Gastrointestinal System Endoscopy and Pathological Findings in Non-Dialysis Chronic Kidney Disease: A Single-Center Study

Introduction

Complications can occur due to decreased kidney function in patients with chronic kidney disease (CKD) [1]. Gastrointestinal problems are one of them. In patients with CKD, hypergastrinemia due to uremic toxins is associated with decreased gastrointestinal motility, hypo- and hypersecretion of stomach acid, polypharmacy, and the risk of gastrointestinal disturbances due to Helicobacter pylori (HP) infection [2,3]. Psychological problems are also an additional risk factor. Therefore, in patients with CKD, dyspeptic symptoms characterized by upper-abdominal discomfort, nausea, vomiting, abdominal pain, anorexia, indigestion, and regurgitation may occur, and the risk of malignancy increases [1,2]. HP can cause serious diseases such as peptic ulcer, gastritis, stomach cancer and lymphoma [2,4]. While it occurs in 15–74.4% of the non-CKD population, this rate is 20–64% in patients with CKD [4]. Patients with CKD who are not end-stage are potential candidates for dialysis and kidney transplantation. After these patients reach the end-stage, their gastrointestinal problems increase due to the effects of uremia and immunosuppressive drugs. Therefore, before moving on to this stage, especially for those who are symptomatic, endoscopic evaluation and treatment may be necessary.

In this study, we assessed endoscopic and pathological findings in patients with severe CKD who underwent upper-gastrointestinal system endoscopy due to dyspepsia.

Material and Methods

STATISTICAL ANALYSIS:

We used the Shapiro-Wilk test to assess the normal distribution of continuous variables. We present descriptive statistics for group comparisons, including mean±standard deviation or median (minimum: maximum) values. We made comparisons of numerical variables in both groups using the Mann-Whitney U test and the independent paired-sample t test. We used a χ² test (or Fisher’s exact test if any expected cell count was <5) and present descriptive statistics as numbers and percentages to analyze categorical data. Binary logistic regression analysis was performed to determine independent risk factors affecting the presence of endoscopic gastritis and duodenitis, pathological chronic gastritis, and HP. The backward elimination method was used in multivariate logistic regression analysis to find the best final model. Data were analyzed using SPSS Software (IBM Corp. Released 2015. IBM SPSS Statistics for Windows, Version 23.0. Armonk, NY: IBM Corp). A P value of <0.05 was considered statistically significant.

Results

A total of 128 patients were included in the study. The mean age in the CKD group was similar to that the non-CKD group (median age was 59) [59.5±11.5 (25–79) vs. 53.6±13.2 (29–86) years] (P>0.05). Sex and primary disease distributions were similar in the 2 groups. When comorbid diseases were evaluated, diabetes mellitus (38.6% vs 23.7%) and hypertension (65.3% vs 34.2%) were significantly higher in the CKD group compared to the non-CKD group (P<0.001). Laboratory values of both groups were given in Table 1. Serum urea, creatinine, potassium, phosphorus, CRP, and ferritin values were significantly higher in the CKD group, whereas eGFR, calcium, total protein, total and HDL cholesterol, erythrocyte, hemoglobin, and hematocrit values were significantly higher in the non-CKD group.

Endoscopic findings are given in Table 2. The 2 groups had similar rates of non-ulcerous and ulcerous lesions, such as gastroesophageal reflux, antral erosive gastritis, duodenitis and fundus polyp. The CKD group had significantly higher rates of antral gastritis, erythematous gastritis, erosive pangastritis, active chronic gastritis, and esophageal varicose veins, whereas the non-CKD group had significantly higher rates of esophagitis LA and bulbitis (Figure 1). Pyloric diverticulum was not detected in either group.

Histopathological findings obtained from biopsy samples are shown in Table 3 and Figures 2–4. The groups had similar rates of intestinal metaplasia and oedema-congestion, atrophic metaplastic gastritis, fundic and hyperplastic polyps, and chronic gastritis. The incidence of inflammation in the CKD group was significantly higher than in the non-CKD group (48.3% vs 20.6%, P=0.008). The incidence of HP in the CKD group was significantly lower than in the non-CKD group (23.5% vs 59.1%, P<0.001). Dysplasia was not detected in either group.

Multivariate logistic regression analysis determined independent risk factors affecting the development of endoscopic gastritis, duodenitis, pathological chronic gastritis, and HP. The logistic regression models obtained in the last stage of the analysis were found to be significant, and the data set was found to be compatible with the models. The variables included in the regression analysis models were determined for each dependent variable. For the presence of endoscopic gastritis: pathological chronic gastritis, HP positivity, hypertension, triglyceride, LDL and HDL cholesterol (Model χ2=11.5, P=0.009; Hosmer-Lemeshow test P=0.903). For the presence of endoscopic duodenitis: age, sodium, hemoglobin, calcium and lipase (Model χ2=5.31, P=0.049; Hosmer-Lemeshow test P=0.724). For the presence of pathological chronic gastritis: sex, antral ulcer, bulbar ulcer, duodenal aphthous ulcer, endoscopic gastritis, urea, sodium, total protein, albumin, LDL cholesterol, ferritin, and phosphorus (Model χ2=12.22, P=0.007; Hosmer-Lemeshow test P>0.99). For HP positivity: endoscopic gastritis, erosive gastritis, urea, total and LDL cholesterol, ferritin, calcium, and gamma-glutamyl transferase (GGT) (Model χ2=12.51, P=0.002; Hosmer-Lemeshow test P=0.632). In multivariate analysis, only gastritis (P=0.028) and calcium (P=0.008) were associated with HP positivity. The variables included in the models created in the last step of the multivariate logistic regression analysis are shown in Table 4.

Discussion

Upper-gastrointestinal system disorders are common in CKD patients due to high ammonium levels, systemic and local circulatory disorders, frequent drug use, and hypergastrinemia [5]. However, in most patients, symptom severity does not correlate with severity of gastroduodenal lesions as assessed by endoscopy, and most patients have mild symptoms [6]. In our previous study of patients receiving hemodialysis (HD) or peritoneal dialysis (PD) maintenance therapy for end-stage renal disease, most did not have any significant upper-gastrointestinal symptoms despite the high rate of gastritis in HD (63.6%) and PD (61.3%) patients [7]. In the current study, antral gastritis, erythematous gastritis, erosive pangastritis, and active chronic gastritis were found to be significantly higher in our non-dialysis uremic patients with Stage 3b and Stage 4 CKD, while the rates of esophageal varicose veins, esophagitis, and bulbitis were higher in the non-uremic control group. Lower albumin, total cholesterol, and LDL and HDL levels are seen in addition to anemia in many patients, often due to infection, inflammation, nutritional deficiency, or erythropoietin deficiency.

In different studies conducted in chronic HD patients, erosive esophagitis 30.2%, gastroesophageal reflux 10%, esophagitis 5.8%, exanthematous pangastritis 57.3%, diffuse antral erythema 27.8%, gastric erosion 5.8%, gastric antral erosion 22.8%, gastric intestinal metaplasia 8.3%, gastritis and duodenitis 42%, duodenal erosion 18–32%, peptic ulcer 7.3%, gastric ulcer 7–14%, duodenal ulcer 7.3–18%, angiodysplasia 4.4%, and gastric polyp were observed at rates of 1.5% [5,8–11]. In our previous study of dialysis patients, we found gastroesophageal reflux in 41.5%, chronic gastritis in 75.5%, non-specific gastritis in 24.5%, intestinal metaplasia in 17%, atrophy in 15.1%, HP in 28.3%, and hyperplastic polyp in 3.8% [7]. Gastric dysplasia was not detected in either of our studies. While rates of antral gastritis, erythematous gastritis, erosive pangastritis, active chronic gastritis, and esophageal varicose veins were detected at lower rates in the CKD group, esophagitis and bulbitis rates were significantly higher in the non-CKD-group.

The frequency of ulcerous lesions was similar in both groups. The frequency of ulcerative lesions in the general population is 10–20% [12]. A retrospective study of dialysis patients found 31.8% had peptic ulcers, and age, low albumin, high GGT, peritoneal dialysis, and diabetes were reported as risk factors [13]. In the present study, intestinal metaplasia was detected in 8.3%, edema-congestion in 10%, inflammation in 48.3%, HP in 23.5%, atrophy metaplastic gastritis in 3.3%, fundus polyp in 3.3%, hyperplastic polyp in 6.7%, and chronic gastritis in 43.3%. The rates in the control group were 14.7% for intestinal metaplasia, 2.9% for edema-congestion, 20.6% for inflammation, 59.1% for HP, 2.9% for fundus polyp, 2.9% for hyperplastic polyp, and 41.2% for chronic gastritis. Inflammation was significantly higher in the CKD group than in the control group. However, although this rate in dialysis patients is reported to be as high as 75.5% in the literature, the frequency of chronic gastritis that we found to be higher in the CKD group in our study did not reach statistical significance [7].

HP was more common in the control group than in the CKD group (59.1% vs 23.5%). While the prevalence of HP is 24–32% in the general population, it was found to be 28.3% in dialysis patients [7,12]. In the present study, the rate of chronic gastritis was 41.2% in the non-uremic patient group with high HP and 43.3% in the CKD group. In a previous study of hemodialysis patients, the prevalence of H. pylori infection decreased as dialysis periods progressed within the first 4-year follow-up after the start of HD [14]. Possible protective mechanisms include antibiotic use or antiacid drugs causing hypochlorhydria, which could contribute to the decreased frequency of HP. Our study’s HP prevalence was 23.5%, less than in the non-CKD population. Also, uremia has a deteriorating effect on glycolipids and external membrane proteins resistant to the external environment, including the lipopolysaccharide cell wall structure of HP colonizing in the gastric mucosa. Our patients with CKD had a higher rate of chronic gastritis despite a lower HP rate, but we did not detect a relationship between HP and chronic gastritis. In multivariate regression analysis, only the presence of gastritis and serum calcium level were associated with HP positivity.

The rates of esophagitis (13.2%) and duodenitis (18.9%) in chronic dialysis patients in our previous study were close to the rates in the current study (13.3%) [7]. In the non-uremic patient group, the rate of esophagitis was higher (53.7%). In studies conducted in HD patients, gastroesophageal reflux was found in 10%, 15% in our patients, and 6% in the control group [5,9,10], while pangastritis was high in HD patients (57.3%), and this rate was lower in our patients (13.3%). Pangastritis is very common in HD patients. While antral erosive gastritis was found in 27.8% of HD patients, this rate was lower (6.7%) in our non-dialysis CKD patients. Antral gastritis, erythematous gastritis, erosive pangastritis, active chronic gastritis, and inflammation were significantly higher in our non-dialysis CKD patients than in the control group. Many drugs used for complications such as high serum urea level, inflammation, stress, mineral bone disorder, anemia, and hypertension in patients with CKD may be effective in this condition, together with uremic toxins [13,15].

A recent study found that gastrointestinal symptoms and signs were common in patients with CKD. Some symptoms and diseases also varied according to whether the renal replacement was peritoneal or hemodialysis type. The prevalence of esophagitis and gastritis diagnosed by symptomatic evaluation varied depending on the type of dialysis and whether it occurred in the pre-dialysis period. Although our study differed from this one in that it included endoscopic diagnostic classification of symptomatic pre-dialysis patients compared to non-uremic patients, the significant difference in the frequency of esophagitis and gastritis histologic subtypes between the groups is compatible with the above study [16].

In our study, no statistically significant independent relationships were detected between hemoglobin, urea, and total cholesterol and the presence of duodenitis, chronic gastritis, and HP in multivariate logistic regression analysis. Our previous study of dialysis patients showed the independent effect of HP and total cholesterol [7].

Conclusions

In uremic patients, risk factors like urea-related toxins, impairment of local and systemic circulations, and hypergastrinemia cause damage to the gastric mucosa. In addition, in chronic uremic patients with cardiovascular damage, anemia and mineral bone disease are frequently seen. These complications and the medications administered to treat them can adversely affect the upper-gastrointestinal system. In the general population, chronic gastritis, inflammation, and gastric and duodenal ulcers are seen frequently. These diseases generally cause symptoms such as vomiting, anorexia, abdominal pain, heartburn, and bleeding. Although most chronic uremic patients have these lesions, they remain asymptomatic.

Even though most chronic uremic patients have these lesions since they remain asymptomatic so, their diagnosis and treatment are often delayed; therefore, the lesions that arise in the gastrointestinal system among pre-dialysis uremic patients should be treated early, bearing in mind that they are potential candidates for kidney transplantation. Adverse effects on the gastrointestinal system of the drugs used, especially in uremic patients, should be minimized.