Comparison of Nutritional Deficiencies after Rouxen-Y Gastric Bypass and after Biliopancreatic Diversion with Roux-en-Y Gastric Bypass

Transcription

1 Obesity Surgery, 12, Comparison of Nutritional Deficiencies after Rouxen-Y Gastric Bypass and after Biliopancreatic Diversion with Roux-en-Y Gastric Bypass George Skroubis, MD 1 ; George Sakellaropoulos, PhD 2 ; Konstantinos Pouggouras, MD 1 ; Nancy Mead, MS, RD 1 ; George Nikiforidis, PhD 2 ; Fotis Kalfarentzos, MD, FACS 1 Nutrition Support and Morbid Obesity Clinic, 1 Department of Surgery and 2 Department of Medical Physics, School of Medicine, University of Patras, Greece Background: Patients undergoing either Roux-en-Y gastric bypass (RYGBP) or biliopancreatic diversion (BPD) with RYGBP are at risk of developing metabolic sequelae secondary to malabsorption. We compared the differences in nutritional complications between these two bariatric operations. Methods: A retrospective analysis of a prospective database was done. From June 1994 to December 2001, 243 morbidly obese patients underwent various bariatric procedures at our institution. Of these patients, 79 (BMI 45.6 ± SD=4.9) who underwent RYGBP (gastric pouch 15 ± 5ml, biliopancreatic limb cm, alimentary limb cm and common limb the remainder of the small intestine), and 95 super obese (BMI 57.2 ± 6.1) who underwent a BPD (gastric pouch 15 ± 5ml, biliopancreatic limb cm, common limb 100 cm and alimentary limb the remainder of the small intestine), were selected and studied for the incidence of micronutrient deficiencies and level of serum albumin at yearly intervals postoperatively. A variety of nutritional parameters including Hb, Fe, ferritin, folic acid, vitamin and serum albumin were measured preoperatively and compared postoperatively at 1, 3, 6, 12, 18 and 24 months, and yearly thereafter. Results: Nutritional parameters were compared preoperatively and at similar periods postoperatively. No statistically significant (P<0.05) difference in the occurrence of deficiency was observed Presented at the 6th Congress of the International Federation for the Surgery of Obesity, Chania, Crete, Greece, September 5-8, Reprint requests to: Fotis Kalfarentzos, MD, Professor of Surgery, Nutrition Support and Morbid Obesity Clinic, University of Patras, Platia Voriou Ipirou 5, Patras, Greece. Fax: ; fkalfar@med.upatras.gr between the groups for any of the nutritional parameters studied, except for ferritin, which showed a significant difference at the 2-year follow-up (37.7% low ferritin levels after RYGBP vs. 15.2% after BPD, P=0.0294). All of these deficiencies were mild, without clinical symptomatology and were easily corrected with additional supplementation of the deficient micronutrient, with no need for hospitalization. Regarding serum albumin, there was only one patient with a level below 3 g/dl in the RYGBP group and two in the BPD group. These three patients were hospitalized and received total parenteral nutrition for 3 weeks, without further complications. Conclusion: There was no significant difference in the incidence of deficiency of the nutritional parameters studied, except for ferritin, following RYGBP vs. BPD with RYGBP. The most common deficiencies encountered were of iron and vitamin. The incidence of hypoalbuminemia was negligible in both groups, with mean values above 4 g/dl. Key words: Morbid obesity, bariatric surgery, gastric bypass, biliopancreatic diversion, nutritional deficiencies Introduction The surgical approach remains the only effective long-term treatment for morbidly obese patients. Various bariatric operations have been used worldwide, being divided generally into two categories: restrictive and malabsorptive procedures. 1 Because of better and more long-lasting results in weight FD-Communications Inc. Obesity Surgery, 12,

2 Skroubis et al loss and resolution of preexisting co-morbidities, malabsorptive operations have been increasing. This has given rise to discussion regarding metabolic consequences, their prevention, and optimal treatment. The Roux-en-Y gastric bypass (RYGBP) causes weight loss by restriction of food intake and by the dumping effect. Biliopancreatic diversion (BPD) promotes weight loss principally by malabsorption of ingested nutrients. Both operations exclude parts of the gastrointestinal tract from alimentation, resulting in potential development of metabolic deficiencies, from malabsorption of certain nutrients including protein, as well as certain minerals and vitamins. 2 The purpose of this study was to investigate and compare potential nutritional complications after RYGBP and BPD. An additional aim was to examine the effectiveness of our postoperative micronutrient supplementation following these operations. Patients and Methods Since June 1994, when the Morbid Obesity Clinic was established at the University Hospital of Patras, 243 morbidly obese patients have undergone various bariatric procedures at our institution. From this prospective database the nutrient status after RYGBP and BPD with RYGBP was studied retrospectively. These two bariatric operations have been used almost exclusively at our institution over the last few years. All patients were evaluated preoperatively using a multidisciplinary approach, in order to optimize their preoperative physical condition and to select the most favorable type of bariatric operation for each patient. The selection criteria for the type of bariatric operation to be performed were the patient s preoperative BMI and personal dietary habits. If the BMI was <50, RYGBP was performed with a 15 ± 5 ml gastric pouch without complete anatomic separation from the bypassed stomach; biliopancreatic limb was cm, alimentary limb of cm and common limb the remainder of the small intestine (Figure 1A). If the BMI was 50, the operation performed was a modification of the BPD with a gastric pouch of 15 ± 5 ml without complete separation from the stomach; biliopancreatic limb was cm, common limb 100 cm and alimentary limb the remainder of the small intestine (Figure 1B). In the study, 79 morbidly obese patients who underwent RYGBP and 95 super obese who underwent the BPD based on these criteria were studied for the occurrence of nutritional complications at yearly intervals postoperatively. The patients preoperative characteristics are shown in Table 1. After surgery all patients received a daily multi- A B Alimentary limb cm Alimentary limb Biliopancreatic limb cm Biliopancreatic limb cm Common channel Figure 1. A. Scheme of the Roux-en-Y gastric bypass. Common channel 100 cm B. Scheme of the biliopancreatic diversion. 552 Obesity Surgery, 12, 2002

3 Table 1. Preoperative characteristics of the patients vitamin and mineral supplement and 2 g of calcium. No additional fat-soluble vitamin supplementation was given other than that included in the multivitamin supplement prescribed to all patients, which contains 4,000 IU vitamin A, 400 IU vitamin D and 10 mg vitamin E. An oral iron supplement was prescribed for all premenopausal women at a dose of 80 mg/day, independently of the type of operation. Starting at 6 months postoperatively, vitamin supplementation was given intramuscularly (IM) at a dose of µg as necessary, depending on measured values. A variety of nutritional parameters including hemoglobin (Hb), Fe, ferritin, folic acid, vitamin and serum albumin were measured preoperatively and postoperatively at 1, 3, 6, 12, 18 and 24 months, and yearly thereafter. Statistical Methods RYGBP BDP No. of patients Sex (male/female) 15/64 24/71 Age (years) 32.6± ±9.8 Weight (kg) 123.2± ±24 BMI 45.6± ±6.1 Results are mean ± SD Values of the nutritional parameters were compared preoperatively and at similar periods postoperatively between the two groups. Variables were Nutritional Deficiencies after RYGBP and after a BPD analyzed using Fisher s exact test for the investigation of difference in proportions (incidence of deficiency). The mean values of all parameters studied were compared at similar periods using the Student s t-test corrected for heteroscedasticity. Estimation of the time of deficiency occurrence was performed, applying a linear regression model to the data up to 6 months after surgery, since no supplementation was given until then. All reported P-values are two-sided and significant at a level of p Results Weight loss results expressed as the percentage of initial excess weight loss and total initial weight loss (mean ± standard deviation) are presented in Table 2. Statistical analysis using Student s t-test corrected for heteroscedasticity showed a significant difference for the percentage of initial excess weight loss in the first and second postoperative years. However, no significant difference was observed in the percentage of total initial weight loss between the two operations. Failure was defined as weight loss <50% of initial excess weight for both procedures. By this definition, failure rates (% of followed patients) for RYGBP were 9.2% (7/76) for the first year, 12.7% (7/55) for the second year, 42.9% (15/35) for the third year, 27.8% (5/18) for the fourth year and 25% (2/8) for the fifth postoperative year. For BPD, respectively, the rates were 21.7% (13/60) Table 2. Comparison study of weight loss RYGBP BPD 1 year 2 years 3 years 4 years 5 years 1 year 2 years 3 years 4 years 5 years Patients in follow-up IEWL% Mean ± SD 71.7±14.9* 68.7± ± ± ± ± ± ± ± IWL% Mean ± SD 36.1± ±8 31.7± ± ± ± ± ± ± * P< , + P=0.002, IEWL% = % of initial excess wt lost; IWL% = % of initial wt lost Obesity Surgery, 12,

4 Skroubis et al for the first year, 33.3% (11/33) for the second year, 43.8% (7/16) for the third year, 66.7% (6/9) for the fourth year and 100% (1/1) for the fifth year. If, however, failure is defined as weight loss <25% of initial weight, the rate of failure is much less. By this definition, failure rates (% of followed patients) for RYGBP were 5.3% (4/76) for the first year, 9.1% (5/55) for the second year, 25.7% (9/35) for the third year, 22.2% (4/18) for the fourth year and 25% (2/8) for the fifth postoperative year. For BPD, respectively, the rates were 10% (6/60) for the first year, 18.2% (6/33) for the second year, 18.8% (3/16) for the third year, 33.3% (3/9) for the fourth year and 0% (0/1) for the fifth year. It must be noted, however, that weight loss results may not be comparable in the two patient groups since the RYGBP group had a BMI kg/m 2 whereas the BPD group included only patients with a BMI >50 kg/m 2. The incidence of deficiency for the nutritional parameters studied is shown in Table 3. Statistical analysis by Fisher s exact test showed no significant difference (P<0.05) between the two types of bariatric operations, with the exception of ferritin at 2-year follow-up. At that time, 37.7% of patients after RYGBP presented with low levels of ferritin vs. 15.2% after BPD, p= There were no significant differences in the other nutritional parameters studied at any period in time. It is noteworthy, however, that the incidence of deficiency increased with time in both patient groups, with the exception of folate for which no deficiency was observed Table 3. Percentages (%) of micronutrient deficiencies RYGBP BPD Hb Fe Ferritin Hb Fe Ferritin Pre year years * years years years There was no folic acid deficiency in either group at any time period studied. *P= (Normal values: Hb: men >13.5 g/dl; women >12.5 g/dl, Fe: >50 mg%, Ferritin: >9 ng/ml, Folic acid: >1.5 ng/ml, B12: >200 pg/ml). in either group. Mean ± SD of the nutritional parameters are shown in Table 4. We found significant differences for iron, ferritin and folate preoperatively, and, as previously mentioned, for ferritin postoperatively. Finally, we studied the occurrence of hypoalbuminemia in both patient groups. There was only one patient with a serum albumin level <3 g/dl in the RYGBP group (during the first postoperative year) and two patients in the BPD group (one during the first and one during the second postoperative year). These three patients were hospitalized and received total parenteral nutrition for a 3-week period. The apparent cause of the hypoalbuminemia in all cases was very low dietary protein intake combined in the two BPD patients with an increased intake of dietary fat causing diarrhea and therefore further protein malabsorption. In all three patients following hospitalization, further instruction and increased dietary protein intake, there were no further complications or recurrence of the problem. Comparison of serum albumin levels following both procedures is shown in Table 5. It is apparent from the data that the mean serum albumin levels were in the upper normal range at all time-periods for both patient groups, even though there was a statistically significant difference between the two groups preoperatively as well as at the first postoperative year. The percentages of men and premenopausal women at each follow-up interval were not significantly different between the two operations (P>0.05). Therefore, nutritional parameters related to iron deficiency could not be due to differences in the patient populations. Regarding vitamin levels, at the 6-month follow-up, before any supplementation was given, the mean ± SD value of was 300 ± for RYGBP and ± for BPD (not significantly different, P<0.05). Using a linear regression model for the data of the first 6 months after surgery, we predicted the time of onset of vitamin deficiency for the two surgical procedures. According to our estimations, vitamin deficiency is most likely to occur at 10.7 months following RYGBP and 7.9 months following the BPD (Figure 2 A, B). 554 Obesity Surgery, 12, 2002

5 Discussion Patients who undergo either RYGBP or BPD are at risk of developing metabolic sequelae as a result of nutrient malabsorption, due to exclusion of parts of the gastrointestinal tract from alimentation. 2 Deficiency in iron, folic acid or vitamin can result in anemia, which is the most common nutritional complication in our patients, a finding that has been reported by others as well. 2-4 Iron deficiency is the most frequent nutritional deficiency following RYGBP, and the major reason appears to be malabsorption of ingested iron due to bypass of the duodenum. Because premenopausal women are the most prone to developing iron deficiency and subsequent anemia, we give prophylactic iron supplementation routinely only to this group, as has also been recommended in previous reports. 3,5 In order to study the occurrence of anemia, we analyzed hemoglobin levels, as opposed to hematocrit values, as being more accurate to define anemia. The difference in the observed relative frequency of occurrence of anemia between the two groups is not statistically significant. The percentage of men and premenopausal women was also not significantly different (P>0.05) between the two groups at any of the studied follow-up intervals. Finally, we had no cases of iron deficiency anemia refractory to treatment and thus did not have to perform hysterectomy due to metrorrhagia in any patient. The only statistically significant difference between the two groups in the nutritional parameters studied was in the levels of ferritin. Although there was not a difference in the incidence of iron deficiency, this difference in ferritin levels, which are indicative of body iron stores, may indicate the need for more aggressive iron supplementation in the early postoperative period. The reason for this difference between these two operations needs further investigation. As has been reported in previous studies, 6,7 we did not have any folic acid deficiency in either group. The difference between our study and others is that in our patients the routine daily multivitamin supplement given to all patients did not contain any folic acid. This was not our intent and as a practice is not recommended. However, due to Nutritional Deficiencies after RYGBP and after a BPD Table 4. Mean values ± standard deviation of nutritional parameters studied RYGBP BPD Mean value ± SD (no. of patients in follow-up) Mean value ± SD (no. of patients in follow-up) Hb Fe Ferritin Folate Hb Fe Ferritin Folate Pre 14±1 (72) 66.8±26.8*(73) 41.4±45 (67) 7.9±7.9 (72) 417±212 (72) 14±2.4 (95) 52±26 (95) 71.6± 82 (93) 6.3±3.1 (95) 446.2±250.1 (95) 1 year 13±1.5 (76) 85±34** (74) 59.3±239 (69) 8.1±4.7 (75) 268±114 (75) 13±1.2 (60) 65±37 (59) 64.6±76.5 +(60) 10.1±5.3 ++(60) 325.6±263.8 (60) 2 years 13±2 (55) 71±34***(55) 61.6±205 (53) 8.4±4.4 (55) 251±100 (55) 12±3.6 (33) 54±32 (33) 49.5±45.2 (33) 10.6±5.3 (32) 269±133 (33) 3 years 12±2 (35) 69±43 (35) 47.8±153 (34) 10±4.9 (35) 295±153 (35) 12±3.1 (16) 59±34 (16) 33.7±31.2 (16) 10.7±5.7 (16) 283±151 (16) 4 years 12±2 (18) 76.7±53.8 (18) 17±22 (18) 8.3±4.7 (18) 251±136 (18) 11.4±4 (9) 46.6±33.6 (9) 29.3± (9) 10.1±5.7 (9) 305.2±159.1 (9) 5 years 13±2.2 (8) 78.4±41.5 (8) 37±38 (8) 7.1±4.8 (8) 246±70.9 (8) 11 (1) 19 (1) 0.5 (1) 8 (1) 492 (1) (Normal values: Hb: men >13.5 g/dl; women >12.5 g/dl, Fe: >50 mg%, Ferritin: >9 ng/ml, Folic acid: >1.5 ng/ml, B12: >200 pg/ml). *P=0.0004, **P=0.002, ***P=0.02, +P=0.003, ++P=0.02 Obesity Surgery, 12,

6 Skroubis et al Table 5. Comparison of serum albumin RYGBP BPD Pre 1 year 2 years 3 years 4 years 5 years Pre 1 year 2 years 3 years 4 years 5 years Patients in follow-up ALB 3 % of patients ALB mean ± SD 2 4.8±0.4* 4.7±0.3** 4.7± ± ± ± ±0.44.5± ± ± ± All the comparisons made with the Fisher s exact test (P<0.05) 2. All the comparisons made with Student s t-test corrected for heteroscedasticity (P<0.05) *P=0.002, **P=0.008 problems in medical coverage by the Greek welfare system, this happens to be the only supplement covered by insurance. Our finding then of the absence of folate deficiency, despite the absence of supplementation, suggests that folate ingested from foods is adequate to cover patient needs and that the routine supplementation used by others is probably more than adequate. On the other hand, we do give additional supplements of folic acid to pregnant women and women expressing a desire to become pregnant. In contrast to folic acid, vitamin deficiency almost always develops after RYGBP operations. 8,9 In our study, despite routine supplementation, there was a rather high frequency of vitamin deficiency. However, there was no statistical difference in occurrence of deficiency between the two surgical procedures, nor were any clinical symptoms observed. On the other hand, because low vitamin levels could potentially lead to anemia and more severe clinical symptomatology, we have attempted to statistically predict as closely as possible when the drop in vitamin levels will most likely occur and we supplement before this hap- BPD B. varia- Figure 2. A. variation with time after RYGBP, including prediction of the time of onset of deficiency. tion with time after BPD, including prediction of the time of onset of deficiency. 556 Obesity Surgery, 12, 2002

7 pens. Others have also reported the appearance of low vitamin levels during the second 6-month postoperative period. 8 Based on the above observations, the majority of patients are given the first dose of vitamin supplementation at the 6- month follow-up visit. Depending on measured values, a dosage of µg was administered IM at the 6-month follow-up visit and routinely thereafter at each yearly visit. Although there are recent studies proposing effective oral treatment of deficiency, 9-11 we prefer the IM route for two reasons. First, there appears to be a prolonged effect over time, ie. there is usually no need for additional administration earlier than 1 year. Second, this type of programmed annual replacement is more convenient for the patients, and, even more important, we are certain that they are actually getting the intended dosage of the vitamin. As shown in our study, this approach leads to values of vitamin which are almost always within normal limits and consequently without clinical symptomatology. Overall, Table 4 shows that even though deficiencies were encountered for all parameters studied with the exception of folic acid, the majority were mild, and mean values were almost always in the upper normal range. This observation, combined with the absence of clinical symptomatology or the need for hospitalization, suggests that our current practice of micronutrient supplementation is effective. It must also be noted that levels of fatsoluble vitamins were not measured, and therefore cannot be commented on based on the present study. However, no clinical symptoms of deficiency were observed, and despite the fact that vitamin K is not contained in the multivitamin supplement prescribed, no patients presented with increased prothrombin time. In general, the incidence of micronutrient deficiencies observed was comparable to that reported by others after RYGBP. 6,12-14 One of the most interesting observations of our study is the extremely low incidence of hypoalbuminemia in the BPD group. The mean values for both groups were almost always above 4.5 g/dl, permitting the hypothesis that adequate protein absorption occurs through the alimentary limb and common channel. In general, the main disadvantage of malabsorptive bariatric operations is Nutritional Deficiencies after RYGBP and after a BPD thought to be hypoalbuminemia, which has been reported by some to be as high as 5-30% following more drastic malabsorptive procedures. 4,15 The BPD as performed at our institution appears to significantly reduce the incidence of this potentially serious complication. Finally, since nutritional deficiencies which occurred were not severe and were similar following RYGBP and the more malabsorptive BPD, it may be wise to investigate the effectiveness and safety of our BPD not only for the super obese, but for non-super obese patients undergoing bariatric surgery as well, with the potential for better longterm weight loss results at little or no extra metabolic cost. Our BPD is a modification of the classical BPD of Scopinaro. 15,16 In the Scopinaro BPD, the proximal gastric remnant is ml, which provides temporary weight loss. Scopinaro divides the small bowel 250 cm proximal to the ileocecal valve, and anastomoses this to the gastric remnant. The biliopancreatic limb is anastomosed to the Roux limb 50 cm proximal to the ileocecal valve. His 50-cm common limb is intended to provide malabsorption to fats and starches, maintaining the weight loss. 16 Our modification of the BPD could serve as a revision operation for a RYGBP (which has a small pouch) into a BPD, providing adequate protein-energy absorption. References 1. Deitel M. Overview of operations for morbid obesity. World J Surg 1998; 22: Brolin RE, Leung M. Survey of vitamin and mineral supplementation after gastric bypass and biliopancreatic diversion for morbid obesity. Obes Surg 1999; 9: Brolin RE, Gorman JH, Gorman RC et al. Prophylactic iron supplementation after Roux-en-Y gastric bypass: a prospective, double-blind, randomized study. Arch Surg 1998; 133: Marceau P, Hould FS, Simard S et al. Biliopancreatic diversion with duodenal switch. World J Surg 1998; 22: Rhode BM, Shustik C, Christou NV et al. Iron absorption and therapy after gastric bypass. Obes Obesity Surgery, 12,

8 Skroubis et al Surg 1999; 9: Brolin RE, Gorman JH, Gorman RC et al. Are vitamin B12 and folate deficiency clinically important after Roux-en-Y gastric bypass? J Gastrointest Surg 1998; 2: Mallory GN, Macgregor AM. Folate status following gastric bypass surgery (the great mystery). Obes Surg 1991; 1: Adachi S, Kawamoto T, Otsuka M et al. Enteral vitamin supplements reverse postgastrectomy deficiency. Ann Surg 2000; 232: Rhode BM, Tamin H, Gilfix BM et al. Treatment of vitamin deficiency after gastric surgery for severe obesity. Obes Surg 1995; 5: Kuzminski AM, Del Giacco EJ, Allen RH et al. Effective treatment of cobalamin deficiency with oral cobalamin. Blood 1998; 92: Rhode BM, MacLean LD. Vitamin and mineral supplementation after gastric bypass. In: Deitel M, Cowan GSM Jr, eds. Update: Surgery for the Morbidly Obese patient. Toronto, Canada: FD- Communications Inc. 2000: Halverson JD, Zuckerman GR, Koehler RE et al. Gastric bypass for morbid obesity: A medical-surgical assessment. Ann Surg 1981; 194: Amaral JF, Thompson WR, Caldwell MD et al. Prospective hematologic evaluation of gastric exclusion surgery for morbid obesity. Ann Surg 1985; 201: Brolin RE, Gorman RC, Milgrim LM et al. Multivitamin prophylaxis in prevention of post-gastric bypass vitamin and mineral deficiencies. Int J Obes 1991; 15: Scopinaro N, Adami GF, Marinari GM et al. Biliopancreatic diversion. World J Surg 1998; 22: Scopinaro N, Adami GF, Marinari GM et al. Bilipancreatic diversion: two decades of experience. In: Deitel M, Cowan GSM, eds, Update: Surgery for the Morbidly Obese Patient. Toronto: FD Communications Inc. 2000: (Received September 6, 2001 accepted March 30, 2002) 558 Obesity Surgery, 12, 2002