Abstract
The present short update reviews the evidence for any association between chronic renal failure (CRF) and hyperacidity states of the upper gastrointestinal tract. Each gastrointestinal entity will be addressed in turn—namely gastroesophageal reflux disease, peptic ulcer disease and Helicobacter pylori, and dyspepsia.
Gastroesophageal Reflux Disease
Gastroesophageal reflux disease (GERD) develops when excessively acidic gastric contents reflux into the esophagus. The condition is believed mainly to be due to an increase in the number of normally occurring transient lower esophageal sphincter (LES) relaxations. Other major mechanisms include decreased clearance of esophageal contents and refluxate owing to impaired peristalsis, decreased gastric emptying with resultant back pressure into the esophagus, and increased gastric acid production with a resultant increase in the potency of the refluxate. To our knowledge, no studies have examined the rate of transient LES relaxation or esophageal clearance in CRF, but some work has been done on the rates of gastroparesis or hyperacidity in uremia.
Uremia and Decreased Gastric Emptying
Nausea and vomiting are often encountered in CRF and can be caused by delayed gastric emptying. The association of CRF with delayed gastric emptying seems to depend on the type of patient. A patient on peritoneal dialysis would be expected to have delayed gastric emptying simply due to the physical impairment caused by the peritoneal fluid. Indeed, it has been shown that, when filled with dialysate, 5 of 10 peritoneal dialysis patients experienced delayed emptying of radio-opaque markers after a test meal. That compares with 0 of 15 normal controls and 1 of 10 dialysis patients when drained (1).
Results in other CRF patients—such as patients with uremia but not on dialysis, or patients on hemodialysis—have yielded conflicting results. Freeman et al (2) showed no significant difference in the half-time for liquid or solid emptying between uremic pre-dialysis patients and controls. However, that small study looked at only 9 patients with uremia. Three other studies demonstrated delayed gastric emptying in such patients (3-5). Ravelli et al (4) found that 70% of CRF patients (7 of 10) experienced delayed emptying with 5% glucose or milk. Kao et al (3) found 80% of non dialyzed uremic patients (16 of 20) had delayed gastric emptying of radiolabelled solid meals. Unfortunately, that study had no non uremic control group, because the comparison was between CRF patients who were on hemodialysis and those who were not. The researchers found no difference between the two groups of patients (95% of hemodialysis patients also had delayed gastric emptying). Others have also documented delayed gastric emptying in hemodialysis patients, particularly patients who are hypoalbuminemic (6,7). Prokinetics administered to the hypoalbuminemic patients not only increased gastric emptying, they actually improved plasma albumin levels. But the situation is muddled by other studies that have not demonstrated decreased gastric emptying in hemodialysis patients (1,5,8,9).
The reasons for the discrepancy between the various results are likely multifactorial. As previously mentioned, one of the confounding factors is whether the patient is pre-dialysis, on peritoneal dialysis, or on hemodialysis. Ko et al (10) demonstrated that uremic patients undergoing dialysis had gastric dysrhythmias that deteriorated after hemodialysis. Hence, the dialysis status of a patient (on hemodialysis, or not) seems to affect gastric rhythm; the rhythm disturbance can, in turn, lead to disturbed gastric emptying.
The presence of an autonomic neuropathy may also confound the issue. Dumitrascu et al (11) demonstrated that patients with CRF had delayed gastric emptying if parasympathetic and sympathetic neuropathy were both present. Diabetics are at increased risk of autonomic neuropathy, and they certainly constitute a large proportion of patients with CRF.
The final confounding factor may be the type of study performed. Some studies are performed with radio-opaque markers; others are performed with radioisotope-labelled food. Some studies use ultrasound tests to measure gastric emptying times. Some studies look not at gastric emptying, but rather at an electrogastrogram.
To summarize, some CRF patients certainly experience delayed gastric emptying. The proportion of the CRF population that such patients represent is extremely difficult to estimate, but it seems to be slightly higher than normal. Delayed gastric emptying can increase gastric reflux and can therefore potentially be a mechanism for GERD in those patients.
Uremia and Acid Production
Another mechanism by which uremia can potentially lead to GERD is increased acid production. Higher acid production can occur secondary to hypergastrinemia, which itself can be a consequence of decreased clearance of gastrin owing to a reduced glomerular filtration rate (GFR). Hypergastrinemia is in fact present (12,13) in CRF patients, but may also be due in part to increased secretion, because the density of G cells is increased in CRF patients (14,15), possibly owing to a hyper-parathyroid state. Elevated serum gastrin causes an increase in acid production by the parietal cells. The resulting elevation in gastric acid would increase the potency of the refluxate. However, when measured in those patients, large variations actually occur in the mean gastric acid output—varying from achlorhydria to hyperchlorhydria (16). Indeed, achlorhydria has been suggested as the stimulus for hypergastrinemia, because a strong feedback mechanism exists between the G cells and the low-acid state that stimulates the secretion of gastrin.
Here, too, many factors can confound the issue of whether uremia is truly associated with increased acid production. Although a generalization cannot be made that increased acid production occurs in CRF, some patients certainly have increased acid production, and that increase may potentially be a mechanism by which CRF can be associated with GERD.
Does GERD Occur More Often in CRF?
Although mechanisms exist that potentially lead to an increase of GERD in CRF, their existence would be insufficient to suggest that the incidence of GERD is increased in CRF patients without further study.
In adults, existing studies have failed to demonstrate an association of increased GERD with CRF. In an endoscopic surveillance study, no difference was seen in the incidence of esophagitis between patients with CRF and healthy controls (17). We are unaware of any 24-hour pH-metry studies in adults; however, in children, some evidence suggests that GERD occurs at fairly high rates and that it seems to have clinical implications. In an uncontrolled study using 24 hour pH-metry, Ruley et al (18) found that 73% (16 of 22) of infants with CRF had significant GERD. That prevalence rate certainly exceeds the norm. Interestingly, the incidence of vomiting was not associated with the presence of GERD. However, in infants, recurring pain owing to GERD could lead to a refusal to eat and, hence, feeding difficulties. Adequate GERD treatment may improve feeding and outcome for such patients.
Peptic Ulcer Disease
Peptic ulcer disease (PUD) has two major causes: Helicobacter pylori infection and the use of non steroidal anti-inflammatory drugs (NSAIDs). One would not expect CRF patients to have an increased exposure to NSAIDs. Hence, more PUD in those patients should not be attributed to that cause. However, infection of the gastric mucosa with H. pylori may theoretically be expected to be increased in CRF patents. The organism's strong urease activity is the reason that it is able to survive the very high acid milieu of the stomach. Urease converts urea to ammonia and CO2. The ammonia produces a relative neutralization in the immediate environment of the organism, thus facilitating its survival (19,20). Given that, in uremia, urea can be expected to be more available, the postulation that circumstances facilitate colonization of a patient with the organism is reasonable. Such colonization would ultimately lead to higher rates of ulcer disease.
A histopathology study in 70 asymptomatic uremic patients under maintenance hemodialysis demonstrated that chronic superficial gastritis was present in 52%, atrophic gastritis in 5.7%, and intestinal metaplasia in 37% (21). Those entities are associated with the presence of H. pylori. Unfortunately, that study did not compare the rate of H. pylori infection in patients with uremia with the rate of infection in a control population without uremia.
A seroprevalence study (22) for H. pylori immunoglobulin G (IgG)–specific antibodies in continuous ambulatory peritoneal dialysis (CAPD) patients as compared with an age-matched Greek general population found no difference between the groups (64% vs 60%). Perhaps more surprising is the fact that several other studies have shown that rates of H. pylori infection in CRF are lower than expected (16,23-25).
Shousha et al (24) found that the prevalence of antral H. pylori was significantly less in patients with renal disease [12 of 50 patients (24%)] than in a control group [51 of 120 controls (42%)]. Kang et al (26) found a similar trend in a serologic assessment of uremic patients as compared with functional dyspepsia patients. The difference did not achieve statistical significance, however. It would appear that, if PUD occurs more often in CRF, it does not seem to be secondary to infection with H. pylori.
Kang et al (16) reviewed all the studies that examined the rates of PUD in CRF. Although some early studies suggested an elevated prevalence of PUD in as many as 20% – 60% of patients with uremia, those results were not confirmed by more recent, better-designed studies. The initial studies were, for the most part, radiologic studies rather than endoscopic studies. They also enrolled relatively small numbers of patients (16,27-32). The more recent, endoscopic studies did not demonstrate a significant increase in the rate of PUD (17,33-38). When the results of the endoscopic studies were combined, duodenal ulcer was present in only 4%, and gastric ulcer in only 1%, of 922 patients. Those rates are in keeping with what would be expected in the general population.
Although the rate of PUD is not higher in patients with uremia than it is in controls without uremia, a difference in the characteristics of PUD in the two groups does seem to exist (26). Two groups of patients with PUD were compared: 43 had uremia, 118 did not. Histologic gastritis was found to be less common in the patients with uremia. H. pylori infection was less common in uremic PUD patients than in non uremic PUD patients, but it was about the same as in uremic patients without PUD. The sex ratio in uremic patients with PUD was equal; in non uremic patients, PUD occurred more often in men. In addition, as compared with non uremic patients with PUD, uremic patients with PUD were more likely to be pain-free, to present with hemorrhage, and to have post-bulbar ulcers or multiple ulcers (26).
Dyspepsia
Dyspepsia refers to pain or discomfort in the upper abdomen (39). It is often associated with heartburn, nausea, bloating, or a feeling of indigestion or slow digestion. The differential diagnosis of the symptom before investigation includes, for the most part, ulcer disease or non ulcer dyspepsia (functional dyspepsia). Gastric cancer is another possibility, but occurs much less frequently. Atypical GERD, pancreatic or biliary disease, and non gastrointestinal diseases are other less-frequent diagnoses.
Non ulcer dyspepsia, defined as a patient presenting with dyspeptic symptoms and having a normal gastroscopy is believed to be secondary either to increased sensitivity of gastric neurofibers (visceral hypersensitivity) or to increased gastric irritation from various potential irritants. The irritants could include H. pylori gastritis, hyperacidity, or prolonged exposure to food antigens owing to gastroparesis (39). As discussed earlier, the latter two entities may occur in CRF and could potentially lead to an increase in non ulcer dyspepsia as compared with a general population. However, no studies that we are aware of have looked at that entity in CRF. The likely reason is that the study would be exceptionally difficult to perform. No single diagnostic test for dyspepsia exists, and uremic patients are often taking many medications, some of which can cause dyspepsia. Given the nonspecific nature of non ulcer dyspeptic symptoms, it is unlikely that a study examining that condition in CRF will ever be performed.
Renal Transplant Patients
The renal transplant patient is different from other CRF patients mainly because of the medications taken. Upper gastrointestinal diseases are definitely associated with those patients, not because of previous or underlying renal failure, but because of the immunosuppression that the transplant requires. Renal transplant patients are at the same risk with regard to upper gastrointestinal diseases as are any transplantation patients receiving immunosuppression. Namely, they are at risk for esophagitis and, in particular, infectious esophagitis from opportunistic infections such as candida, Herpes simplex virus, or cytomegalovirus. With regard to ulcer disease, the rate of complicated PUD is increased in transplantation patients as compared with controls. Bleeding, for example, is said to occur at a rate of 4.4%, and perforation of an ulcer at 0.5% (16). The patients are also at risk for infectious ulcer disease with cytomegalovirus.
Summary
The prevalence of H. pylori infection and PUD seem not to be different in CRF patients as compared with the general population. However, PUD in CRF patients seems to have some unique features—namely, lack of pain and higher associations with bleeding, with post-bulbar location, and with multiple ulcers. No increase in GERD has been proven in adults, but several studies demonstrate increased GERD in pediatric CRF patients. The causes of the increase in GERD may include delayed gastric emptying owing to altered myoelectric activity, or perhaps to an increased production of gastric acid, but evidence for the latter is small. Importantly, treating the problem may lead to better nutrition and higher albumin levels, thus improving patient prognosis.