Introduction

Roux-en-Y gastric bypass (RYGB) is the second commonest bariatric procedure worldwide [1] and commonest in the UK [2]. Bariatric procedures are traditionally classified as those that physically restrict food intake (the so-called restrictive procedures), those that prevent absorption of consumed calories (malabsorptive), and those that work through a combination of these mechanisms. RYGB is classified as a combined restrictive and malabsorptive procedure. Some authors argue that this classification of bariatric procedures is unscientific and may lead to ineffective and potentially dangerous surgical modifications [3].

This has led to calls to abandon the abovementioned classification of bariatric procedures [4]. Others [5] have suggested that restriction and malabsorption make little contribution to the overall effect of gastric bypass and that other mechanisms might be at play. Many authors have attempted to study absorption of one or more macronutrient after RYGB, but there is however to date no published review systematically examining the contribution of malabsorption to weight loss following RYGB.

One also recognizes that there may be differences between proximal and distal RYGB [6]. Since proximal gastric bypass is the commoner variation of RYGB, this review focuses on understanding the contribution of malabsorption towards weight loss with proximal RYGB in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.

Methods

An online search of PubMed was carried out using key-words like “bariatric surgery,” “gastric bypass,” “Roux en Y Gastric Bypass,” “Malabsorption,” “Malabsorptive,” “Absorption,” “Carbohydrate,” “Fat,” and “Protein” to identify all articles on contribution of macronutrient malabsorption to outcomes associated with RYGB. Articles were also identified from references of relevant articles. Last of these searches was carried out on April 1, 2017.

We excluded non-human studies, studies only evaluating micronutrient malabsorption, studies on distal RYGB, and review articles with no new data. We also excluded studies that presented data on protein-calorie malnutrition after RYGB without specifically exploring the contribution of malabsorption.

A total of 22 articles were included as they made a contribution to our understanding of malabsorption with proximal RYGB. Figure 1 gives a PRISMA flow chart for article selection. Since the quality of the data would not lend itself to a meta-analysis, we then performed a full narrative review of these studies. We attempt to study malabsorption of each macronutrient—carbohydrate, protein, and fat.

Fig. 1
figure 1

PRISMA flow chart for article selection

Full size image

Results

A total of 22 articles studied the role of malabsorption towards weight loss with RYGB. We present here the salient characteristics of these studies depending on the macronutrient studied.

Studies on Fat Absorption After RYGB

Table 1 [7,8,9,10,11,12,13,14] lists all eight studies on malabsorption of fat after RYGB. Important characteristics of these studies are listed below.

  1. 1.

    Pihlajamäki et al. [7] studied cholesterol metabolism including markers of cholesterol synthesis and absorption preoperatively and 1 year after surgery in 29 patients after RYGB and 26 patients who underwent gastric banding (GB). Both operations resulted in a decrease in serum levels of cholesterol synthesis markers by 12.0–28.0%, but a decrease in cholesterol absorption markers was only observed after RYGB. Authors suggested that reduced levels of plant sterols, a surrogate marker they used to study cholesterol absorption rather than studying cholesterol absorption directly, could be due to a lower intake. It was suggested that studies using labeled isotopes might help clarify the situation.

  2. 2.

    Carswell et al. [8] studied seven RYGB (30 ml gastric pouch, 150 ml alimentary limb (AL), and 70 cm biliopancreatic limb (BPL)) patients 8–20 months (mean 10) after surgery and compared a number of parameters with patients undergoing gastric banding (n = 6), duodenal switch (n = 5), and obese controls (n = 7) with body mass index >30 kg/m2. Authors found that the RYGB patients had higher fecal fat excretion than the obese controls (p = 0.033). Fecal fat concentration was 18.5 mmol/day (12.5–28.9) in obese controls as opposed to 49.4 mmol/day (38.6–83.3) in RYGB patients (p = 0.033). Fecal elastase was lower in RYGB vs obese (p = 0.002) controls. Plasma citrulline levels (indicative of functional enterocyte mass) were similar in RYGB and obese control groups. Fecal calprotectin was higher in the RYGB group compared to obese controls (p = 0.016). Authors interpreted the data as showing “evidence of impaired pancreatic exocrine function, some residual inflammation and some fat malabsorption in RYGB.”

  3. 3.

    Odstrcil et al. [9] found that the average fat intake was 156 g/day before bypass, 50 g/day 5 months after bypass, and 82 g/day 14 months after bypass. At 5 months after bypass, the total reduction in fat absorption averaged 108.6 g/day, of which 11.2 g/day (10%) was due to malabsorption and 97.4 g/day (90%) was due to a reduction in dietary fat intake. At 14 months, the total reduction was 85.4 g/day, of which 16.9 g/day (20.0%) was due to malabsorption and 68.5 g/day was due to reduced intake. Authors also observed a positive correlation between the length of biliopancreatic limb and reduction in fat absorption due to malabsorption (r = 0.84, p = 0.001). The coefficient of fat absorption (CFA) decreased from 92.1% before bypass to 68.1% at 14 months after surgery.

  4. 4.

    Moreland et al. [10] found that 24 out of 26 patients had steatorrhoea 1 year after “long limb” RYGB compared to 12 out of 26 patients preoperatively. The mean combined length of the AL and BPL was 223 cm, and the mean common channel was 597 cm. Authors found that the fat intake in prebypass patients averaged 173 g/day (104–276) and fecal fat output was 7.6 g/day (2.2–17.7). The average CFA was 95.6%. After RYGB, the average fat intake was reduced to 81.0 g/day and the average coefficient of fat absorption came down to 75.3%. Because of the reduction in CFA, 24/26 patients had steatorrhoea after RYGB despite a reduction in total fat intake. The average fecal fat output was approximately 17.0 g/day (range 2.2–46.8 g/day) after surgery.

  5. 5.

    Griffo et al. [11] studied the effect of RYGB (n = 10) and sleeve gastrectomy (n = 15) in 25 obese type 2 diabetes mellitus (T2DM) patients (mean age 46 ± 8 years; mean BMI 44 ± 7 kg/m2). Lipids were evaluated 3 h after a standardized 304 kcal liquid meal containing 9 g of fat, before and 2 weeks after surgery. RYGB was constructed with a 40-ml gastric pouch, 100–150-cm AL, and 100–150-cm BPL. The area under the curve (AUC) for plasma glucose and triglyceride was significantly lower after surgery. Authors did not find any significant difference between two procedures for plasma triglyceride AUC even though the levels were lower for RYGB than for sleeve gastrectomy. Authors remarked that the two procedures appeared “to exert a similar influence on Postprandial Triglyceride metabolism” but acknowledged that the numbers were too small for a meaningful comparison of the effect of surgery on postprandial lipemia between two procedures.

  6. 6.

    Borbély et al. [12] found that 19.0% of patients undergoing proximal gastric bypass had features of exocrine pancreatic insufficiency. Interestingly, 81.0% of proximal RYGB patients did not have any evidence of exocrine pancreatic insufficiency. Authors did not study weight loss difference in the two groups.

  7. 7.

    Kumar et al. [13] studied fecal fat excretion in nine RYGB patients (15–20-ml pouch, 100–150-cm AL, and 30–60-cm BPL) and found an increase in fecal fat at 6 months (6.3 ± 3.1 g/72 h, p = 0.027) and at 12 months (7.2 ± 4.4 g/72 h, p = 0.015) compared to that before surgery (4.0 ± 3.8 g/72 h) despite a decrease in fat intake.

  8. 8.

    Forbes et al. [14] studied plasma phospholipid fatty acid profile in 13 RYGB patients compared with five patients who underwent gastric banding. Dietary intake of fats decreased equally at 1 and 6 months in both the groups of patients. The concentration of plasma free fatty acid increased (+62%) in the RYGB group at 1 month following surgery, but values came down to baseline by 6 months after surgery. Measures of essential fatty acid status showed improvement in RYGB group, at 1 and 6 months following surgery; the levels did not change in participants who underwent GB.

Table 1 Studies on fat absorption after RYGB
Full size table

Interpretation

There appears to be some fat malabsorption after RYGB. CFA decreases to approximately 68.0–75.0% after RYGB compared to 92.0–95.0% preoperatively [9, 10]. Malabsorption accounts for approximately 10.0–12.0 g of fecal fat loss daily after surgery [9, 10].

Studies on Protein Absorption After RYGB

Table 2 [9, 15,16,17] lists the four studies on malabsorption of proteins after RYGB. Important characteristics of these studies are listed below.

  1. 1.

    Bojsen-Møller et al. [15] studied leucine and phenylalanine kinetics in response to a test meal, before and 3 months after RYGB in nine obese, glucose-tolerant patients. Basal plasma leucine concentration was unchanged, but postprandially there was a 1.7-fold rise in peak concentration and an unchanged total AUC in comparison with presurgery values. The time to peak was much shorter after surgery (65 vs 173 min). Peak plasma phenylalanine levels were also higher, but basal levels were lower after surgery. Authors suggested that these findings indicate accelerated protein digestion and amino acid absorption resulting in faster and higher but more transient postprandial elevation of plasma amino acids after RYGB. The study concluded that “protein digestion was not impaired after RYGB.”

  2. 2.

    Khoo et al. [16] studied the metabolic response to a standard 230-kcal test meal in 10 T2DM obese subjects pre and 10–14 days post RYGB. A matched group of volunteers maintained on a calorie restriction served as controls. Significant reductions in fasting proline, histidine, valine, and phenylalanine and molar sum of branched-chain amino acid (AA), aromatic AA, and total AA were observed after RYGB but not with calorie restriction. Significant reductions of 20–30% were observed in multiple AA, molar sum of branched-chain AA, aromatic AA, and total AA in the RYGB, 2 h after the test meal, but not the calorie restriction group. The AA trajectories were not different between groups in the preintervention period, which authors said was due to “a unique effect of RYGB on the trajectories of these amino acids with meal challenge.” The post-RYGB trajectories exhibited a steep increase in the first phase (0–60 min) with a rapid decline in the second phase (60–120 min) after the test meal. Authors attributed this to “improved peripheral uptake and utilization” of AA because the postprandial trajectories of amino acids tracked closely with “either increased insulin secretion or improved insulin action, or both.”

    The study also found that multiple intermediates and end points of fatty acid oxidation like acetylcarnitine, total ketones, β-hydroxybutyrate, and long-chain acylcarnitines tended to be higher in the fasted state in RYGB subjects. Authors interpreted these results to indicate a higher rate and efficiency of fatty acid oxidation after RYGB.

  3. 3.

    Laferrère et al. [17] compared metabolic parameters in obese diabetics 1 month after RYGB and a matched group who lost 10 kg on a 1000-kcal/day diet. The weight loss was similar in both groups as was the rise in ketones and β-hydroxybutyrate. Authors found that total AAs and branched-chain AAs decreased significantly after RYGB, but not after dietary intervention. Plasma concentrations of total AAs and branched-chain AAs decreased by 19.7% (p = 0.008) and 38.3% (p < 0.001), respectively, following RYGB, but only by 12.8% (- = 0.025) and 12.6% (p = 0.083), respectively, following calorie restriction. Branched-chain AAs were lower after RYGB compared to diet (p < 0.001). In addition to the branched-chain AAs leucine, isoleucine, and valine, the aromatic AAs phenylalanine and tyrosine, as well as ornithine, citrulline, and histidine, all decreased after RYGB. This study did not report on AUC for AA so we cannot be certain of the reason for lower circulating AA levels. It could be due to lower protein ingestion, impaired absorption, or increased catabolism. Protein intake was 70 g per day in the RYGB group compared to 100 g per day in the diet group. Authors further observed an increase in two key branched-chain AA catabolic enzymes and several other changes that suggested an “enhanced oxidation of Branched Chain AA.”

    Odstrcil et al. [9] found that protein malabsorption accounted for 13% of the total reduction in protein absorption at both 5 and 14 months after bypass for patients who did not take any supplements. For these seven patients who did not take any protein supplement (excluding two who did), the coefficient for protein absorption decreased from 81.8% before surgery to 75.9% 14 months after bypass (p = 0.136).

Table 2 Studies on protein absorption after RYGB
Full size table

Interpretation

Coefficient for protein absorption decreases by approximately <5.0% after RYGB [0–10], and overall, there does not appear to be much impairment of protein digestion and absorption after RYGB [15].

Studies on Carbohydrate Absorption After RYGB

Table 3 lists the four studies on malabsorption of proteins after RYGB. Important characteristics of these studies are listed below [9, 18–28].

  1. 1.

    Andalib et al. [18]: This is a retrospective study of 63 individuals who underwent breath hydrogen testing after oral glucose (to study carbohydrate malabsorption) to evaluate abdominal symptoms following RYGB. The study found that 51 (81%) patients had a late rise (≥45 min) in breath hydrogen or methane, suggesting unabsorbed glucose reaching terminal ileum or colon (i.e., glucose malabsorption), and 46 (90%) of these 51 subjects also had an early rise (≤30 min) in breath hydrogen or methane indicating “upper gut bacterial overgrowth.” Authors also found glucose malabsorption to be more frequent in subjects with upper gut bacterial overgrowth compared to subjects with no evidence for bacterial overgrowth (<0.001).

    The reason for including this study in our review was because authors attempted to study the effect of glucose malabsorption on weight loss. In the 26 individuals with evidence for glucose malabsorption, the average percentage (±SEM) of excess body weight lost at 3 years was 66% (±3.8). In the seven subjects with no findings supporting glucose malabsorption, the average percentage (±SEM) of excess body weight lost at 3 years was 56% (±8.1). There was no significant increase in the percentage of excess weight loss in subjects with evidence of glucose malabsorption (unpaired t test = 0.24). Authors mentioned that this study was not adequately powered to study the effect of glucose malabsorption on weight loss

  2. 2.

    Nguyen et al. [19] studied glucose absorption in 10 non-diabetic patients following RYGB and 10 healthy volunteers. RYGB patients underwent two studies, separated by at least 7 days with either an oral glucose drink consumed over 3 min (50 g glucose in 150 ml water mixed with 3 g 3-omethyl-d-glucopyranose (3-OMG) or (ii) the same solution infused directly into the proximal Roux-limb over 50 min (at 4 kcal/min). The healthy subjects received a 4-kcal/min intraduodenal glucose infusion over 50 min. In response to the small intestinal glucose infusion, both peak and integrated AUC(0–120 min) plasma 3-OMG levels were higher in RYGB subjects than in healthy volunteers (p = 0.06). Among RYGB subjects, both absolute and integrated AUC(0-120min) plasma 3-OMG concentrations after oral glucose were lower than after glucose infusion and were comparable to those after glucose infusion in healthy subjects. Authors hypothesized that “greater intestinal glucose absorption might represent small intestinal adaptation after RYGB.”

  3. 3.

    Nguyen et al. [20] collected intestinal biopsies for mRNA expression of STR (sweet taste receptor) T1R2 and GTs (glucose transporters) SGLT-1 and GLUT2 from 11 non-diabetic RYGB, 13 non-diabetic obese, and 11 healthy subjects, at baseline and following a 30-min small intestinal glucose infusion (30 g/150 ml water with 3 g 3-O-methyl-d-glucopyranose (3-OMG)). Blood glucose, plasma 3-OMG, and insulin were measured for 270 min. Authors found that the baseline levels of SGLT-1 and GLUT2 transcripts were higher in RYGB than in either morbidly obese or healthy subjects (p < 0.001). The rate of increase in plasma 3-OMG and peak concentrations was higher in RYGB and morbidly obese subjects than in healthy controls. Both absolute (p < 0.001) and integrated AUC(0-270min) plasma 3-OMG concentrations were higher in RYGB patients than in healthy subjects (p = 0.05). Authors concluded that the finding of upregulation of intestinal GTs associated with increased glucose absorption in RYGB patients was evidence of adaptation of the small intestine in these patients to prevent carbohydrate malabsorption.

  4. 4.

    Anderwald et al. [21] studied glucose absorption and other parameters in six patients before and 7 ± 1 months after RYGB and compared results with obese and lean controls. Authors found, on an oral glucose tolerance test, that glucose absorption was not different before and after surgery, as well as between both controls. Authors were surprised with the findings as “they rather point away from a pronounced malabsorption after RYGB.”

  5. 5.

    Odstrcil et al. [9] found that there was almost no carbohydrate malabsorption after RYGB and the carbohydrate absorption coefficients remained unchanged after surgery. This in authors’ view was at least partly due to absorption of carbohydrates in the AL. This may also be due to upregulation of intestinal glucose transporters in RYGB patients [L].

  6. 6.

    Jacobsen et al. [22] studied glucose absorption and metabolism before and at 3 months following RYGB in three men and nine women. One of the key findings of this study was that the percentage of oral glucose absorbed into the systemic circulation increased significantly from 54 ± 4 to 80 ± 2% (p = 0.004) before and after RYGB (p = 0.004 for IAUC). Approximately 27.0 and 40.0 g of the ingested glucose load was absorbed within the 4-h study period before and after RYGB.

  7. 7.

    Camastra et al. [23] studied glucose kinetics and hormonal response to a test meal before and 1 year following RYGB in 10 T2DM and 11 non-diabetic patients. Authors found that most of the oral glucose was absorbed within the first hour. The AUC(0–60 min) was significantly higher (p = 0.04) postoperatively compared to the preoperative values in both groups of patients.

  8. 8.

    Falkén et al. [24] studied glucose metabolism in 12 non-diabetic obese subjects before and at 3 days, 2 months, and 1 year after RYGB. AUC(0–180 min) for glucose was similar postoperatively at all time periods compared to values pre surgery. Authors also found that plasma glucose peaked around 30 min both before and after the operation, but the whole curve was shifted leftward with an earlier rise in plasma glucose and a steeper decline after the operation.

  9. 9.

    Rodieux et al. [25] studied glucose kinetics in eight RYGB patients 9–48 months after surgery in comparison with gastric banding patients and age- and weight-matched controls. Authors found a significantly earlier (p < 0.01) and exaggerated (p = 0.0001) rise in blood glucose in the RYGB group compared to both the other groups. Incremental AUC was, however, similar among the three groups as plasma glucose also returned to baseline significantly earlier in the RYGB group (p = 0.0001). Authors did not observe any difference among the three groups for the AUC for total exogenous glucose appearance in the systemic circulation over the 4-h period following ingestion. The study found that 78.5% of the oral glucose load was absorbed in the RYGB group compared to 73.5% in the banding group and 76.6% in the control group.

  10. 10.

    Wang et al. [26] studied seven patients 1 year post RYGB and found no significant difference in absorption of d-xylose compared to that before surgery. Authors concluded that these findings suggested “an absence of carbohydrate malabsorption” 1 year after RYGB.

  11. 11.

    Wilms et al. [27] studied levels of circulating glucose among other things following a test liquid meal in a cohort of 10 women who had undergone RYGB 41.9 months before and compared the findings with 8 severely obese women and 10 lean women. Authors found that the rise in blood levels was markedly higher in the RYGB group in comparison with both the control groups (p = 0.001). AUC(0–90 min) for blood glucose was significantly higher in the RYGB group compared to lean women; the difference was not significant with obese control.

  12. 12.

    Savassi-Rocha et al. [28] studied 16 RYGB patients (40-ml gastric pouch, 110-cm AL, and 50–80-cm BPL). The mean length of the BP limb was 62.3 cm (50–80 cm). The mean jejunoileal length was 668.9 cm (485–787 cm), and the mean length of the common limb was 496.6 cm (298–610 cm). Authors found a mean mannitol urinary excretion of 7.47% at 1 month after surgery compared to 11.41% before surgery (p = 0.03). Interestingly, however, at 6 months, it increased to 10.66%. Authors suggested that loss of absorptive surface in BP limb and mucosal atrophy in AL (due to altered microbiota) might have been responsible for decreased mannitol absorption and excretion at 1 month. They further surmised that intestinal adaptation might be responsible for the return of mannitol excretion to near preoperative values by 6 months. There was no significant increase in lactulose excretion at 1 and 6 months for the whole cohort, but six patients showed a considerable increase in lactulose excretion rate of 4–73 times. Authors extrapolated that a “1.0-m long AL would have little or no influence on long-term weight loss after RYGB” and that a subgroup of patients who undergo RYGB show a “pronounced increase in intestinal permeability.”

Table 3 Studies on carbohydrate absorption after RYGB
Full size table

Interpretation

The coefficient of carbohydrate absorption remains largely unchanged after RYGB [9, 10], and there is no carbohydrate malabsorption after RYGB [26]. At the same time, there are significant changes to glucose kinetics with a higher peak observed quicker after meals and a lower baseline level.

Studies on Total Calorie Malabsorption

There are only two studies assessing the contribution of macronutrient malabsorption towards weight loss with RYGB in some detail. Both suggest that the total contribution of malabsorption towards energy intake after RYGB is approximately 10.0–11.0%, most of which comes from reduced coefficients for fat absorption.

Odstrcil et al. [9] found that combustible energy intake averaged 3754 kcal/day before bypass, 1556 kcal/day 5 months after bypass (a 59% reduction), and 2241 kcal/day 14 months after RYGB (a 40% reduction from prebypass energy intake). The reduction in energy absorption due to malabsorption averaged 124 kcal/day, 5 months (6.0%) after bypass and 172 kcal/day 14 months (11.0%) after bypass. Reduced energy absorption due to a reduction in dietary energy intake was much greater, averaging 2062 kcal/day at 5 months after bypass and 1418 kcal/day at 14 months after bypass.

Moreland et al. [10] found that the total energy absorption before the bypass was 3781 cal/day. The average total absorption 1 year after RYGB was reduced by 1944 cal/day. Authors estimated that 182 cal was reduced due to malabsorption and 1762 cal was reduced due to lack of intake.

Discussion

Gastric bypass is a hugely successful weight loss intervention with demonstrated results over a long period of time, but some patients can suffer adverse consequences. For bariatric surgery, and gastric bypass to become more acceptable, we do not need to spend much time thinking about how to improve weight loss further as it is already several times better and longer-lasting than any available alternative. All we need to do is to continue to improve the safety of surgical interventions.

By definition, gastric bypass bypasses a large part of the stomach and some proximal intestine. From a recent systematic review of small bowel limb lengths with RYGB [6], we know that bypassing a cumulative length of 150 cm as AL and BPL is sufficient to achieve most of its desired benefits and bypassing longer lengths of small bowel is not associated with a proportionate increase in benefits. On the contrary, it may be associated with a higher risk of micronutrient as well as macronutrient malnutrition.

Despite this, many surgeons bypass much longer lengths of small bowel for primary RYGB as well as those seeking revisions [29], presumably to maximize its gains. The perception that malabsorption is one of the key mechanisms of action behind the success of gastric bypass is a key reason behind such surgical modifications. It is hence important to understand how important malabsorption is to the functioning of the RYGB.

This systematic review shows that most of the benefits of RYGB come from a reduction in calorie intake, and in the early period after surgery, malabsorption contributes no more than 10.0–11.0% to the entire calorie deficit. It is conceivable that bowel adaptation would reduce this contribution even further in the long term. Nguyen et al. [20] found that intestinal glucose transporters were upregulated after RYGB in association with increased glucose absorption and suggested that their data provided “evidence of a molecular adaptation of the intestine in RYGB patients, to prevent carbohydrate malabsorption.” Savassi-Rocha et al. [28] found that mannitol excretion showed a significant decrease at 1 month but came to near normal levels by 6 months. Authors suggested this was due to intestinal adaptation.

We further found that there is little carbohydrate or protein malabsorption after RYGB even though there are significant alterations to glucose kinetics. It would hence follow that improvement in diabetes seen almost immediately after RYGB is not due to the weight loss but altered glucose metabolism and gastro-intestinal neuro-hormonal milieu [30].

On the other hand, RYGB does seem to be associated with some fat malabsorption, possibly linked to exocrine pancreatic insufficiency. Interestingly, steatorrhoea associated with RYGB is not usually associated with diarrhea, possibly because of lack of corresponding increase in fecal water [10].

It would be very interesting to know how much of this fat malabsorption with RYGB is simply an effect of Roux-en-Y reconstruction and not a function of the limb lengths. Bradley et al. [31] studied precisely that in a different group of patients. Authors examined the effect of Roux-en-Y reconstruction after total gastrectomy in five patients and compared the results with five healthy controls. They found that the malabsorption of both fat and nitrogen was significantly greater in Roux-en-Y patients compared to controls (p < 0.01). The study found moderate malabsorption of fat (19.2 ± 2.2%) and nitrogen (22 ± 2.5%) after the Roux-en-Y reconstruction. The loss was approximately three times the normal values for both fat and nitrogen. Authors also found a significant decrease in both trypsin and lipase concentration and a marked delay in secretion of these enzymes in response to a test meal (p < 0.01). This led authors to conclude that the mechanism for malabsorption in these patients was pancreatic insufficiency. This was further corroborated by the fact that malabsorption of both fat and nitrogen improved after exogenous pancreatic enzymes, but not after administration of tetracycline.

There is another angle to this debate. If malabsorption were a dominant factor with RYGB, one would expect patients to become hungrier due to negative calorie imbalance [3, 32]. What we instead see is that patients become less hungry after RYGB [33] and that RYGB is associated with a remarkable decrease in calorie intake [34]. Flancbaum et al. [34] found that the energy intake, which was 2603 kcal/day (±982) before the surgery, came down to 815 kcal/day (±196) at 3 months, 969 kcal/day (±241) at 6 months, 1095 kcal/day (±307) at 12 months, 1259 kcal/day (±466) at 18 months, and 1373 kcal/day (±620) at 24 months. Borg et al. [35] found progressively increasing levels of peptide YY and other hormones after RYGB and concluded that absence of expected increase in appetite and hence calorie intake “may be explained by gut adaptation and the consequent graded rise in the levels of gut hormones that promote satiety.”

Human small bowel length in obese subjects is highly variable and can range from 302 to 1140 cm [6]. Inevitably, this means that most patients have several meters of common channel after a proximal RYGB [28, 36]. It is inconceivable how significant malnutrition can take place in such a physiological environment. Moreover, though it is intuitive to think of any potential malabsorption with RYGB to be solely a function of the small bowel bypass, there may be other factors at play. We see from the study by Andalib et al. [18] that bacterial overgrowth can also be associated with some malabsorption.

There are several weaknesses to this review. Many of the studies in this review were conducted following a test meal. It cannot hence be ruled out that physiological responses would be different under normal circumstances but short of a test meal, and it would be difficult to compare findings at different time periods in the same individual or different individuals. Furthermore, it is possible that variations in limb lengths used for RYGB in different studies would have some bearing on the outcome. To overcome this problem, we concentrated on studies on proximal gastric bypass only and it is known from a large body of data that weight loss for proximal gastric bypass, beyond a certain minimum, is largely independent of the lengths of AL and BPL [6]. One also has to recognize the scarcity of studies evaluating fat and protein malabsorption after RYGB.

Conclusion

This systematic review concludes that there is little or no malabsorption of carbohydrates or protein after RYGB, but there is some fat malabsorption. Overall, malabsorption makes little (approximately 11.0% in the early period) overall contribution to weight loss after RYGB. There is a scarcity of long-term data on contribution of malabsorption to weight loss outcomes after RYGB.

AA, amino acid; AL, alimentary limb; AUC, area under the curve; BPL, biliopancreatic limb; CFA, coefficient of fat absorption; GB, gastric banding; GT, glucose transporter; PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses; RYGB, Roux-en-Y gastric bypass; STR, sweet taste receptor; T2DM, type 2 diabetes mellitus.