Reoperative Surgery for the Morbidly Obese A
University Experience
Joe A.
Cates, MD; Ernst J. Drenick, MD; Mohammed Z. Abedin, PhD; Jeff E. Doty, MD; Kimberly D. Saunders, MD; Joel J. Roslyn, MD
\s=b\ Patients who undergo surgery for morbid obesity are often subjected to reoperation for a wide array of indications. To evaluate outcome following revisional procedures, we reviewed the records of 32 such patients treated at UCLA between April 1986 and May 1989. Twenty-five women (78%) and 7 men (22%) with a mean age of 44 years underwent 76 reoperations (2.4 per patient) for complications of prior obesity surgery. Indications for initial surgical revision consisted primarily of metabolic derangements (12 patients) and weight-related problems (11 patients). In contrast, indications for the patients' final surgical procedure were commonly for bowel obstruction (41%), intra-abdominal sepsis (12%), and gastrointestinal bleeding (6%). Following initial revision, 23 patients (71.8%) required further surgery for major complications and four patients died (12.5%). While initial revisions are frequently indicated for metabolic problems, final reoperations are more frequently undertaken for urgent, life-threatening complications. Revisional procedures for morbid obesity should be carefully considered, and the potential for major complications and/or death should be weighed heavily against proposed
benefits.
(Arch Surg. 1990;125:1400-1404)
obesity" specific subpopulation Morbid ideal weight weigh kg body weight.12 unique
refers to a of obese individuals who in excess of 200% of or 45 over ideal Because this condi¬ tion has been associated with many systemic illnesses includ¬
ing hypertension, coronary artery disease, degenerative ar¬ thritis, diabetes, and others, morbid obesity has been itself recognized as a disease process.3"6 In the United States, it is estimated that 4.5 million individuals
or
5% of the adult
population are morbidly obese.1 Inability to achieve satisfactory weight reduction coupled with the recognized decrease in life expectancy of the morbid¬ ly obese prompted an early interest in surgical intervention as an alternative means of therapy.7"9 In 1978, a National Insti¬ tutes of Health conference affirmed the role of surgery in
selected
patients with morbid obesity.10 It was, however, suggested that these procedures should be restricted to pa¬ tients in whom traditional therapy had failed. The proposed selection criteria limited candidates to morbidly obese pa¬ tients who had been given suitable treatments of calorie
Accepted for publication May 23,1990. From the Departments of Surgery (Drs Cates, Abedin, Doty, Saunders, and Roslyn) and Medicine (Dr Drenick), UCLA School of Medicine. Read before the 61st Annual Meeting of the Pacific Coast Surgical Association, Laguna Niguel, Calif, February 21,1990. Reprint requests to Division of General Surgery, UCLA School of Medicine, Los Angeles, CA 90024 (Dr Roslyn).
restriction and/or behavioral therapy, but failed to improve or relapsed after short periods. Despite these apparently re¬ strictive criteria, the number of bariatric procedures per¬ formed each year continues to increase. Depending on the procedure performed, various aspects of alimentary function are altered and a number of long-term complications may arise.11"13 The reality of these sequelae, coupled with the un¬ certainty regarding adequacy of weight loss, has resulted in many obese patients undergoing revisionai or "conversion" operations.14"17 The aim of this study was to evaluate in a university setting the outcome of patients undergoing these reoperative bariatric procedures, and to define the scope of postoperative complications in this subset of patients. PATIENTS AND METHODS
Study Design A retrospective analysis was performed of all patients who under¬ went a bariatric procedure at UCLA Medical Center over the 37month period from April 1986 to May 1989. Those patients undergo¬ ing a single successful bariatric procedure were excluded from further study. The study population included only those patients requiring one or more subsequent procedures for the management of complications related to the primary bariatric procedure. For the purpose of this review, the first reoperation has been designated as "initial revision." In those cases in which additional procedures were required, they have been classified as either "subsequent" or "final" procedures as appropriate. Following admission or transfer to UCLA, all patients were managed on a single resident teaching service supervised by some of us. The average length of patient follow-up from time of initial revision was 39 months. Postoperative complications were defined as any morbidity that prolonged hospital¬ ization or caused readmission to the hospital. Revisionai procedures were considered early if they occurred less than 1 year following primary bariatric procedure and late if they occurred thereafter. Data inquiry was based on total number and type of surgical proce¬ dures, indications for initial and subsequent surgeries, and postoper¬ ative complications and treatments.
Patient Population From April 1986 to May 1989, 32 patients underwent reoperative bariatric procedures at UCLA and form the basis of this report. There were 25 women (78%) and seven men (22%) with an average age of 44 years (range, 30 to 58 years). These 32 patients underwent a total of 76 reoperations (2.4 per patient) for problems arising from their initial bariatric procedure. Nine patients (28%) of the study group had their initial bariatric procedure and all subsequent opera¬ tions performed at UCLA. Twenty-three patients (72%) were re-
Downloaded From: http://archsurg.jamanetwork.com/ by a DALHOUSIE UNIVERSITY-DAL-11762 User on 06/19/2015
Metabolic
Insufficient Weight Loss
Technical Problems
Excessive Weight Loss
Symptom Intolerance 8 6 10 No. of Patients
Fig 1.—Distribution of initial bariatric procedures. JIB indicates jejunoileal bypass; VBG, vertical banded gastroplasty; HG, horizontal gas¬ troplasty; BPBP, biliopancreatic bypass; and R-Y GJ, Rou-en-Y gastrojejunostomy.
12
14
16
Fig 3. —Indications for revision of the patients' initial bariatric proce¬ dure. Although there were 32 patients in the series and therefore 32 initial reoperations, one patient had two indications for revision, there¬ by accounting for the 33 indications listed.
Indications for Surgical Revision Based on
Primary Procedure*
Type of Operation Indications
JIB
Initial reoperation Metabolic
H-G
GBP
VBG
BPBP
10
Insufficient weight loss Excessive
weight loss
Technical
Symptom intolerance Subsequent reoperation Metabolic/weight-related Bowel obstruction
Intra-abdominal
sepsis
Enterocutaneous fistulae
Gastrointestinal
bleeding
1
1
*Data are expressed in numbers of patients. JIB indicates jejunoileal bypass: H-G, horizontal gastroplasty; GBP, gastric bypass; VBG, vertical banded gastroplasty; and BPBP, biliopancreatic bypass. or definitive operative procedures in the patient population. BPBP indicates biliopancreatic bypass; VBG, vertical banded gastroplasty; Rest Anat, restoration of anatomic or physiologic gastrointestinal continuity; and Gastrect, subtotal or total gastrectomy.
Fig
2.—Distribution of final
same
ferred to UCLA for management of complications arising from a bariatric procedure. Only four patients underwent reoperation with¬ in the first year, and the remaining 28 (88%) patients had late reopera¬ tions, ie, more than 1 year after the primary bariatric procedure. The distribution of initial bariatric procedures (Fig 1) was as follows: jejunoileal bypass (JIB) in 11 patients (34%); vertical banded gastro¬ plasty (VBG) in 11 patients (34%); biliopancreatic bypass (BPBP) in four patients (13%); and Roux-en-Y gastrojejunostomy (9%) and horizontal gastroplasty (9%) in two patients each). Figure 2 repre¬ sents the distribution of final procedures on the same patients. Six¬ teen patients (50%) underwent BPBP, four patients (13%) VBG, five patients (16%) either subtotal or total gastrectomy, and seven pa¬ tients (22%) were restored to physiologically "normal" anatomy. In patients restored to normal anatomy, gastrointestinal continuity was achieved via gastrogastrostomy (VBG, 2) or jejunojejunostomy 2). In cases in which a BPBP was dismantled, the recon¬ (JIB, struction consisted of a short Roux-en-Y gastrojejunostomy (n 3). This latter procedure achieved nearly normal physiologic function. Anatomic restoration was not feasible since these three patients had undergone distal gastric resection as part of their primary BPBP. =
=
=
RESULTS
Surgical Indications The various indications for initial surgical revision of baria¬ tric procedures in our study group are displayed in Fig 3. Metabolic complications were the primary indication in 44% of the patients studied. The most frequently cited problems included hyperuricemia, accelerated osteoarthritis, vitamin deficiencies, and renal lithiasis. Other reported indications included insufficient or excessive weight loss (total, 38%) and symptom intolerance (9%). Overall, 81% of the initial reoper¬ ations were performed for metabolic or weight-related prob¬ lems. The reasons for the first reoperation and subsequent procedures were further separated according to the type of primary operation performed (Table). Metabolic problems were the predominant factor responsible for reoperation in patients who had undergone an initial JIB. In contrast, weight-related and technical problems led to reoperation in the study patients with an initial VBG. Overall, the indica¬ tions for subsequent reoperation were vastly different as compared with indications for initial revision. The indications for subsequent reoperations included small-bowel obstrue-
Downloaded From: http://archsurg.jamanetwork.com/ by a DALHOUSIE UNIVERSITY-DAL-11762 User on 06/19/2015
intra-abdominal sepsis associated with multisystem organ failure after having been transferred to UCLA for complica¬ tions related to their revisionai procedure. One patient died of overwhelming septic shock and malnutrition after having been unavailable for follow-up. The final death occurred as a result of cardiac arrest in a patient who was recovering from a small-bowel resection for intestinal obstruction secondary to an internal hernia. Seventy-two percent of all patients undergoing revisionai bariatric procedures required further intervention for major gastrointestinal complications. Furthermore, 41% of all pa¬ tients referred to UCLA following a bariatric procedure had a life-threatening illness that required urgent intervention.
Bowel Obstruction
Metabolic Weight Loss Intra-abdominal Sepsis
Gastrointestinal Fistulae
Gastrointestinal Hemorrhage
Among
0
2
4
6
8
10
12
14
16
No. of Patients
Fig 4.
Indications for final operative intervention in all 32 patients. As in Fig 3, one patient had two surgical indications. —
enterocutaneous fistulae, sepsis, and gastrointestinal hemorrhage. Within each operative category, the limited number of patients precludes more definitive analysis. Figure 4 summarizes the frequency of indications for final procedures in the same 32 patients. Bowel obstruction was the surgical indication in 13 patients (41%), intra-abdominal sepsis in four patients (13%), and gastrointestinal hemor¬ rhage and enterocutaneous fistulae in two patients each (12%). These indications represented life-threatening illness¬ es, and urgent surgery was required in most cases. Final operative procedures were undertaken for metabolic or weight-related problems in only 12 patients (38%). Postoperative Complications Seventy-six reoperations were performed following an ini¬ tial bariatric procedure during the 37-month period of this study. Following initial revision, 23 (72%) of the 32 patients required further interventions for major complications. Of the nine patients who underwent a single revision, two (22%) died secondary to postoperative complications. The following
tion,
tabulation summarizes the total number of postoperative complications that occurred during this time. There were 78 complications in all, and while metabolic complications were the most frequent, occurring in 18 patients (56%), 13 (41%) had major wound or intra-abdominal infectious complications. Forty percent of patients experienced a postoperative intesti¬ nal obstruction and 12 patients (38%) developed an incisional hernia. Pancreatitis, manifested clinically as an elevation in serum amylase and abdominal pain, was experienced by six
patients (19%).
Type of Complication
Reoperation
Metabolic Infectious Bowel obstruction Hernia Pancreatitis Gastrointestinal bleeding Cardiac Central nervous system Death
No. of Patients 23 18 13 13 12 6 6 4 2 4
All of these individuals responded to conservative mea¬ sures, and no pancreatic pseudocyst or abscess formation was
reported. Cardiac complications consisted primarily of ar¬ rhythmias and contributed to one of the four reported mortali¬ ties. Two individuals experienced postoperative cerebrovas¬ cular accidents. Four (12.5%) of the patients undergoing revisionai bariatric procedures died. Two patients died of
the patients with life-threatening complications, 3.5 complications per patient and the associated mortality rate was 25%. In contrast, the complication rate among patients undergoing elective admission was 1.4 per patient and there were no deaths in this group. Four patients died following surgical revision, constituting an overall mor¬ tality rate of 12.5%. Three of the four reported deaths oc¬ curred in patients who had undergone BPBP either as an initial weight-loss procedure or as a subsequent revision.
there
were
COMMENT Bariatric surgery continues to be an evolving discipline. The ideal operation for the morbidly obese has yet to be identified and this whole area has been the focus of some controversy.18 The alterations in normal gastrointestinal physiology produced by many of the currently performed procedures, in combination with the difficulty in predicting weight loss, is probably responsible for the number of pa¬ tients who ultimately require some type of revision or reoper¬ ation. Data from the present study suggest that indications for initial surgical revision commonly include weight-related complaints or metabolic disorders. In contrast, when subse¬ quent procedures are performed, the indications are general¬ ly much more serious. Furthermore, reoperation following revisionai bariatric procedures is associated with significant morbidity and mortality. During a 3-year period, 32 patients treated at UCLA un¬ derwent revisionai surgery or reoperation following an initial bariatric procedure. Overall, 76 reoperations were performed in these patients for complications of prior obesity surgery. Postoperative morbidity was considerable in this group and the mortality rate was 12.5%. Three (75%) of the four deaths occurred in patients who had undergone a BPBP. While the death rate in the 16 patients who underwent some other form of bariatric procedure was 6.3%, the rate in patients undergo¬ ing reoperations following BPBP was 18.8%. The frequency with which bariatric procedures are being performed in this country is unclear. Although the need for reoperation for insufficient or excessive weight loss or for associated metabolic disturbances is well recognized, it is nearly impossible to quantify the frequency with which this occurs. The risks associated with conversion surgery have recently been reviewed by Yale.15 In this study, 120 patients undergoing conversion surgery between 1977 and 1987 were retrospectively analyzed. The early complication rate in this series was 18%, a figure similar to that reported by Linner16 and Forse et al." These authors concluded that conversion surgery is safe and effective. The results of our current study do not support this statement, but it may well be that we are talking about two very different patient populations. This is suggested by the large number of patients transferred to our institution with severe, potentially life-threatening prob¬ lems. Undoubtedly, this observation reflects the tertiary referral nature of our practice. Additionally, previous reports have focused on revision of a gastric partitioning or JIB
Downloaded From: http://archsurg.jamanetwork.com/ by a DALHOUSIE UNIVERSITY-DAL-11762 User on 06/19/2015
bypass to a gastric bypass, unbanded gastrogastrostomy, or VBG.1""17 Conspicuously absent from these reports has been information dealing with patients who have undergone a BPBP. This procedure, first described by Scopinaro et al in 1976,19 is being performed by a number of surgeons through¬ out this
country. There is limited information, however, fo¬
long-term outcome and complications associated with this procedure. Some of the most catastrophic complications observed in the current series occurred in patients who had had a BPBP. Three patients with a functioning BPBP performed else¬ where underwent revisionai procedures for metabolic or weight-related problems at an outside facility. All three pa¬ tients had major complications with recurrent intra-abdomi¬ nal sepsis and complex enterocutaneous fistulae. They were ultimately transferred to UCLA where, after lengthy hospitalizations in excess of 6 months each, two have undergone definitive reconstructive surgery. The third patient contin¬ ues to have an enterocutaneous fistula and is being managed with home total parenteral nutrition. Two of these patients ultimately underwent takedown of their bypasses with phys¬ iologic (as opposed to anatomic) restoration of their gastroin¬ testinal tracts. Four patients requiring reoperation following BPBP developed severe fungal sepsis, of which one died. The etiologic features of fungemia are not clear in this subgroup of patients, but it is interesting to speculate about the role of bacterial overgrowth in the biliary limb. cusing
on
The vagaries of bariatric surgery suggest that revisionai procedures will continue to be done until, perhaps, the "per¬ fect" weight-reducing procedure is identified. It is difficult to identify specific criteria that should be used in the selection of an appropriate revisionai procedure. This decision should be based on the surgeon's particular expertise, patient's previ¬ ous procedures, current anatomy, eating habits, and overall medical condition. Although revisionai or conversional sur¬ gery can generally be performed by experienced surgeons with minimal morbidity and mortality, one should be cogni¬ zant of the potential risks involved. Many of these patients, at the time of reoperation, are malnourished and not in optimal condition. Moreover, the anatomy of certain gastrointestinal arrangements may make dissection difficult and predispose patients to bacterial overgrowth secondary to "blind limbs." These factors should be carefully considered when contem¬ plating reoperative surgery for the morbidly obese. While we accept the possibility that our population is very likely skewed, nonetheless, we must conclude that complications arising from reoperative bariatric procedures can be devas¬ tating. This appears to be particularly true for patients under¬ going revision of a BPBP. In conclusion, our data suggest that revisionai procedures for morbid obesity should be subjected to the utmost scrutiny, and the potential for major complica¬ tions and/or death should be weighed heavily against pro¬ posed benefits.
References S, Johnson CL. Prevalence of severe obesity in adults in the United States. Am J Clin Nutr. 1980;33:364-369. 2. Kral JG, Heymsfield S. Morbid obesity: definitions, epidemiology, and methodological problems. Gastroenterol Clin North Am. 1987;16:197-205. 3. Carey LC, Martin EW Jr, Mojzisik C. The surgical treatment of morbid obesity. Curr Probl Surg. 1984;21:8-78. 4. Foster WR, Burton BT, Van Itallie TB. Health implications of obesity: National Institutes of Health consensus development conference. Ann Intern Med. 1985;103:977-1077. 5. Drenick EJ. Definition and health consequences of morbid obesity. Surg Clin North Am. 1979;59:963-976. 6. Yale CE. Gastric surgery for morbid obesity: complications and long-term weight control. Arch Surg. 1989;124:941-946. 7. Van Itallie TB. Morbid obesity: a hazardous disorder that resists conservative treatment. Am J Clin Nutr. 1980;33:358-363. 8. Buchwald H, Schwartz MZ, Varco RL. Surgical treatment of obesity. Adv Surg. 1973;7:235-255. 9. Linner JM. A summary of 24 years' experience with surgery for morbid obesity. Am J Clin Nutr. 1980;33:504-505. 10. Van Itallie TB, Burton BT. National Institutes of Health consensus development conferences on surgical treatment of morbid obesity. Ann Surg. 1979;189:455-457. 1. Abraham
11. Buckwalter JA, Herbst CA. Complications of gastric bypass for morbid obesity. Am JSurg. 1980;139:55-60. 12. Buckwalter JA, Herbst GA. Perioperative complications of gastric restrictive operations. Am J Surg. 1983;146:613-618. 13. Peltier G, Hermreck AS, Moffat RE, Hardin CA, Jewell WR. Complications following gastric bypass procedures for morbid obesity. Surgery. 1979;86:648-654. 14. Scopinaro N, Gianetta E, Friedmen D, Adami G, Traverso E, Vitale B. Surgical revision of the bilio-pancreatic diversion. Gastroenterol Clin North
Am. 1987;16:529-531. 15. Yale CE. Conversion surgery for morbid obesity: complications and long-term weight control. Surgery. 1989;106:474-480. 16. Linner JH. Revisional surgery. In: Surgery for Morbid Obesity. New York, NY: Springer-Verlag NY Inc; 1984:109-126. 17. Forse RA, Deitel M, MacLean LD. The conversion of a horizontal to vertical banded gastroplasty: a hazardous procedure. Gastroenterol Clin North Am. 1987;16:533-535. 18. MacLean LD. Surgery for obesity: where do we go from here? Am Coll
Surg Bull. 1989;74:10-23.
19. Scopinaro N, Gianetta E, Pandolfo Minerva Chir. 1976;31:560-566.
N,
et al.
Biliopancreatic bypass.
Discussion
STABILE, MD, San Diego, Calif: The high morbidity and mortality attending late reoperation after weight-reducing surgery for the morbidly obese is often ignored. Since JIB operations are becoming more obsolete, relatively little information has been forth¬ coming on the late reoperation rates and risks following gastric bypass, gastroplasty, and the BPBP procedure. These operations have long-term failure rates and serious complications that have been minimized or not addressed at all. The true outcome data for these sometimes casually applied procedures would be particularly sober¬ ing. I remain to be convinced that bariatric surgery as widely prac¬ ticed under the currently accepted eligibility criteria has durable benefits that clearly outweigh the perioperative and long-term mor¬ tality and morbidity. This is most particularly the case if one includes failure to sustain an acceptable weight loss among the numerous complications. Experience has shown that the laudable results from centers with particular interest and expertise in bariatric surgery are attributable in substantial part to the interest, expertise, and vigi¬ lance in the dietary management and counseling that postoperative patients receive for the duration of their lives. Current investigations into the metabolism of adipocytes and the neuroendocrine control of appetite and satiety offer the promise of new nonsurgical approaches to both prevention and management. BRUCE E.
Specific regulatory abnormalities in triglycéride storage and mobili¬ zation and in plasma cholecystokinin levels have been identified in a variety of pathologic states including obesity, and therapeutic strate¬ gies are already being devised. It is disturbing that a surgical proce¬ dure as deforming and potentially morbid as the BPBP has evolved as treatment for a fundamentally metabolic or psychologic disorder. Hopefully, pharmacologie manipulations will relegate such high-risk operations to the bariatric museum shelf. How often do patients require late surgical intervention for compli¬ cations or failure of bariatric operations? Are these procedures being applied too casually and by unqualified practitioners? Is follow-up often grossly inadequate? The 41% rate of referred patients requiring urgent intervention raises the question of neglect or abandonment of these difficult patients who need lifelong attention and support. Another issue is the extremely high rate of multiple reoperations and their attendant morbidity and mortality. Do the authors believe these results to be representative and expected? These data do not support the notion that bariatric surgery renders obese people safe for future operations. Do the authors believe the BPBP to be a justifiable operation for morbid obesity? Do they accept the criterion of a body weight 45 kg over the ideal weight as a valid indication for a weight-reducing
Downloaded From: http://archsurg.jamanetwork.com/ by a DALHOUSIE UNIVERSITY-DAL-11762 User on 06/19/2015
operation
in the absence of other specific complications? Do they believe that the long-term salutary effects outweigh the combined total short- and long-term morbidity and mortality attending baria¬ tric operations? DAVID TAPPER, MD, Seattle, Wash: Thirty-four percent of the patients had JIBs. Was the operation performed as their only opera¬ tion, or did they come from some other group? Were the complica¬ tions uniform across the group? The patients with BPBPs had a substantial number but where did the complications, particularly the metabolic ones, relate to the JIB? Could you break the major and minor complications down? Were there wound complications such as hernias? Were they a result of wound infection? CLAUDE H. ORGAN, Jr, MD, Oakland, Calif: Many operations applauded in one decade have been discarded in the next. Bariatric surgery is such an example. The second most difficult decision in surgery is when to operate, the most difficult one being when to reoperate. These patients present a challenge for reoperation. The principal fault of bariatric surgery has been that the operations have been too easy. The jejunoileal bypass was too easy. The stapler made the gastric bypass easy. It was our experience that two thirds of the patients who came in for reoperation were malnourished but could be brought to a good nutritional state before reoperation. The biliary pancreatic bypass is a Mann-Williamson procedure (Ann Surg. 1923;77:409-422). It conforms very closely to what Mann and William¬ son described from the Mayo Clinic in the 1920s. Why is there such a high percentage of bowel obstructions in these patients? The second question has to do with the diabetic patient. What was your experience with diabetes? These are patients who have an incredible turnaround in their insulin requirements after surgery. Even before they left the hospital, some required no insulin, hypogly¬ cémie agents, or dietary management. Did you have any experience with these groups, and were you reluctant to correct their bypass and return the gastrointestinal tract to normal? DR ROSLYN: I hope that the general surgeons in the audience are interested in bariatric surgery, not necessarily as a clinical interest, but as an area that all general surgeons should be familiar with. In an article in the September 1989 Bulletin of the American College of Surgeons, Lloyd McLean, MD clearly addresses many of the problems that deal with bariatric surgery as a discipline. Dr Stabile, in the January 1990 issue of Surgery, Dr McLean suggests that in their series of 200 patients undergoing VBG, the reoperation rate was 36%. There truly are a large number of patients who undergo bariatric procedures and ultimately require reopera¬ tion. In most cases, the indication relates to insufficient or excessive weight loss. It is hard for me to comment on follow-up. When we were actively doing bariatric procedures at our institution, the only way we would operate on any of these patients was if they were willing to be seen and followed up postoperatively by a specialist in metabolism. Are these results representative, and how serious are some of these complications? Let me describe three of the more difficult patients who were treated in our series. A young lady underwent a
VBG at an outside institution and developed a leak from her staple line. She was reoperated on several times and was finally transferred to UCLA in florid sepsis with a huge gastrocutaneous fistula. Subse¬ quent tests done at UCLA demonstrated that she had a small pitu¬ itary tumor and that her obesity was really due to Cushing's disease. She died as a result of ongoing sepsis. A second patient had a BPBP performed at an outside institution, and was reoperated on 20 months later because of problems with diarrhea. At reoperation, several enterotomies were made and she was ultimately transferred to UCLA with eight fistulas. She was hospitalized for 5 months before we were able to reconstruct her gastrointestinal tract by taking down her BPBP and restoring her normal anatomy. The final patient had a BPBP performed in Italy. She came back to the United States and ultimately required takedown of her bypass revision. A gastroduodenostomy attempt resulted in anastomotic breakdown. She was transferred to UCLA with no abdominal wall and 16 fistulas. She was hospitalized for 8 months before her defini¬ tive surgery was done. Dr Stabile, BPBP does achieve weight loss; but in terms of safety, it is still experimental, and further data are needed to establish its role. I will not address the issue as to whether patients who are morbidly obese should have bariatric procedures. If this discipline is to survive in the surgical world, it needs to focus on results not just in terms of weight loss; they need to look hard at results in terms of metabolic and postoperative complications. Most of the bariatric procedures have become very easy. About half of the bowel obstructions were really gastric outlet obstructions and the remaining patients had small-bowel obstructions. All the small-bowel obstructions occurred in patients who had a BPBP. The very long distal Roux-en-Y gastrojejunostomy obviously has a great propensity for internal herniation. These operations are very effec¬ tive in reversing diabetes in the morbidly obese patient. In my experience, a number of patients who were receiving 60 units or more of isophane insulin suspension per day were able, within 2 months, to have insulin therapy discontinued completely. The last bariatric operation I performed was about 5 months ago. It was in a 181-kg physician who had such profound sleep apnea that he was dozing off while I was interviewing him in my office. Within 3 weeks after he was discharged from the hospital, he had absolutely no evidence of sleep apnea. Dr Tapper asked about JIBs. All the patients undergoing a reoper¬ ation for JIB were referred from other institutions. We have not performed a JIB at UCLA in the last 14 to 15 years. Most of the catastrophic complications were in patients who had undergone a BPBP. Most of the wound complications were septic in nature and most of the hernias occurred in that setting. At UCLA our experience with "virgin" bariatric procedures has really been limited because it has not been a major focus. The only patients that I see have had bariatric procedures and now have problems, not in terms of requiring revision for weight-related prob¬ lems, but other sorts of problems.
Downloaded From: http://archsurg.jamanetwork.com/ by a DALHOUSIE UNIVERSITY-DAL-11762 User on 06/19/2015
University Experience
Joe A.
Cates, MD; Ernst J. Drenick, MD; Mohammed Z. Abedin, PhD; Jeff E. Doty, MD; Kimberly D. Saunders, MD; Joel J. Roslyn, MD
\s=b\ Patients who undergo surgery for morbid obesity are often subjected to reoperation for a wide array of indications. To evaluate outcome following revisional procedures, we reviewed the records of 32 such patients treated at UCLA between April 1986 and May 1989. Twenty-five women (78%) and 7 men (22%) with a mean age of 44 years underwent 76 reoperations (2.4 per patient) for complications of prior obesity surgery. Indications for initial surgical revision consisted primarily of metabolic derangements (12 patients) and weight-related problems (11 patients). In contrast, indications for the patients' final surgical procedure were commonly for bowel obstruction (41%), intra-abdominal sepsis (12%), and gastrointestinal bleeding (6%). Following initial revision, 23 patients (71.8%) required further surgery for major complications and four patients died (12.5%). While initial revisions are frequently indicated for metabolic problems, final reoperations are more frequently undertaken for urgent, life-threatening complications. Revisional procedures for morbid obesity should be carefully considered, and the potential for major complications and/or death should be weighed heavily against proposed
benefits.
(Arch Surg. 1990;125:1400-1404)
obesity" specific subpopulation Morbid ideal weight weigh kg body weight.12 unique
refers to a of obese individuals who in excess of 200% of or 45 over ideal Because this condi¬ tion has been associated with many systemic illnesses includ¬
ing hypertension, coronary artery disease, degenerative ar¬ thritis, diabetes, and others, morbid obesity has been itself recognized as a disease process.3"6 In the United States, it is estimated that 4.5 million individuals
or
5% of the adult
population are morbidly obese.1 Inability to achieve satisfactory weight reduction coupled with the recognized decrease in life expectancy of the morbid¬ ly obese prompted an early interest in surgical intervention as an alternative means of therapy.7"9 In 1978, a National Insti¬ tutes of Health conference affirmed the role of surgery in
selected
patients with morbid obesity.10 It was, however, suggested that these procedures should be restricted to pa¬ tients in whom traditional therapy had failed. The proposed selection criteria limited candidates to morbidly obese pa¬ tients who had been given suitable treatments of calorie
Accepted for publication May 23,1990. From the Departments of Surgery (Drs Cates, Abedin, Doty, Saunders, and Roslyn) and Medicine (Dr Drenick), UCLA School of Medicine. Read before the 61st Annual Meeting of the Pacific Coast Surgical Association, Laguna Niguel, Calif, February 21,1990. Reprint requests to Division of General Surgery, UCLA School of Medicine, Los Angeles, CA 90024 (Dr Roslyn).
restriction and/or behavioral therapy, but failed to improve or relapsed after short periods. Despite these apparently re¬ strictive criteria, the number of bariatric procedures per¬ formed each year continues to increase. Depending on the procedure performed, various aspects of alimentary function are altered and a number of long-term complications may arise.11"13 The reality of these sequelae, coupled with the un¬ certainty regarding adequacy of weight loss, has resulted in many obese patients undergoing revisionai or "conversion" operations.14"17 The aim of this study was to evaluate in a university setting the outcome of patients undergoing these reoperative bariatric procedures, and to define the scope of postoperative complications in this subset of patients. PATIENTS AND METHODS
Study Design A retrospective analysis was performed of all patients who under¬ went a bariatric procedure at UCLA Medical Center over the 37month period from April 1986 to May 1989. Those patients undergo¬ ing a single successful bariatric procedure were excluded from further study. The study population included only those patients requiring one or more subsequent procedures for the management of complications related to the primary bariatric procedure. For the purpose of this review, the first reoperation has been designated as "initial revision." In those cases in which additional procedures were required, they have been classified as either "subsequent" or "final" procedures as appropriate. Following admission or transfer to UCLA, all patients were managed on a single resident teaching service supervised by some of us. The average length of patient follow-up from time of initial revision was 39 months. Postoperative complications were defined as any morbidity that prolonged hospital¬ ization or caused readmission to the hospital. Revisionai procedures were considered early if they occurred less than 1 year following primary bariatric procedure and late if they occurred thereafter. Data inquiry was based on total number and type of surgical proce¬ dures, indications for initial and subsequent surgeries, and postoper¬ ative complications and treatments.
Patient Population From April 1986 to May 1989, 32 patients underwent reoperative bariatric procedures at UCLA and form the basis of this report. There were 25 women (78%) and seven men (22%) with an average age of 44 years (range, 30 to 58 years). These 32 patients underwent a total of 76 reoperations (2.4 per patient) for problems arising from their initial bariatric procedure. Nine patients (28%) of the study group had their initial bariatric procedure and all subsequent opera¬ tions performed at UCLA. Twenty-three patients (72%) were re-
Downloaded From: http://archsurg.jamanetwork.com/ by a DALHOUSIE UNIVERSITY-DAL-11762 User on 06/19/2015
Metabolic
Insufficient Weight Loss
Technical Problems
Excessive Weight Loss
Symptom Intolerance 8 6 10 No. of Patients
Fig 1.—Distribution of initial bariatric procedures. JIB indicates jejunoileal bypass; VBG, vertical banded gastroplasty; HG, horizontal gas¬ troplasty; BPBP, biliopancreatic bypass; and R-Y GJ, Rou-en-Y gastrojejunostomy.
12
14
16
Fig 3. —Indications for revision of the patients' initial bariatric proce¬ dure. Although there were 32 patients in the series and therefore 32 initial reoperations, one patient had two indications for revision, there¬ by accounting for the 33 indications listed.
Indications for Surgical Revision Based on
Primary Procedure*
Type of Operation Indications
JIB
Initial reoperation Metabolic
H-G
GBP
VBG
BPBP
10
Insufficient weight loss Excessive
weight loss
Technical
Symptom intolerance Subsequent reoperation Metabolic/weight-related Bowel obstruction
Intra-abdominal
sepsis
Enterocutaneous fistulae
Gastrointestinal
bleeding
1
1
*Data are expressed in numbers of patients. JIB indicates jejunoileal bypass: H-G, horizontal gastroplasty; GBP, gastric bypass; VBG, vertical banded gastroplasty; and BPBP, biliopancreatic bypass. or definitive operative procedures in the patient population. BPBP indicates biliopancreatic bypass; VBG, vertical banded gastroplasty; Rest Anat, restoration of anatomic or physiologic gastrointestinal continuity; and Gastrect, subtotal or total gastrectomy.
Fig
2.—Distribution of final
same
ferred to UCLA for management of complications arising from a bariatric procedure. Only four patients underwent reoperation with¬ in the first year, and the remaining 28 (88%) patients had late reopera¬ tions, ie, more than 1 year after the primary bariatric procedure. The distribution of initial bariatric procedures (Fig 1) was as follows: jejunoileal bypass (JIB) in 11 patients (34%); vertical banded gastro¬ plasty (VBG) in 11 patients (34%); biliopancreatic bypass (BPBP) in four patients (13%); and Roux-en-Y gastrojejunostomy (9%) and horizontal gastroplasty (9%) in two patients each). Figure 2 repre¬ sents the distribution of final procedures on the same patients. Six¬ teen patients (50%) underwent BPBP, four patients (13%) VBG, five patients (16%) either subtotal or total gastrectomy, and seven pa¬ tients (22%) were restored to physiologically "normal" anatomy. In patients restored to normal anatomy, gastrointestinal continuity was achieved via gastrogastrostomy (VBG, 2) or jejunojejunostomy 2). In cases in which a BPBP was dismantled, the recon¬ (JIB, struction consisted of a short Roux-en-Y gastrojejunostomy (n 3). This latter procedure achieved nearly normal physiologic function. Anatomic restoration was not feasible since these three patients had undergone distal gastric resection as part of their primary BPBP. =
=
=
RESULTS
Surgical Indications The various indications for initial surgical revision of baria¬ tric procedures in our study group are displayed in Fig 3. Metabolic complications were the primary indication in 44% of the patients studied. The most frequently cited problems included hyperuricemia, accelerated osteoarthritis, vitamin deficiencies, and renal lithiasis. Other reported indications included insufficient or excessive weight loss (total, 38%) and symptom intolerance (9%). Overall, 81% of the initial reoper¬ ations were performed for metabolic or weight-related prob¬ lems. The reasons for the first reoperation and subsequent procedures were further separated according to the type of primary operation performed (Table). Metabolic problems were the predominant factor responsible for reoperation in patients who had undergone an initial JIB. In contrast, weight-related and technical problems led to reoperation in the study patients with an initial VBG. Overall, the indica¬ tions for subsequent reoperation were vastly different as compared with indications for initial revision. The indications for subsequent reoperations included small-bowel obstrue-
Downloaded From: http://archsurg.jamanetwork.com/ by a DALHOUSIE UNIVERSITY-DAL-11762 User on 06/19/2015
intra-abdominal sepsis associated with multisystem organ failure after having been transferred to UCLA for complica¬ tions related to their revisionai procedure. One patient died of overwhelming septic shock and malnutrition after having been unavailable for follow-up. The final death occurred as a result of cardiac arrest in a patient who was recovering from a small-bowel resection for intestinal obstruction secondary to an internal hernia. Seventy-two percent of all patients undergoing revisionai bariatric procedures required further intervention for major gastrointestinal complications. Furthermore, 41% of all pa¬ tients referred to UCLA following a bariatric procedure had a life-threatening illness that required urgent intervention.
Bowel Obstruction
Metabolic Weight Loss Intra-abdominal Sepsis
Gastrointestinal Fistulae
Gastrointestinal Hemorrhage
Among
0
2
4
6
8
10
12
14
16
No. of Patients
Fig 4.
Indications for final operative intervention in all 32 patients. As in Fig 3, one patient had two surgical indications. —
enterocutaneous fistulae, sepsis, and gastrointestinal hemorrhage. Within each operative category, the limited number of patients precludes more definitive analysis. Figure 4 summarizes the frequency of indications for final procedures in the same 32 patients. Bowel obstruction was the surgical indication in 13 patients (41%), intra-abdominal sepsis in four patients (13%), and gastrointestinal hemor¬ rhage and enterocutaneous fistulae in two patients each (12%). These indications represented life-threatening illness¬ es, and urgent surgery was required in most cases. Final operative procedures were undertaken for metabolic or weight-related problems in only 12 patients (38%). Postoperative Complications Seventy-six reoperations were performed following an ini¬ tial bariatric procedure during the 37-month period of this study. Following initial revision, 23 (72%) of the 32 patients required further interventions for major complications. Of the nine patients who underwent a single revision, two (22%) died secondary to postoperative complications. The following
tion,
tabulation summarizes the total number of postoperative complications that occurred during this time. There were 78 complications in all, and while metabolic complications were the most frequent, occurring in 18 patients (56%), 13 (41%) had major wound or intra-abdominal infectious complications. Forty percent of patients experienced a postoperative intesti¬ nal obstruction and 12 patients (38%) developed an incisional hernia. Pancreatitis, manifested clinically as an elevation in serum amylase and abdominal pain, was experienced by six
patients (19%).
Type of Complication
Reoperation
Metabolic Infectious Bowel obstruction Hernia Pancreatitis Gastrointestinal bleeding Cardiac Central nervous system Death
No. of Patients 23 18 13 13 12 6 6 4 2 4
All of these individuals responded to conservative mea¬ sures, and no pancreatic pseudocyst or abscess formation was
reported. Cardiac complications consisted primarily of ar¬ rhythmias and contributed to one of the four reported mortali¬ ties. Two individuals experienced postoperative cerebrovas¬ cular accidents. Four (12.5%) of the patients undergoing revisionai bariatric procedures died. Two patients died of
the patients with life-threatening complications, 3.5 complications per patient and the associated mortality rate was 25%. In contrast, the complication rate among patients undergoing elective admission was 1.4 per patient and there were no deaths in this group. Four patients died following surgical revision, constituting an overall mor¬ tality rate of 12.5%. Three of the four reported deaths oc¬ curred in patients who had undergone BPBP either as an initial weight-loss procedure or as a subsequent revision.
there
were
COMMENT Bariatric surgery continues to be an evolving discipline. The ideal operation for the morbidly obese has yet to be identified and this whole area has been the focus of some controversy.18 The alterations in normal gastrointestinal physiology produced by many of the currently performed procedures, in combination with the difficulty in predicting weight loss, is probably responsible for the number of pa¬ tients who ultimately require some type of revision or reoper¬ ation. Data from the present study suggest that indications for initial surgical revision commonly include weight-related complaints or metabolic disorders. In contrast, when subse¬ quent procedures are performed, the indications are general¬ ly much more serious. Furthermore, reoperation following revisionai bariatric procedures is associated with significant morbidity and mortality. During a 3-year period, 32 patients treated at UCLA un¬ derwent revisionai surgery or reoperation following an initial bariatric procedure. Overall, 76 reoperations were performed in these patients for complications of prior obesity surgery. Postoperative morbidity was considerable in this group and the mortality rate was 12.5%. Three (75%) of the four deaths occurred in patients who had undergone a BPBP. While the death rate in the 16 patients who underwent some other form of bariatric procedure was 6.3%, the rate in patients undergo¬ ing reoperations following BPBP was 18.8%. The frequency with which bariatric procedures are being performed in this country is unclear. Although the need for reoperation for insufficient or excessive weight loss or for associated metabolic disturbances is well recognized, it is nearly impossible to quantify the frequency with which this occurs. The risks associated with conversion surgery have recently been reviewed by Yale.15 In this study, 120 patients undergoing conversion surgery between 1977 and 1987 were retrospectively analyzed. The early complication rate in this series was 18%, a figure similar to that reported by Linner16 and Forse et al." These authors concluded that conversion surgery is safe and effective. The results of our current study do not support this statement, but it may well be that we are talking about two very different patient populations. This is suggested by the large number of patients transferred to our institution with severe, potentially life-threatening prob¬ lems. Undoubtedly, this observation reflects the tertiary referral nature of our practice. Additionally, previous reports have focused on revision of a gastric partitioning or JIB
Downloaded From: http://archsurg.jamanetwork.com/ by a DALHOUSIE UNIVERSITY-DAL-11762 User on 06/19/2015
bypass to a gastric bypass, unbanded gastrogastrostomy, or VBG.1""17 Conspicuously absent from these reports has been information dealing with patients who have undergone a BPBP. This procedure, first described by Scopinaro et al in 1976,19 is being performed by a number of surgeons through¬ out this
country. There is limited information, however, fo¬
long-term outcome and complications associated with this procedure. Some of the most catastrophic complications observed in the current series occurred in patients who had had a BPBP. Three patients with a functioning BPBP performed else¬ where underwent revisionai procedures for metabolic or weight-related problems at an outside facility. All three pa¬ tients had major complications with recurrent intra-abdomi¬ nal sepsis and complex enterocutaneous fistulae. They were ultimately transferred to UCLA where, after lengthy hospitalizations in excess of 6 months each, two have undergone definitive reconstructive surgery. The third patient contin¬ ues to have an enterocutaneous fistula and is being managed with home total parenteral nutrition. Two of these patients ultimately underwent takedown of their bypasses with phys¬ iologic (as opposed to anatomic) restoration of their gastroin¬ testinal tracts. Four patients requiring reoperation following BPBP developed severe fungal sepsis, of which one died. The etiologic features of fungemia are not clear in this subgroup of patients, but it is interesting to speculate about the role of bacterial overgrowth in the biliary limb. cusing
on
The vagaries of bariatric surgery suggest that revisionai procedures will continue to be done until, perhaps, the "per¬ fect" weight-reducing procedure is identified. It is difficult to identify specific criteria that should be used in the selection of an appropriate revisionai procedure. This decision should be based on the surgeon's particular expertise, patient's previ¬ ous procedures, current anatomy, eating habits, and overall medical condition. Although revisionai or conversional sur¬ gery can generally be performed by experienced surgeons with minimal morbidity and mortality, one should be cogni¬ zant of the potential risks involved. Many of these patients, at the time of reoperation, are malnourished and not in optimal condition. Moreover, the anatomy of certain gastrointestinal arrangements may make dissection difficult and predispose patients to bacterial overgrowth secondary to "blind limbs." These factors should be carefully considered when contem¬ plating reoperative surgery for the morbidly obese. While we accept the possibility that our population is very likely skewed, nonetheless, we must conclude that complications arising from reoperative bariatric procedures can be devas¬ tating. This appears to be particularly true for patients under¬ going revision of a BPBP. In conclusion, our data suggest that revisionai procedures for morbid obesity should be subjected to the utmost scrutiny, and the potential for major complica¬ tions and/or death should be weighed heavily against pro¬ posed benefits.
References S, Johnson CL. Prevalence of severe obesity in adults in the United States. Am J Clin Nutr. 1980;33:364-369. 2. Kral JG, Heymsfield S. Morbid obesity: definitions, epidemiology, and methodological problems. Gastroenterol Clin North Am. 1987;16:197-205. 3. Carey LC, Martin EW Jr, Mojzisik C. The surgical treatment of morbid obesity. Curr Probl Surg. 1984;21:8-78. 4. Foster WR, Burton BT, Van Itallie TB. Health implications of obesity: National Institutes of Health consensus development conference. Ann Intern Med. 1985;103:977-1077. 5. Drenick EJ. Definition and health consequences of morbid obesity. Surg Clin North Am. 1979;59:963-976. 6. Yale CE. Gastric surgery for morbid obesity: complications and long-term weight control. Arch Surg. 1989;124:941-946. 7. Van Itallie TB. Morbid obesity: a hazardous disorder that resists conservative treatment. Am J Clin Nutr. 1980;33:358-363. 8. Buchwald H, Schwartz MZ, Varco RL. Surgical treatment of obesity. Adv Surg. 1973;7:235-255. 9. Linner JM. A summary of 24 years' experience with surgery for morbid obesity. Am J Clin Nutr. 1980;33:504-505. 10. Van Itallie TB, Burton BT. National Institutes of Health consensus development conferences on surgical treatment of morbid obesity. Ann Surg. 1979;189:455-457. 1. Abraham
11. Buckwalter JA, Herbst CA. Complications of gastric bypass for morbid obesity. Am JSurg. 1980;139:55-60. 12. Buckwalter JA, Herbst GA. Perioperative complications of gastric restrictive operations. Am J Surg. 1983;146:613-618. 13. Peltier G, Hermreck AS, Moffat RE, Hardin CA, Jewell WR. Complications following gastric bypass procedures for morbid obesity. Surgery. 1979;86:648-654. 14. Scopinaro N, Gianetta E, Friedmen D, Adami G, Traverso E, Vitale B. Surgical revision of the bilio-pancreatic diversion. Gastroenterol Clin North
Am. 1987;16:529-531. 15. Yale CE. Conversion surgery for morbid obesity: complications and long-term weight control. Surgery. 1989;106:474-480. 16. Linner JH. Revisional surgery. In: Surgery for Morbid Obesity. New York, NY: Springer-Verlag NY Inc; 1984:109-126. 17. Forse RA, Deitel M, MacLean LD. The conversion of a horizontal to vertical banded gastroplasty: a hazardous procedure. Gastroenterol Clin North Am. 1987;16:533-535. 18. MacLean LD. Surgery for obesity: where do we go from here? Am Coll
Surg Bull. 1989;74:10-23.
19. Scopinaro N, Gianetta E, Pandolfo Minerva Chir. 1976;31:560-566.
N,
et al.
Biliopancreatic bypass.
Discussion
STABILE, MD, San Diego, Calif: The high morbidity and mortality attending late reoperation after weight-reducing surgery for the morbidly obese is often ignored. Since JIB operations are becoming more obsolete, relatively little information has been forth¬ coming on the late reoperation rates and risks following gastric bypass, gastroplasty, and the BPBP procedure. These operations have long-term failure rates and serious complications that have been minimized or not addressed at all. The true outcome data for these sometimes casually applied procedures would be particularly sober¬ ing. I remain to be convinced that bariatric surgery as widely prac¬ ticed under the currently accepted eligibility criteria has durable benefits that clearly outweigh the perioperative and long-term mor¬ tality and morbidity. This is most particularly the case if one includes failure to sustain an acceptable weight loss among the numerous complications. Experience has shown that the laudable results from centers with particular interest and expertise in bariatric surgery are attributable in substantial part to the interest, expertise, and vigi¬ lance in the dietary management and counseling that postoperative patients receive for the duration of their lives. Current investigations into the metabolism of adipocytes and the neuroendocrine control of appetite and satiety offer the promise of new nonsurgical approaches to both prevention and management. BRUCE E.
Specific regulatory abnormalities in triglycéride storage and mobili¬ zation and in plasma cholecystokinin levels have been identified in a variety of pathologic states including obesity, and therapeutic strate¬ gies are already being devised. It is disturbing that a surgical proce¬ dure as deforming and potentially morbid as the BPBP has evolved as treatment for a fundamentally metabolic or psychologic disorder. Hopefully, pharmacologie manipulations will relegate such high-risk operations to the bariatric museum shelf. How often do patients require late surgical intervention for compli¬ cations or failure of bariatric operations? Are these procedures being applied too casually and by unqualified practitioners? Is follow-up often grossly inadequate? The 41% rate of referred patients requiring urgent intervention raises the question of neglect or abandonment of these difficult patients who need lifelong attention and support. Another issue is the extremely high rate of multiple reoperations and their attendant morbidity and mortality. Do the authors believe these results to be representative and expected? These data do not support the notion that bariatric surgery renders obese people safe for future operations. Do the authors believe the BPBP to be a justifiable operation for morbid obesity? Do they accept the criterion of a body weight 45 kg over the ideal weight as a valid indication for a weight-reducing
Downloaded From: http://archsurg.jamanetwork.com/ by a DALHOUSIE UNIVERSITY-DAL-11762 User on 06/19/2015
operation
in the absence of other specific complications? Do they believe that the long-term salutary effects outweigh the combined total short- and long-term morbidity and mortality attending baria¬ tric operations? DAVID TAPPER, MD, Seattle, Wash: Thirty-four percent of the patients had JIBs. Was the operation performed as their only opera¬ tion, or did they come from some other group? Were the complica¬ tions uniform across the group? The patients with BPBPs had a substantial number but where did the complications, particularly the metabolic ones, relate to the JIB? Could you break the major and minor complications down? Were there wound complications such as hernias? Were they a result of wound infection? CLAUDE H. ORGAN, Jr, MD, Oakland, Calif: Many operations applauded in one decade have been discarded in the next. Bariatric surgery is such an example. The second most difficult decision in surgery is when to operate, the most difficult one being when to reoperate. These patients present a challenge for reoperation. The principal fault of bariatric surgery has been that the operations have been too easy. The jejunoileal bypass was too easy. The stapler made the gastric bypass easy. It was our experience that two thirds of the patients who came in for reoperation were malnourished but could be brought to a good nutritional state before reoperation. The biliary pancreatic bypass is a Mann-Williamson procedure (Ann Surg. 1923;77:409-422). It conforms very closely to what Mann and William¬ son described from the Mayo Clinic in the 1920s. Why is there such a high percentage of bowel obstructions in these patients? The second question has to do with the diabetic patient. What was your experience with diabetes? These are patients who have an incredible turnaround in their insulin requirements after surgery. Even before they left the hospital, some required no insulin, hypogly¬ cémie agents, or dietary management. Did you have any experience with these groups, and were you reluctant to correct their bypass and return the gastrointestinal tract to normal? DR ROSLYN: I hope that the general surgeons in the audience are interested in bariatric surgery, not necessarily as a clinical interest, but as an area that all general surgeons should be familiar with. In an article in the September 1989 Bulletin of the American College of Surgeons, Lloyd McLean, MD clearly addresses many of the problems that deal with bariatric surgery as a discipline. Dr Stabile, in the January 1990 issue of Surgery, Dr McLean suggests that in their series of 200 patients undergoing VBG, the reoperation rate was 36%. There truly are a large number of patients who undergo bariatric procedures and ultimately require reopera¬ tion. In most cases, the indication relates to insufficient or excessive weight loss. It is hard for me to comment on follow-up. When we were actively doing bariatric procedures at our institution, the only way we would operate on any of these patients was if they were willing to be seen and followed up postoperatively by a specialist in metabolism. Are these results representative, and how serious are some of these complications? Let me describe three of the more difficult patients who were treated in our series. A young lady underwent a
VBG at an outside institution and developed a leak from her staple line. She was reoperated on several times and was finally transferred to UCLA in florid sepsis with a huge gastrocutaneous fistula. Subse¬ quent tests done at UCLA demonstrated that she had a small pitu¬ itary tumor and that her obesity was really due to Cushing's disease. She died as a result of ongoing sepsis. A second patient had a BPBP performed at an outside institution, and was reoperated on 20 months later because of problems with diarrhea. At reoperation, several enterotomies were made and she was ultimately transferred to UCLA with eight fistulas. She was hospitalized for 5 months before we were able to reconstruct her gastrointestinal tract by taking down her BPBP and restoring her normal anatomy. The final patient had a BPBP performed in Italy. She came back to the United States and ultimately required takedown of her bypass revision. A gastroduodenostomy attempt resulted in anastomotic breakdown. She was transferred to UCLA with no abdominal wall and 16 fistulas. She was hospitalized for 8 months before her defini¬ tive surgery was done. Dr Stabile, BPBP does achieve weight loss; but in terms of safety, it is still experimental, and further data are needed to establish its role. I will not address the issue as to whether patients who are morbidly obese should have bariatric procedures. If this discipline is to survive in the surgical world, it needs to focus on results not just in terms of weight loss; they need to look hard at results in terms of metabolic and postoperative complications. Most of the bariatric procedures have become very easy. About half of the bowel obstructions were really gastric outlet obstructions and the remaining patients had small-bowel obstructions. All the small-bowel obstructions occurred in patients who had a BPBP. The very long distal Roux-en-Y gastrojejunostomy obviously has a great propensity for internal herniation. These operations are very effec¬ tive in reversing diabetes in the morbidly obese patient. In my experience, a number of patients who were receiving 60 units or more of isophane insulin suspension per day were able, within 2 months, to have insulin therapy discontinued completely. The last bariatric operation I performed was about 5 months ago. It was in a 181-kg physician who had such profound sleep apnea that he was dozing off while I was interviewing him in my office. Within 3 weeks after he was discharged from the hospital, he had absolutely no evidence of sleep apnea. Dr Tapper asked about JIBs. All the patients undergoing a reoper¬ ation for JIB were referred from other institutions. We have not performed a JIB at UCLA in the last 14 to 15 years. Most of the catastrophic complications were in patients who had undergone a BPBP. Most of the wound complications were septic in nature and most of the hernias occurred in that setting. At UCLA our experience with "virgin" bariatric procedures has really been limited because it has not been a major focus. The only patients that I see have had bariatric procedures and now have problems, not in terms of requiring revision for weight-related prob¬ lems, but other sorts of problems.
Downloaded From: http://archsurg.jamanetwork.com/ by a DALHOUSIE UNIVERSITY-DAL-11762 User on 06/19/2015
