Symptom-Based Diagnosis—The Functional Mainstay

The perception of ill health reported by the patient when consulting a physician is the basis for all diagnostic decision making except in certain emergency situations. The combined outcome of how the individual physician judges the medical history and the results of diagnostic tests and treatment trials affects the way a diagnosis is seen. Traditionally, what cannot be seen, measured, or assessed by a positive treatment response is regarded as less confirmative of a “real” disease compared with diagnoses that are obvious from 1, or preferably all 3, of these viewpoints. Even if empiric treatment is the mode of diagnosis, we tend to accept it as proof for somatic disease without too much hesitance, as long as there is a symptom improvement, as is the case with acid suppression for reflux symptoms.

A major problem regarding the 28 adult FGIDs as defined today is that most of them do not, in any convincing way, meet our traditional view of clinical diagnoses. In the first instance, the labeling of FGIDs as disorders creates some confusion because this is a common term in psychiatry but not in somatic medicine. The routine practice to exclude organic disease before making a diagnosis of a FGID further emphasizes this; rule out “diseases” and the “disorders” are what remains.

During the last 2 decades, the Rome process has changed our ways of thinking about FGIDs quite a lot. The Rome process started out by using consensus of opinion but has developed into a more-or-less worldwide scientific joint venture for creating evidence-based and improved knowledge regarding FGIDs. One of the objectives of the Rome process was to create the means for a positive diagnosis of FGIDs; exclusion of organic disease should no longer be needed. The message from Rome is that clusters of symptoms with a minimum duration of six months and without alarm symptoms can safely be used for diagnosing a benign disorder with a reasonably well-defined prognosis. The dissemination of this strategy, in particular to community providers of health care, may still have a long way to go. Although the stability over time of a given FGID is poor, with FGIDs tending to change labels, eg, from irritable bowel syndrome (IBS) to functional dyspepsia or functional constipation, new organic diseases do not appear to develop more often than in the general population. The major diagnostic drift stays within the framework defined by FGID. The diagnostic certainty conveyed by symptom criteria is an important step forward in everyday work and a trustworthy basis on which to start in the therapeutic relation with a patient.

Several problems seem inherent to FGID. To start with, those who seek medical advice for FGID have either more severe symptoms than nonconsulting FGID patients or carry more psychological problems, like anxiety, depression, somatization, and general health concerns. It is vital to understand the patient’s reasons for seeking medical advice and to also address contributing factors. Medical therapy for FGIDs is hampered by the lack of efficacious drugs. Symptomatic treatment for patients with diagnoses of unknown etiology and uncertain pathophysiology is a challenge. A confident and skilled physician seems to increase the chance for improvement, even if the treatment modes are not particularly effective. This is exemplified by the high placebo response to treatment interventions in IBS, which is good for short-term success but may increase the risk for continued use of ineffective drugs unless careful follow-up is performed.

Physiology-Based Diagnosis

The diagnosis of dysmotility requires some means for measuring motor activity of the gastrointestinal tract. Any such measurement, in turn, requires that the boundaries of normal motility are known, and that this measurement is performed using an agreed upon or standardized technique. This is where the concept of physiology-based diagnosis starts to become complicated. Measurement techniques are constantly evolving, and only a few techniques have finally made it into clinical practice. The market for measurement systems is relatively small, and, although producers of measurement instruments aim for solutions that will sell enough numbers of units to make a profit, they depend on medical researchers and research groups for defining clinically relevant measures. In many areas of gastrointestinal physiology, it has been difficult to establish clinical relevance. A typical example is electrogastrography, ie, the measurement of gastric electrical activity from cutaneous electrodes. Despite developments in hardware and software for automated signal analysis, the clinical relevance of measuring gastric electrical activity has remained unproven. Expertise in a number of techniques, such as 3-dimensional ultrasonography for gastric accommodation and emptying and intraluminal manometry of the left colon, has been confined to a few groups. At any given point it has been very difficult to predict the winners among currently available techniques.

As a corollary, only a few measurement techniques have gained widespread use. These include 24-hour pH-monitoring of esophageal exposure to acidic reflux, esophageal manometry, gastric emptying using a radionuclide-labeled meal, whole-gut or colonic transit using radio-opaque markers, and anorectal function testing using a catheter with a balloon. In specialized centers, a few more techniques have been used to some extent: combined impedance and pH monitoring for gastroesophageal reflux, small bowel manometry, satiety drinking tests, breath tests for orocecal transit, scintigraphic colon transit, and defecating proctography. Currently available techniques are reasonably good at detecting well-defined cases of dysmotility (achalasia, gastroesophageal reflux, gastroparesis, pseudo-obstruction, and slow transit constipation). Three of these are actually diagnosed from physiologic parameters: (1) achalasia by absent relaxation of the lower esophageal sphincter (for further details see Chapter 9), (2) gastroparesis by delayed emptying of the stomach, and (3) slow transit constipation by delayed transit through the colon. The diagnosis of gastroesophageal reflux disease can be made from endoscopic findings of erosive esophagitis but also from typical symptoms (heartburn and acid regurgitation) in combination with symptom improvement after antacid or antisecretory medication, whereas 24-hour pH monitoring, impedance, and manometry are seldom used in routine diagnostic work. Pseudo-obstruction is diagnosed on the basis of clinical history (subocclusive events), findings on abdominal roentgenograms (dilated bowel and air/fluid levels), and exclusion of a mechanical cause for obstruction (see Chapter 7). Physiological measurements, such as small bowel manometry, can possibly be used for screening purposes or for differentiation between myopathic and neuropathic types of pseudo-obstruction. The latter, however, remains an issue primarily for the pathologist, if full-thickness bowel tissue can be obtained.

The detection of less well-defined cases of dysmotility is more difficult either because no firm definition exists in terms of motor disturbances or because methods for their detection are lacking. If a patient has alternating diarrhea and constipation, one can hypothesize that these symptoms result from intestinal dysmotility. It is, however, very difficult to prove dysmotility in such a patient using currently available techniques (small bowel transit, small bowel manometry, and colon transit). Even so, because one goal of the Rome process has been to better understand the underlying pathophysiology, groups of patients with FGIDs that no longer are purely without objective findings have been identified. This creates a basis for discussing if dysmotility-associated FGIDs should be defined and if such definitions are useful from a clinical or research point of view?

Physiology-Based Diagnosis

The diagnosis of dysmotility requires some means for measuring motor activity of the gastrointestinal tract. Any such measurement, in turn, requires that the boundaries of normal motility are known, and that this measurement is performed using an agreed upon or standardized technique. This is where the concept of physiology-based diagnosis starts to become complicated. Measurement techniques are constantly evolving, and only a few techniques have finally made it into clinical practice. The market for measurement systems is relatively small, and, although producers of measurement instruments aim for solutions that will sell enough numbers of units to make a profit, they depend on medical researchers and research groups for defining clinically relevant measures. In many areas of gastrointestinal physiology, it has been difficult to establish clinical relevance. A typical example is electrogastrography, ie, the measurement of gastric electrical activity from cutaneous electrodes. Despite developments in hardware and software for automated signal analysis, the clinical relevance of measuring gastric electrical activity has remained unproven. Expertise in a number of techniques, such as 3-dimensional ultrasonography for gastric accommodation and emptying and intraluminal manometry of the left colon, has been confined to a few groups. At any given point it has been very difficult to predict the winners among currently available techniques.

As a corollary, only a few measurement techniques have gained widespread use. These include 24-hour pH-monitoring of esophageal exposure to acidic reflux, esophageal manometry, gastric emptying using a radionuclide-labeled meal, whole-gut or colonic transit using radio-opaque markers, and anorectal function testing using a catheter with a balloon. In specialized centers, a few more techniques have been used to some extent: combined impedance and pH monitoring for gastroesophageal reflux, small bowel manometry, satiety drinking tests, breath tests for orocecal transit, scintigraphic colon transit, and defecating proctography. Currently available techniques are reasonably good at detecting well-defined cases of dysmotility (achalasia, gastroesophageal reflux, gastroparesis, pseudo-obstruction, and slow transit constipation). Three of these are actually diagnosed from physiologic parameters: (1) achalasia by absent relaxation of the lower esophageal sphincter (for further details see Chapter 9), (2) gastroparesis by delayed emptying of the stomach, and (3) slow transit constipation by delayed transit through the colon. The diagnosis of gastroesophageal reflux disease can be made from endoscopic findings of erosive esophagitis but also from typical symptoms (heartburn and acid regurgitation) in combination with symptom improvement after antacid or antisecretory medication, whereas 24-hour pH monitoring, impedance, and manometry are seldom used in routine diagnostic work. Pseudo-obstruction is diagnosed on the basis of clinical history (subocclusive events), findings on abdominal roentgenograms (dilated bowel and air/fluid levels), and exclusion of a mechanical cause for obstruction (see Chapter 7). Physiological measurements, such as small bowel manometry, can possibly be used for screening purposes or for differentiation between myopathic and neuropathic types of pseudo-obstruction. The latter, however, remains an issue primarily for the pathologist, if full-thickness bowel tissue can be obtained.

The detection of less well-defined cases of dysmotility is more difficult either because no firm definition exists in terms of motor disturbances or because methods for their detection are lacking. If a patient has alternating diarrhea and constipation, one can hypothesize that these symptoms result from intestinal dysmotility. It is, however, very difficult to prove dysmotility in such a patient using currently available techniques (small bowel transit, small bowel manometry, and colon transit). Even so, because one goal of the Rome process has been to better understand the underlying pathophysiology, groups of patients with FGIDs that no longer are purely without objective findings have been identified. This creates a basis for discussing if dysmotility-associated FGIDs should be defined and if such definitions are useful from a clinical or research point of view?

The Utility of Physiology-Based Diagnosis

A diagnosis is, in the ideal situation, the assignment of a patient’s complaints to a category that links symptoms with a pathological process and, in some cases, with a specific cause. The classification of patients and their complaints into diagnostic categories is most useful if the diagnosis facilitates an understanding of the nature and causation of the complaints and aids therapeutic or prognostic decision making. The diagnosis of a functional disorder partly helps us in this respect by defining a benign prognosis and some therapeutic options worth trying. It also helps us avoid things not to do, like unnecessary and potentially harmful diagnostic investigations. A major drawback is that our understanding of disease mechanisms is not much enhanced by symptom-based diagnosis, and it is questionable if the symptom-based approach can identify diagnoses with similar etiology, pathogenesis, or pathophysiology.

As an alternative to the Rome process, an international working team presented a pragmatic view on physiology-based diagnosis for the 2002 World Congresses of Gastroenterology in Bangkok. The Bangkok classification divides motility disorders into well-defined entities, entities with a variable dysfunction-symptom relationship, and questionable entities. The well-defined entities in the esophagus comprised gastroesophageal reflux disease, achalasia, and esophageal spasm. Only the diagnosis of dumping syndrome qualified as a well-defined gastric motility disorder and intestinal pseudo-obstruction as a well-defined small bowel motility disorder. Finally, megacolon, Hirschsprung’s disease, and slow-transit constipation were classified as diagnoses with a firm relationship between disordered colorectal motility and symptoms.

Few, if any, patients who currently are labeled as FGIDs would receive a physiology-based diagnosis among the well-defined entities of the Bangkok classification but a fair proportion might qualify for entities with a variable dysfunction-symptom relationship. These include gastroparesis, gastric dysrelaxation, sphincter of Oddi dyskinesia, the new diagnostic entity, enteric dysmotility, and fecal incontinence. Several studies have found that 20% to 30% of patients with functional dyspepsia have delayed gastric emptying, thus qualifying for the diagnosis of gastroparesis. However, a significant proportion of patients with functional dyspepsia (27%–43%) exhibit an increased rate of emptying, in particular, during the early postprandial phase, and about 40% of patients with functional dyspepsia may have gastric dysrelaxation. The relation of these pathophysiologic findings to symptoms is weak and inconsistent between studies, but impaired gastric accommodation may be associated with early satiation and delayed gastric emptying with postprandial fullness, nausea, and vomiting. Recently, a large study assessing colon transit by scintigraphy in a cohort of patients with functional constipation and constipation-predominant IBS, concluded that an underlying motor disorder could be found in about 30%. As in functional dyspepsia, a lack of meaningful associations between colon transit and gastrointestinal symptoms other than the stool habits, particularly stool consistency, have been hard to find and have not been extensively investigated. Even more problematic is the observation that about one quarter of patients with IBS change their predominant bowel pattern, at least once within a year.

A new entity with a variable dysfunction-symptom relationship that seems to make a difference in prognostic and therapeutic decision making is enteric dysmotility. The increased use of small bowel motility testing in patients with bowel symptoms has made it clear that there exist patients who have disturbances of small bowel motor activity that are characteristic of intestinal pseudo-obstruction but that lack the radiologic features of this diagnosis. Such patients may represent a transitional state between the severe end of the spectrum of functional bowel disorders and pseudo-obstruction or a previously unrecognized subgroup of patients with FGID. The latter is supported by the finding of abnormal motor patterns in 39% of patients with IBS. Patients with enteric dysmotility have a better prognosis and less need for parenteral nutrition than those with intestinal pseudo-obstruction. The meaningfulness of distinguishing enteric dysmotility from IBS remains to be shown.