Exocrine pancreatic function and dynamic of digestion after restrictive and malabsorptive bariatric surgery: a prospective, cross-sectional, and comparative study

1 Research Article EXOCRINE PANCREATIC FUNCTION AND DYNAMIC OF DIGESTION AFTER RESTRICTIVE AND MALABSORPTIVE BARIATRIC SURGERY: A PROSPECTIVE, CROSS-SECTIONAL, AND COMPARATIVE STUDY. Laura Uribarri-Gonzaleza, Laura Nieto-Garcíaa, Aurelio Martis-Sueirob, J. Enrique Dominguez-Muñoza a Department of Gastroenterology, Health Research Institute (IDIS), University Hospital of Santiago de Compostela, Spain. b Department of Endocrinology, University Hospital of Santiago de Compostela, Spain. Short Title: Pancreatic function after bariatric surgery. Corresponding Author: Full name Laura Uribarri-Gonzalez Department of Gastroenterology Hospital Universitario de Santiago de Compostela Rúa da Choupana, s/n Santiago de Compostela, 15706, Spain Tel: 981950000 E-mail: luribarrigon[email protected] Number of Tables: 3 Number of Figures: 3 Word count: Keywords: Breath test, pancreatic exocrine insufficiency, malabsorption, gastric bypass, sleeve gastrectomy, duodenal switch. 2 Abstract 1 Introduction: Gastrointestinal anatomical changes after restrictive and malabsorptive 2 bariatric surgery lead to important disturbances in the process of digestion and absorption 3 of nutrients, and could lead to pancreatic exocrine insufficiency (PEI). Aim of the present 4 study was to evaluate and to compare pancreatic function and the dynamic of digestion and 5 absorption of nutrients after restrictive and malabsorptive bariatric surgical procedures. 6 Methods: A prospective, observational, cross-sectional, comparative study of patients after 7 sleeve gastrectomy (SG), Roux-en-Y gastric bypass (RYGB) and biliopancreatic diversion with 8 duodenal switch (BPD/DS) was carried out. Non-operated obese patients were included as 9 control group. Pancreatic function and the dynamic of digestion and absorption of nutrients 10 were evaluated by the 13C-mixed triglyceride (13C-MTG) breath test. Six-hour 13C-cumulative 11 recovery rate (13C-CRR), 13C exhalation peak and 1-hour maximal 13C-CRR were calculated. 12 Results: 105 patients were included (mean age 49.8 years, 84 females). Six-hour 13C-CRR was 13 significantly reduced after BPD/DS (p<0.001), but not after SG and RYGB. PEI was present in 14 80% of patients after BPD/DS, 8.3% of patients after RYGB and 4.3% of patients after SG. 15 Compared to non-operated controls, digestion and absorption of nutrients tend to occur 16 earlier after SG, whereas it is delayed after RYGB and mainly after BPD/DS (p<0.001). 17 Conclusion: Bariatric surgery alters significantly the dynamic of the digestive process. PEI is 18 very common after BPD/DS, frequent after RYGB and less frequent after SG. This information 19 is clinically relevant since PEI is a treatable condition associated with symptoms, nutritional 20 deficiencies and complications.21 3 Introduction 22 Pancreatic exocrine insufficiency (PEI) is defined as the primary or secondary disturbance of 23 the exocrine pancreatic function leading to maldigestion [1]. PEI can be the consequence of 24 pancreatic diseases as well as extrapancreatic conditions, mainly gastrointestinal and 25 pancreatic surgery [2-4]. 26 The prevalence of obesity has increased dramatically over the last years, and the relevance 27 and frequency of bariatric surgery is steadily increasing [2,3]. Surgical approaches for 28 bariatric surgery can be divided into restrictive (e.g., gastric band and sleeve gastrectomy 29 [SG]), malabsorptive (Roux-en-Y intestinal bypass), and mixed procedures (Roux-en-Y gastric 30 bypass [RYGB] and biliopancreatic diversion with duodenal switch [BPD/DS]). 31 Alterations of the digestive process induced by bariatric surgery are complex and frequently 32 overlapping among different procedures. Whereas SG is considered a pure restrictive 33 procedure, it is associated with disturbed postprandial neural stimulation of pancreatic 34 secretion [5]. Moreover, the loss of the antrum crushing capacity of the food impairs the 35 digestive ability of pancreatic juice. On the other hand, not only digestion and absorption of 36 nutrients in the small intestine but gastric physiology is altered after RYGB and BPD/DS [6]. 37 Exocrine pancreatic physiology is also disturbed after bariatric surgery. Cholecystokinin (CCK) 38 is secreted by I cells that are mainly located in the mucosa of the duodenum and proximal 39 jejunum. Since the proximal gut is excluded from the nutrient passage following RYGB and 40 BPD/DS, postprandial CCK release and, therefore, biliopancreatic secretion is expected to be 41 reduced after surgery. However, this is controversial since some authors have reported an 42 increased CCK release after RYGB in response to a mixed meal [7]. Mixed procedures lead in 43 4 addition to an asynchrony between gastric emptying of nutrients and bilio-pancreatic 44 secretion, limiting the appropriate mixing of nutrients with digestive enzymes [8]. All these 45 alterations explain the potential development of PEI after bariatric surgery, which may play a 46 role in weight loss and malabsorption of nutrients in patients after RYGB and BPD/DS [9]. 47 PEI of any etiology is a well-known cause of abdominal and bowel symptoms, and nutritional 48 deficiencies [10,11]. In addition, PEI is associated with complications such as osteoporosis, 49 sarcopenia, cardiovascular events, and even mortality [12-14]. Diagnosis of PEI in clinical 50 practice is limited by the lack of simple and accurate methods. The coefficient of fat 51 absorption (CFA) after fecal fat quantification is the method of choice for fat malabsorption, 52 but it is cumbersome and unpleasant. Quantification of fecal elastase-1 (FE-1) is a simple and 53 non-invasive test of pancreatic secretion, but its accuracy after GI and pancreatic surgery is 54 limited [15]. The 13C-mixed triglyceride (MTG) breath test is a simple and accurate 55 alternative to CFA that does not only quantify pancreatic function, but also evaluate the 56 dynamic of digestion and absorption of nutrients, which is of special interest in patients after 57 bariatric surgery [16]. MTG is specifically digested by pancreatic lipase, which is the limiting 58 factor of the test [17]. 13CO2 recovery at each time point represents therefore the amount of 59 chyme that has been digested by pancreatic secretion and absorbed in the gut at that time, 60 thus providing with dynamic information of the digestive process. 61 Main aim of the present study was to evaluate pancreatic function and the dynamic of 62 digestion and absorption of nutrients after restrictive and malabsorptive bariatric surgical 63 procedures. 64 65 5 Materials and Methods 66 Study design 67 A prospective, observational, cross-sectional, single center, comparative study was designed. 68 The study was conducted at the Departments of Gastroenterology and Endocrinology of the 69 University Hospital of Santiago de Compostela, Spain. Patients older than 18 years who 70 underwent bariatric surgery, either RYGB, SG or BPD/DS, for morbid obesity were considered 71 for inclusion. Morbid obese patients in the waiting list for bariatric surgery were included as 72 a control group. Previous GI or pancreatic surgery, chronic GI diseases (celiac disease, 73 diabetic gastroparesis), any severe concomitant disease limiting life expectancy, and inability 74 or refusal to perform the study-related procedures or to sign the written informed consent 75 were considered exclusion criteria. Demographic and clinical data were recorded at 76 inclusion. Pancreatic function was evaluated by the 13C-MTG as previously described [16]. In 77 short, 250 mg of 13C-MTG was orally given together with a solid test meal containing 16 g of 78 fat after overnight fasting. Metoclopramide 10 mg was given orally 20 minutes before test 79 meal ingestion. Naturally 13C-enriched foods (corn, pineapple, broccoli, and sugarcane) were 80 avoided for at least 48 hours before to avoid interferences with the breath test. Smoking 81 was prohibited from the night before until the end of the test. In addition, patients were 82 asked to remain fasting and seated during the test, although limited physical activities as 83 visiting the toilet were allowed. Breath samples were collected twice just before test meal 84 ingestion (basal) and at 15-minute intervals for 6 hours thereafter into 10-mL glass tubes 85 (Exeteiner; Labco Limited, UK) using a disposable plastic straw. Subjects exhaled into the 86 straw during a complete expiration, and the tube was capped immediately thereafter. The 87 quotient 13CO2/12CO2 was quantified in these end-expiratory breath samples by isotope ratio 88 6 mass spectrometry (Breath MAT plus; Thermo Finnigan, Germany). The total cumulative 89 recovery rate of 13C over the 6-hour time (13C-CRR %) was calculated as previously described 90 and used as the result of the test [16]. For the purposes of the study, PEI was defined as a 6-91 hour 13C-CRR below the 5th percentile of the control group. The time and height of the 13C 92 exhalation peak (% dose/15 min) and the one-hour period of maximal 13C recovery rate (one-93 hour 13C-CRR, % dose/h) were calculated as markers of the dynamic process of digestion and 94 absorption of nutrients. 95 Data analysis 96 Categorical variables are shown as absolute values and percentages and compared using χ² 97 test or Fischer’s exact test as appropriate. Continuous data are presented as mean ± 98 standard deviation (SD) or median and interquartile range (IQR) for normally or non-99 normally distributed data, respectively. Data distribution was analyzed by the Shapiro–Wilk 100 test. Student-t test or Mann-Whitney test for comparing two groups, and ANOVA or the 101 Kruskal-Wallis test for comparing more than two groups, were used as appropriate according 102 to data distribution. 103 Ethical Aspects 104 The study was approved by the Clinical Research Ethics Committee of the Galician Ministry 105 of Health (Comité Ético de Investigación Clínica de Galicia, Consellería de Sanidade, 106 www.ceic.sergas.es) with the approval number AD-04-014. All patients provided written 107 informed consent to the study. The study protocol conforms to the ethical guidelines of the 108 1975 Declaration of Helsinki and the study was conducted in accordance with the 109 Declaration of Helsinki, its amendments, and Good Clinical Practice guidelines. 110 7 111 Results 112 A total of 122 patients who had undergone restrictive or malabsorptive bariatric surgical 113 procedure for morbid obesity were considered for inclusion. Twenty-seven patients were 114 excluded due to the presence of other GI diseases (n = 15) or unwillingness to participate in 115 the study (n = 12). Finally, 95 patients were included (figure 1). Ten morbid obese patients 116 from the waiting list of bariatric surgery were included as controls. Mean age was 49.8 years 117 (range 22-70 years), and 84 patients were female (80%). Thirty-six patients (37.9%) had 118 undergone RYGB, 36 (37.9%) BPD/DS and 23 (24.2%) patients SG. Age and gender 119 distribution of operated patients and controls were similar. Mean time from bariatric 120 surgery to inclusion into the study was longer after BPD/DS than after RYGB and SG (Table 121 1). 122 Breath test was well tolerated in all patients and controls, and no side effects were recorded. 123 13C-CRR after 6 hours was similar in patients after RYGB (44.0 ± 9.6 %), SG (47.6 ± 9.3 %) and 124 non-operated controls (45.0 ± 6.1 %), whereas it was significantly reduced in patients after 125 BPD/DS (25.6 ± 13.5 %) (BPD/DS vs each other group, p <0.001). The 5th percentile of the 6-126 hour 13C-CRR in the control group was 32.9%. A 6-hour 13C-CRR below that limit was present 127 in 78.8% of patients after BPD/DS, 8.3% of patients after RYGB and 4.3% of the patients after 128 SG (p<0.01). The 13C exhalation peak was significantly reduced in patients after BPD/DS (1.94 129 ± 0.86 % dose/15min) compared to patients after RYGB (2.85 ± 0.59 % dose/15min), SG (3.15 130 ± 0.63 % dose/15min) and controls (3.06 ± 0.49 % dose/15min) (p<0.001), and in patients 131 after RYGB compared to those after SG (p=0.05) (figure 2). Similarly, the maximal one-hour 132 13C-CRR was significantly lower in patients after BPD/DS than in those after RYGB, SG or 133 8 controls (p<0.001) (figure 2). 134 Dynamic of the digestion process according to the 13C exhalation curves after different 135 bariatric surgical procedures and in control patients is shown in figure 3. In non-operated 136 controls, the 13C exhalation peak occurred at 127 min (IQR 120-157 min) (figure 4). Digestion 137 and absorption of nutrients occurred earlier in patients after SG (13CO2 exhalation peak at 138 105 min, IQR 90-135 min) than in those after RYGB (13CO2 exhalation peak at 157 min, IQR 139 94-240 min) (p<0.001). The 13CO2 exhalation peak is significantly delayed in patients after 140 BPD/DS (330 min, IQR 270-360 min, p<0.001 vs any other group) (figure 4). 141 142 Discussion/Conclusion 143 The present study shows that exocrine pancreatic function and the dynamic of digestion and 144 absorption of nutrients are altered differently after different bariatric surgical procedures. 145 The proportion of ingested fat digested by pancreatic lipase over 6 postprandial hours is 146 similar in patients after SG and RYGB than in non-operated obese patients. Dynamic of 147 digestion is however different. Digestion tends to occur earlier after SG, whereas it is 148 delayed after RYGB compared to non-operated obese subjects. Digestion and absorption of 149 fat is significantly reduced and delayed in patients after BPD/DS compared to SG, RYGB and 150 non-operated patients. The development of PEI is very frequent after BPD/DS, whereas it 151 occurs in about one out of ten patients after RYGB, and in one out of twenty patients after 152 SG. This is important since PEI causes malabsorption-related symptoms, nutritional 153 deficiencies, complications, and even increased mortality that can be limited by the 154 appropriate use of oral pancreatic enzymes. 155 9 Bariatric surgery is the most effective treatment for morbid obesity and metabolic related 156 conditions (diabetes mellitus, arterial hypertension, and hypercholesterolemia). Surgical 157 procedures may reduce gastric capacity to accommodate food (gastric banding and SG), thus 158 limiting food intake, may disturb digestion and absorption of nutrients (intestinal bypass) or 159 may alter both gastric accommodation, and digestion and absorption of food (RYGB and 160 BPD/DS). Weight loss depends on the degree of gastric restriction and the severity of 161 intestinal malabsorption of nutrients associated with the different procedures [18]. 162 Exocrine pancreatic function is disturbed after upper gastrointestinal surgery [19]. Exclusion 163 of the duodenum and proximal jejunum from the passage of nutrients after gastrectomy or 164 gastric bypass may alter postprandial release of CCK and secretin, thus impairing stimulation 165 of pancreatic secretion. Neurally mediated stimulation of the exocrine pancreas is in 166 addition altered after surgical disruption of antro-fundic and duodenogastric vagal reflexes. 167 Finally, Roux-en-Y anatomical reconstruction is associated with a postcibal asynchrony 168 between the gastric emptying of nutrients and the biliopancreatic secretion, all together 169 leading to PEI [20]. The degree of exocrine pancreatic dysfunction is different among 170 different surgical procedures and factors like disruption of the gastrointestinal passage, 171 single or Roux-en-Y anastomosis and length of biliopancreatic and alimentary limbs play a 172 role [21]. 173 Evaluation of the exocrine pancreatic function is not an easy task. Fat maldigestion is at best 174 quantified by the CFA after 72h stool collection. This method is however cumbersome and 175 unpleasant both for patients and people processing the samples in the lab. FE-1 is a simple 176 and widely available test, but its accuracy in patients after surgery is limited [2,13,22]. In 177 addition, none of these two methods allows evaluating the dynamic of the digestive process, 178