The current culture that values patient satisfaction and the oversimplification of approaches to pain, (pain is a vital sign), fuels the development of addictive behaviors in vulnerable patients.
Personality issues and gastroparesis
Difficult patients have been a part of medicine since the origin of our field. Thomas Sydenham famously said, “All is caprice. They love without measure those they will soon hate without reason. Now they will do this, now that; ever receding from their purpose .” This quote emphasizes the observation that these patients sabotage their own efforts to get better, much to the frustration of the clinicians who care for them.
Discussions of personality and temperament are often confounded by entrenched psychological theories and dogma about the basis personality. The concept of borderline personality originates in psychoanalytic ideas of the early 20th century that have become doctrinaire. To avoid these issues, we will discuss personality in simple terms; personality is the usual way that a person will respond to a stimulus or problem. Personality is a product of underlying temperament or disposition, which then interacts with the learned behavior from experience to produce the characteristic pattern we observe in other people. People vary in personality in specific measurable ways. What is important or motivates people is largely determined by elements of temperament and certain features of the world have more or less ‘salience’. People vary in their sensitivity to rewards, consequences, emotions, how they are perceived by others, rules, threats, their bodily sensations, and interactions with other people. These factors tend to group together and have been measured in a variety of ways.
Although overly simplistic, patients can be described as having endowments or temperament along a dimensional curve such as shown in Fig. 27.3 . When problems become difficult or frustrating, patients tend to tenaciously cling to their core style of responding. If their core style does not help solve the problem, it becomes very difficult to persuade them to try an alternative approach. Personality and temperament can be parsed in many ways, but we discuss the approach we have found most useful, derived from Hans Eysenck and other personality theorists. We describe introversion and extraversion along a normal distribution and another curve that describes stability and instability.