Re-examining the BMI threshold for bariatric surgery in the USA

BACKGROUND

The optimal BMI threshold above which gastric bypass surgery should be offered to obese patients is controversial. The objective of this study was to compare the impact of Roux-en-Y gastric bypass (RYGB) vs. diet and exercise (D&E) on life expectancy to find the BMI at which patients experience an improvement in their life expectancy by undergoing surgery.

METHODS

A Markov state transition model was designed to implement a decision tree that simulated the lives of obese patients. Life expectancies following RYGB and 2 years of D&E were estimated and compared. Ten thousand patients' lives were simulated in each weight-loss intervention group in the model. In addition to base case analysis (45 kg/m(2) BMI pre-intervention), sensitivity analysis of initial BMI at the start of the study was completed. Markov model parameters were extracted from the literature.

RESULTS

The impact of RYGB on survival relative to D&E depended on the patient's initial BMI. Compared to patients who underwent 2 years of "optimal" diet and exercise (7 % total body weight loss/year), RYGB improved long-term survival for patients above a BMI of 31.3 kg/m(2).

CONCLUSIONS

Roux-en-Y gastric bypass can improve long-term survival for patients with class I obesity. This study suggests that RYGB should not be reserved solely for patients with class II or III obesity.

Routine laparoscopic ultrasound can significantly reduce the need for selective intraoperative cholangiography during cholecystectomy

BACKGROUND

The use of intraoperative cholangiography (IOC), routinely rather than selectively, during laparoscopic cholecystectomy (LC) is controversial. Recent findings have shown laparoscopic ultrasound (LUS) to be safe, quick, and effective not only for screening of the bile duct for stones, but also for evaluating the biliary anatomy. This study aimed to evaluate, on the basis of the LC outcome and the cost of LUS and IOC, whether and how much the routine use of LUS would be able to reduce the need for IOC.

METHODS

During LC, LUS was used routinely to screen the bile duct for stones and to evaluate the biliary anatomy, whereas IOC was used selectively only when LUS was unsatisfactory or unsuccessful.

RESULTS

For 193 (96.5%) of 200 patients, LUS was completed successfully, whereas IOC was needed for 7 patients (3.5%). Bile duct stones were identified in 20 patients (10%). For the detection of bile duct stones, LUS yielded 19 true-positive, 175 true-negative, 0 false-positive, and 1 false-negative results. It had a sensitivity of 95%, a specificity of 100%, a positive predictive value of 100%, and a negative predictive value of 99.4%. The postoperative complications included bile leaks from the liver bed in two patients and a retained bile duct stone in one patient. If IOC had been used selectively in a traditional manner on the basis of preoperative risk factors, IOC would have been needed for 77 patients (38.5%). The total cost of LUS plus IOC for the current 200 patients was 26,256 dollars. The total estimated cost of selective IOC, if it had been performed for the 77 patients, would have been 31,416 dollars.

CONCLUSIONS

Routine LUS accurately diagnosed bile duct stones and significantly reduced the need for selective IOC from a potential 38.5% to an actual 3.5% without adversely affecting the outcome of the LC or increasing the overall cost. The routine use of LUS during LC is accurate and cost effective.

Hand dominance and performance in a laparoscopic skills curriculum

BACKGROUND

This study evaluated the influence of hand dominance on skill acquisition during a basic laparoscopic skills curriculum.

METHODS

A total of 27 surgical residents (5 postgraduate year 3 [PGY-3] and 22 PGY-2 residents) participated in a 4-week laparoscopic skills curriculum. The residents were pre- and posttested on six laparoscopic tasks during weeks 1 and 4. During weeks 2 and 3, the residents attended a proctored practice session. The results were compared using analysis of variance (ANOVA), (with significance determined by a p value less than 0.05.

RESULTS

The posttest scores were significantly higher than the pretest scores. On the pretest, lefthand-dominant (LHD) surgeons (n = 4) performed significantly better than righthand-dominant (RHD) surgeons (n = 23). In the analysis of individual task pretest scores, LHD surgeons performed significantly better on pattern cutting and vessel loop application. Posttest analysis of overall performance did not show significant differences between the RHD and LHD surgeons.

CONCLUSIONS

Participation in a laparoscopic skills curriculum improved overall performance. The LHD surgeons demonstrated better initial performance, but posttest comparison showed no difference between the two groups.

Decreased type IV collagenase activity in experimental pancreatic fibrosis

BACKGROUND

Severe hyperstimulation and duct obstruction pancreatitis (SHOP) is characterized by pancreatic fibrosis and loss of acinar cell mass. MMP-2 and MMP-9 are type IV collagenases and gelatinases. We hypothesized that fibrosis results from disruption of the normal collagen homeostasis and that altered activity of the type IV collagenases may contribute to pancreatic fibrosis in SHOP.

METHODS

SHOP rats (n = 15) were prepared with pancreatic duct obstruction and cerulein (50 microg/kg/d, ip) hyperstimulation. Pancreas from unoperated control (n = 8), 48 h SHOP (n = 8), and 96 h SHOP (n = 7) rats was harvested, homogenized, and assayed for protein concentration (BCA method). Type IV collagenase (MMP-2 and MMP-9) expression was measured by zymography using gelatin as substrate. Type IV collagenase activity was quantified with a fluorescence assay.

RESULTS

Expression of the active form of MMP-9 decreased while latent MMP-9 and active and latent MMP-2 increased on gelatin zymography. Activity of type IV collagenases (MMP-2 and MMP-9) progressively decreases with SHOP injury. The differences between expression and activity are likely due to posttranslational regulators such as MT-MMPs and TIMPs.

CONCLUSIONS

Collagenase expression and activity are decreased in the SHOP model of pancreatitis, suggesting a decrease in the homeostatic mechanisms for type IV collagen in the extracellular matrix. Therefore, early fibrosis in the SHOP model is, at least in part, due to alterations in collagen homeostasis and not simply increased collagen production.

A state-wide evaluation of appendectomy in children

BACKGROUND

Traditional management of appendicitis in children involves open appendectomy (OA), an operation that is relatively inexpensive and carries few risks and complications. However, little information is available regarding the use, cost, and complication of laparoscopic appendectomy (LA) in children.

METHODS

Our initial aim was to determine if LA is frequently performed in children (<15 years). We then compared the surgical results of OA versus LA. In conjunction with the Missouri Department of Health, we evaluated 793 children treated for appendicitis throughout the state between January 1997 and June 1997. The authors were blinded to the patient, surgeon, and hospital; no children were excluded.

RESULTS

LA was infrequently performed in children with advanced disease. Overall, children undergoing LA were older and had a shorter hospitalization but no difference in hospital charge. When separated by child age, LA was associated with a shorter length of stay in all groups (0 to 5, 6 to 10, and 11 to 15 years) but only children in the 6 to 10 year range had a lower hospital charge when compared with patients undergoing OA.

CONCLUSIONS

LA is becoming a common surgical approach for older children with simple appendicitis. Furthermore, these data suggest that LA, independent of individual surgeon or medical center, is associated with a decreased length of hospitalization without a significant difference in hospital charge.

Characterization of a novel model of pancreatic fibrosis and acinar atrophy

Significant fibrosis and acinar atrophy are characteristics of chronic pancreatitis; however, because of the lack of a reproducible model, early phases of these changes are poorly understood. We have developed a model of severe hyperstimulation and obstruction pancreatitis (SHOP) to better define the mechanisms of early pancreatic fibrogenesis. Sprague-Dawley rats were used and SHOP was induced by complete pancreatic duct obstruction and daily cerulein hyperstimulation (50 microg/kg intraperitoneally). Animals were killed at 24, 48, 72, and 96 hours. Control animals underwent sham operation and received no cerulein. Pancreata were prepared for hematoxylin and eosin and sirius red (collagen-specific) staining and for hydroxyproline assay (measure of total collagen content). We found moderate amounts of edema and inflammation but minimal parenchymal necrosis. Significant loss of acinar cell mass was noted by 48 hours, and normal acinar cells were essentially absent by 96 hours. Tissue collagen content increased with time and large amounts of interstitial collagen were detected by 72 hours. In conclusion, SHOP is a novel model of early pancreatic fibrosis associated with minimal necrosis and a significant decrease in acinar cell mass, making it an ideal model to study the early cellular mechanisms of pancreatic fibrogenesis.

Staple occlusion of the infrarenal aorta: a videoscopic retroperitoneal approach

PURPOSE

The purposes of this study were to determine whether available laparoscopic stapling devices could be used to interrupt the diseased human aorta, and to develop a videoscopic technique for retroperitoneal exposure and control of the infrarenal aorta in pigs. Our long-term goal is to develop a minimally invasive approach to the treatment of abdominal aortic aneurysms by exclusion and extraanatomic bypass.

METHODS

Ten diseased, formalin-preserved human cadaver aortas underwent stapling using a laparoscopic stapling device. The aortas were then pressurized to superphysiologic levels to assess the integrity of the staple line. Ten swine underwent retroperitoneal video-assisted exploration with control and staple occlusion of the aorta and iliac artery.

RESULTS

The staple line was complete and remained intact after pressurization in nine of 10 cadaver aortas, despite the presence of complex calcified disease. One aorta had a 2-mm opening through the staple line. Through the left retroperitoneal approach, the infrarenal aorta and left iliac artery could be dissected and controlled. A modified pledgeted technique used for stapling resulted in hemostasis of the staple line and exclusion of flow without injury to adjacent structures.

CONCLUSIONS

The diseased human aorta can be occluded using available laparoscopic staplers. These swine experiments demonstrate the feasibility of the retroperitoneal approach for exclusion of infrarenal aortic aneurysms.

Laparoscopic gastrostomy and jejunostomy are safe and effective for obtaining enteral access

BACKGROUND

Laparoscopic gastrostomy (lap g-tube) and jejunostomy (lap j-tube) are relatively new procedures that do not require laparotomy. Our aim was to determine the role of laparoscopic feeding tube placement for enteral access and the safety of these techniques.

METHODS

We reviewed our experience with attempted laparoscopic placement of 93 enteral tubes in 81 patients over a 3-year period. Patients received either a lap g-tube (n = 64), lap j-tube (n = 5), or both lap g/j-tube (n = 12). When enteral access was needed for nutritional support, the choice of lap g-tube or lap j-tube was based on risk of gastroesophageal reflux.

RESULTS

The most common underlying conditions requiring tube placement were head and neck cancer (49%), neurologic disorders (19%), and trauma (11%). Mean operative times (minutes) were as follows (mean +/- SD): lap g-tube (39 +/- 7), lap j-tube (63 +/- 10), and lap g/j-tube (85 +/- 13). Lap g-tube placement was successful in 73 (96%) of 76 patients and lap j-tube in all 17 patients. The major complication rate for all tubes was 8% (7 of 93) and included gastrointestinal bleeding, wound infection, and failed placement. Five patients died in the 30-day period following surgery, but none of the deaths was procedure related.

CONCLUSIONS

Laparoscopic tube placement should be considered for patients in whom endoscopy is not feasible or undesirable or who are undergoing other operative procedures. Lap g-tube and lap j-tube are safe procedures that avoid the potential risk of a laparotomy, and they can be done with a high success rate. This is a valuable approach for patients with head and neck cancer or neurologic disorders and for trauma patients with multiple disease processes.

Video-assisted, retroperitoneal approach for abdominal aortic aneurysm exclusion

Abdominal aortic aneurysm (AAA) repair is a common procedure associated with significant morbidity and mortality. Although attempts have been made to reduce operative risk in patients with significant comorbid disease by combining aneurysm exclusion with axillofemoral bypass, the morbidity is not greatly reduced when the standard operative approach is required for exclusion. The authors describe a technique for staple exclusion of AAA using a minimally invasive, video-assisted retroperitoneal approach.

Laparoscopic gastrostomy versus open gastrostomy in head and neck cancer patients

Alimentation in the surgically treated head and neck cancer patient frequently requires bypassing the upper aerodigestive tract. The laparoscopic gastrostomy fulfills this criterion. The authors compared 25 laparoscopic gastrostomies (group 1) with 18 open gastrostomies (group 2) performed on head and neck cancer patients. The length of operation, morbidity, mortality, and cost were evaluated. Operative time was significantly shorter in group 1 (40 +/- 2 minutes) than in group 2 (56 +/- 4 minutes), with P=.003. The major complication rate was 9% for group 1 and 11% for group 2. There was no procedure-related mortality in group 1, but 1 patient died in the immediate postoperative period in group 2. The cost was not significantly different. It is concluded that the laparoscopic gastrostomy is a safe and cost-effective alternative to open gastrostomy in this patient group.

Laparoscopic gastrostomy: a safe method for obtaining enteral access

Laparoscopic gastrostomy is a relatively new procedure which does not require a laparotomy but which allows direct visualization of the gastrostomy tube entering the stomach and the stomach-parietal peritoneum interface. The role of the laparoscopically placed gastrostomy for enteral access has yet to be defined. Our aim was to compare laparoscopic gastrostomy with open surgical gastrostomy to determine the effectiveness and safety of laparoscopic gastrostomy as a method for obtaining long term enteral access. We reviewed the records of 32 patients (December 1992 to April 1994) who had laparoscopic gastrostomy (group I) and 37 patients (January 1987 to December 1993) who had open gastrostomy (group II). The indications, length of operation, and morbidity and mortality rates in both groups were compared. The underlying diseases and indications for gastrostomy placement were similar in both groups. Both groups included primarily patients who were not candidates for upper endoscopy. Operative time was significantly shorter in the laparoscopic gastrostomy group (38 +/- 7 min) than in the open gastrostomy group (62 +/- 19), P < 0.0001. Major complication for laparoscopic gastrostomy was 6% and for open gastrostomy was 11%. There was no procedure-related mortality in group I and three patients in group II died in the immediate postoperative period. Major morbidity and mortality rates were not significantly different between the two groups. Laparoscopic gastrostomy is a safe and effective alternative to open gastrostomy, particularly in patients unable to undergo upper endoscopy or in patients undergoing a concomitant laparoscopic procedure.

Pancreatic complications after distal splenorenal shunt

Pancreatic complications after the distal splenorenal shunt have not been commonly recognized. Between January 1978 and June 1993, 154 patients underwent a distal splenorenal shunt, and 11 patients (7%) developed pancreatic complications, of which 4 had pancreatitis alone, and 7 developed pancreatitis-related complications. Etiology of cirrhosis, Child's classification and timing of surgery were not predictive of pancreatic complications. Eight patients (5%) were found to have chronic pancreatitis at the time of surgery, and four of these patients (50%) developed pancreatic complications following distal splenorenal shunt. Eleven early postoperative deaths in our series resulted in an overall operative mortality rate of 7%. Of these eleven patients, 6 (55%) had postoperative pancreatic complications. The operative mortality rate of patients who developed pancreatic complications (55%) after distal splenorenal shunt was significantly greater than that of patients who did not develop pancreatic complications (3%), P < .001. When compared with patients without pancreatitis, those with pancreatitis had significantly greater incidences of complete or partial portal vein thrombosis (55% v 20%, P < .02), severe ascites (64% v 13%, P < .001), and encephalopathy (45% v 3%, P < .001). We reach the following conclusions: (1) although not a frequent complication after distal splenorenal shunt in general, pancreatitis was commonly present in early postoperative deaths and was most likely a major contributor to the demise of those patients; (2) survivors with postdistal splenorenal shunt pancreatitis had a markedly increased morbidity rate; and (3) pancreatic complications after distal splenorenal shunt are more likely to occur in patients with pre-existing chronic pancreatitis.

Pancreaticoduodenectomy for suspected but unproven malignancy

BACKGROUND

Pancreaticoduodenectomy is an accepted surgical option for certain benign conditions and biopsy proven cancer. Whether this procedure should be performed when malignancy of the pancreas and periampullary region is suspected but not confirmed represents a fairly common intraoperative dilemma.

PATIENTS AND METHODS

Sixty-seven patients who had undergone pancreaticoduodenectomy during a 15-year period were evaluated retrospectively.

RESULTS

The indications for resection were symptomatic benign conditions (n = 10, 15%), proven pancreatic or periampullary cancer (n = 37, 55%), and suspected but unproven malignancy (n = 20, 30%). The patients with suspected malignancy ranged in age from 27 to 73 years. Common findings in this group were abdominal pain (75%), jaundice (70%), weight loss (65%), and alcohol use (45%). There were 14 pancreatic and 6 ampullary masses. Biopsies obtained preoperatively (n = 15) and intraoperatively (n = 11) were nonconfirmatory. Postoperatively 9 patients (45%) were found to have tumors, including 6 pancreatic adenocarcinoma, 2 duodenal adenocarcinoma, and 1 islet cell tumor. Six of the 8 adenocarcinomas (75%) were stage I. Seven patients were alive 11 to 108 months later. The most common benign diagnosis was pancreatitis. There were 8 complications and 1 death.

CONCLUSIONS

Pancreaticoduodenectomy performed based on suspicion alone frequently reveals malignancy. Immediate and long-term outcomes are acceptable. These findings justify a continued aggressive approach to suspected pancreatic and periampullary malignancy.

Increased circulating cholecystokinin in obstruction-induced acute pancreatitis. I. Bile duct obstruction with and without pancreatic duct obstruction

Bile exclusion from the gut exacerbates pancreatic duct obstruction-induced acute pancreatitis. We hypothesized that obstruction-induced acute pancreatitis involves an increase in circulating cholecystokinin (CCK), as bile and pancreatic juice exclusion from the gut stimulates duodenal CCK release. We studied 54 rats after the following operations: (1) sham operation (n = 18), (2) hepatic bile duct obstruction alone (n = 18), (3) hepatic bile duct and common bile-pancreatic duct obstruction (n = 18). Rats recovered and were killed in subgroups of six rats each at 3, 6, and 18 hr after operation; blood was collected for measurement of plasma CCK and amylase concentrations. Each pancreas was excised, weighed, and processed for histological examination; an acute pancreatitis score was determined. Combined bile and pancreatic duct obstruction induced acute pancreatitis and was associated with a marked increase of circulating CCK concentration. Bile duct obstruction alone did not induce acute pancreatitis but was associated with an increase of circulating CCK of lower magnitude. The time course of circulating CCK increase showed an early peak. These findings support our hypothesis and suggest that CCK plays a role in the pathogenesis of obstruction-induced acute pancreatitis.

Bile exclusion from the gut exacerbates acute pancreatitis caused by pancreatic duct obstruction in rats

Acute pancreatic duct obstruction causes hyperamylasemia and mild pancreatic inflammation. We hypothesized that bile exclusion from the gut, which stimulates pancreatic secretion, exacerbates acute pancreatitis caused by pancreatic duct obstruction. Rats were surgically prepared with gastric, duodenal, bile, and pancreatic fistula catheters and a jugular vein catheter. After a 4-day recovery, groups of rats (a) served as controls, (b) had complete pancreatic duct obstruction for 6 h, or (c) had bile excluded from the gut for 24 h and then, during the final 6 h, complete pancreatic duct obstruction. Plasma amylase was measured, and all rats were euthanized at the end of experiments. Each pancreas was excised and weighed, and portions were fixed in formalin and glutaraldehyde. In blind fashion, each pancreas was examined under light microscopy and assigned a pancreatitis score based on presence of edema and severity of acinar cell changes and inflammation. Acute pancreatic duct obstruction was associated with increased pancreas weight, hyperamylasemia, and elevated pancreatitis score; moderate acinar cell vacuoles, which were observed in the cytoplasm near the basolateral membrane, and loss of microvilli were noted with electron microscopy. Bile exclusion from the gut exacerbated the acute pancreatitis caused by pancreatic duct obstruction; acinar cell distortion and destruction, as well as marked inflammation, were seen microscopically. These observations suggest that the absence of intestinal bile contributes to the pathogenesis of acute pancreatitis associated with pancreatic duct obstruction.

Does somatostatin analogue prevent experimental acute pancreatitis?

Because somatostatin is a potent inhibitor of pancreatic secretion, we hypothesized that pretreatment with somatostatin analogue octreotide (SMS 201-995) might prevent cerulein-induced edematous pancreatitis. We studied 18 rats prepared with jugular vein catheters. The following agents were administered intravenously to groups of four rats for 6 hours: 1 mL/h (control) crystalloid solution; 1-microgram/kg bolus then 1 microgram/kg per hour of octreotide; and 5 micrograms/kg per hour of cerulein; also, in a fourth group of six rats, octreotide and cerulein were administered simultaneously. At the end of experiments, blood was drawn for plasma amylase determinations; rats were killed and pancreata were examined. Supramaximal cerulein administration to conscious rats induced hyperamylasemia and edematous pancreatitis, confirming previous observations; in both groups of rats receiving cerulein, there was prominent interstitial edema, acinar vacuolization, and mild-to-moderate acute inflammation. While octreotide pretreatment of rats with cerulein-induced acute pancreatitis was associated with a lesser increase of wet pancreas weight and plasma amylase concentration, there was little overall benefit of octreotide pretreatment in this form of experimental acute pancreatitis.

Cholecystokinin antagonist prevents hyperamylasemia and improves pancreatic exocrine function in cerulein-induced acute pancreatitis

Supramaximal cerulein administration induces acute pancreatitis, which markedly impairs pancreatic secretion in conscious rats. We hypothesized that pretreatment with the potent cholecystokinin antagonist, L-364,718, improves the pancreatic secretory impairment associated with cerulein-induced acute pancreatitis. Rats were surgically prepared with gastric, duodenal, bile, and pancreatic fistulas and jugular vein catheters. On postoperative day 4, groups of rats were administered (a) L-364,718 1 mg/kg intraduodenally, (b) cerulein 5 micrograms/kg/h for 6 h intravenously, (c) L-364,718 1 mg/kg intraduodenally followed by cerulein 5 micrograms/kg/h for 6 h intravenously, and (d) safflower oil carrier intraduodenally. On postoperative day 5, we studied cholecystokinin (CCK)-stimulated pancreatic secretion. Plasma amylase was measured at the time of surgery and at the conclusion of experiments on postoperative days 4 and 5. The duodenally administered CCK antagonist had no effect, 24 h later, on CCK-evoked protein secretion and prevented the pancreatic exocrine impairment and hyperamylasemia caused by supramaximal cerulein administration. These observations suggest that cerulein-induced acute pancreatitis is mediated by a CCK-receptor mechanism.

Acute edematous pancreatitis impairs pancreatic secretion in rats

There are few observations of in vivo pancreatic secretory changes that accompany acute pancreatitis. We hypothesized that acute pancreatitis impairs pancreatic exocrine function. We developed a conscious-rat experimental preparation with gastric, duodenal, bile, and pancreatic fistulas. We studied cholecystokinin-stimulated pancreatic secretion in conscious rats before and after inducing acute pancreatitis with supramaximal administration of caerulein--5 micrograms/kg/hr intravenously for 6 hours. Marked hyperamylasemia developed in all rats immediately after administration of caerulein. Basal and cholecystokinin-stimulated pancreatic juice flow and protein (enzyme) secretion decreased significantly 24 hours after acute pancreatitis was induced even though plasma amylase returned to basal levels. We conclude that acute pancreatitis markedly impairs pancreatic secretion.