Serum levels of ghrelin and obestatin in children with symptoms suggestive of delayed gastric emptying of unclear etiology

Study subjects

All children referred for further investigation at a tertiary center (Pediatric Gastronterology Unit; 4th Pediatric Department, Papageorgiou Hospital; Aristotle University of Thessaloniki; Greece) for symptoms (duration >6 months) indicative of delayed GET between March 2007 and March 2013 were included in the study.

Twenty healthy, age-matched children without any GI symptoms or evidence of other organic disease were also recruited as controls among patients examined at the pediatric outpatient clinic for routine assessment. Inclusion and exclusion criteria are shown in (Table 1).

Table 1 Inclusion and exclusion study criteria

This study had institutional approval and was conducted according to Helsinki Declaration. Written informed consent was obtained from all study participants.

Assessment for potential organic disorders

All children of the present cohort were investigated for the presence of organic diseases with detailed clinical history, careful physical examination and a number of laboratory tests. The latter included complete blood count, baseline biochemistry with liver, pancreatic and renal function tests, C-reactive protein, erythrocyte sedimentation rate, urinalysis with culture, stool microscopy, and screening for celiac disease (total IgA and IgA anti-tissue transglutaminase antibodies) and H. pylori infection (serum anti-H. pylori IgG and IgM antibodies). Mechanical upper GI obstruction or anatomical abnormality was excluded in all patients with a combination of upper GI contrast series and/or esophagogastroduodenoscopy (EGD).

Gastric emptying scintigraphy

After an overnight fast children underwent a 2-h gastric emptying scintigraphy with a standardized meal of solids. The meal consisted of 2 AA scrambled eggs labeled with technetium 99m (99mTc) sulfur colloid (1 mCi), two thin slices of white bread, 1 teaspoon of butter (5 g) and 150mL of water to ease the ingestion process as the patients had to consume the aforementioned meal within 10 min. The amount of the test meal was calculated according to patient body surface area (test meal in g/1.73 m2).

Immediately after meal ingestion all children were placed in the upright position in front of a γ-camera with parallel-hole collimator (Vertex-Plus Epic, ADAC/Philips). Anterior and posterior abdominal images were acquired at 15 min intervals for 1 min for a total duration of 2 h. Areas of interest were subsequently set in the stomach area and the rate at which the radioactive material left the stomach was calculated based on the geometric mean.

Gastric half-emptying time (T_1/2) was defined as the time by which 50% of the radiolabeled meal of solids had left the stomach. GET was considered delayed if T_1/2 was >90 min.

In the intervals of image acquisition children were allowed to walk outside the γ-camera room; however, and until the test was over, they were neither allowed to eat or drink nor to engage in intense play.

Blood sample collection and analysis

Venous blood samples (4 mL) were collected between 08:00-09:00 am immediately prior to the ingestion of the meal for the scintigraphic gastric emptying study and 2 h later, upon completion of the test. The blood samples were collected into biochemistry tubes (not containing EDTA, aprotinin or hydrochloric acid), and were immediately centrifuged at 4° C for 5 min at 3000 rpm. Sera were subsequently extracted and stored at -24° C.

Serum total GHR and OB were measured using the human GHR and OB RIA commercially available kits (Phoenix Pharmaceuticals, Inc., Burlingame, California, USA). Due to ethical reasons gastric emptying scintigraphy was not performed and only morning fasting blood samples were collected for the measurement of preprandial levels of total GHR and OB in the control group.

Statistical analysis

Patient characteristics, serum levels of GHR and OB, GHR/OB ratio and GET T_1/2are expressed as mean ± standard deviation and as median and range depending on the normality of the distributions of continuous variables as assessed with Kolmogorov-Smirnov test.

Comparisons of preprandial GHR and OB as well as GHR/OB ratio between patients and controls were performed with Student’s t-test, whereas paired t-test was used for comparisons between pre- and postprandial levels. Finally, t-test was also used to compare both pre- and postprandial levels between patients with normal and delayed GET. Categorical variables were compared using chi-square test. Furthermore the adjusted effect of different factors [gender, age, body mass index (BMI z-score)] on GET, levels of GHR and OB, GHR/OB (pre- and postprandial) was explored using multiple regression analysis. Statistical analyses were performed with SPSS version 17.0 software (SPSS Inc, Chicago, IL). P values <0.05 were considered statistically significant.

Demographics and clinical characteristics

Twenty prepubertal children (7 male, 35%) with symptoms suggestive of delayed GET (early satiety, postprandial fullness, nausea, vomiting) that remained of unexplained etiology (duration of symptoms >6 months) were identified and included in the study. Of note, the initial diagnostic work-up which included a complete blood count, baseline biochemistry with liver, pancreatic and renal function tests, C-reactive protein, erythrocyte sedimentation rate, urinalysis with culture, stool microscopy, celiac disease (total IgA and IgA antitissue transglutaminase antibodies) and H. pylori (serum IgM and IgG antibodies) screening was within normal range. Furthermore, no mechanical upper GI obstruction or anatomical abnormality was revealed in any patient by means of a combination of upper GI contrast series and/or EGD. Of note, the histopathology of the biopsies taken during all EGD procedures was also without any abnormalities. Demographic data and baseline patient characteristics are presented in (Table 2). Twenty healthy, age-matched controls were also included.

Table 2 Study participants’ clinical characteristics and laboratory findings

GET

Nine of 20 patients had delayed GET (T_1/2 >90 min). The median time for gastric emptying rate was 80.5 (range: 55-210) min. GET was not significantly affected by age, gender and BMI (adjusted effect as assessed with regression analysis).

Serum total GHR and OB levels

The preprandial levels of GHR, OB and the GHR/OB ratio were significantly higher in patients compared to controls (P<0.05) (Fig. 1). Postprandially (patient group) there was a decrease in GHR and an increase in OB levels which were not statistically significant; however the postprandial GHR/OB ratio was found to be significantly lower (Fig. 2). Moreover, the postprandial levels of GHR and the GHR/OB ratio (but not OB) were significantly lower in the patients who had delayed GET (P<0.05) (Fig. 3). Pre- and postprandial levels of GHR, OB were not significantly affected by patient/control characteristics (adjusted effect, regression analysis).

Figure 1 Preprandial levels of GHR, OB and GHR/OB ratio in patient and control groups

GHR, ghrelin; OB, obestatin; GHR, OB and GHR/OB values represent mean values

Figure 2 Pre- and postprandial levels of GHR, OB and GHR/OB ratio in patient group

GHR, ghrelin; OB, obestatin; GHR, OB and GHR/OB values represent mean values

Figure 3 Pre- and postprandial levels of GHR, OB and GHR/OB ratio in patients with normal and delayed gastric emptying time (GET)

GHR, ghrelin; OB, obestatin; GHR, OB and GHR/OB values represent mean values