Search

Total parenteral nutrition (TPN) for the nonoperative treatment of acute pancreatic pseudocyst has been of hypothetical benefit. We reviewed pseudocyst hospital admissions in 40 patients treated with TPN who had serial imaging studies. The mean cyst size was 7.4 cm on presentation, decreasing to 5.6 cm after nonoperative treatment with TPN (mean 32.5 days). After a nonoperative period, 68 per cent of cysts regressed, completely in 14 per cent, partially in 54 per cent. Except for a patient with cyst-related obstructive jaundice, there were no complicated pseudocysts. Only 12 (28%) patients underwent cyst drainage. Fifteen patients (35%) sustained catheter-related complication, which included sepsis (26%), pneumothorax (9%), hydropneumothorax (2%), and septic right atrial thrombosis (2%), in the course of hospitalization. The majority of TPN-treated patients had a clinical and radiographic regression of their pseudocyst. However, the increased risk of catheter-related complications in this group suggests that this therapy should be limited to patients who are unable to sustain enteral nutrition.

493 African-American children between the ages of three and four years were examined for evaluation of molar occlusion with frequency of anterior and posterior crossbite, fractured and/or darkened teeth from trauma and dental anomalies in the primary dentition. Overall results from both age groups revealed 90% had a mesial step, 8% a flush and 2% a distal step molar occlusion. Crossbite, either anterior or posterior, occurred in 12% of all children examined. Posterior crossbite was seen in 7% and anterior crossbite in 5%. Anterior openbite was seen in 8% of the children. Trauma to the primary dentition occurred in 23% of the children, 16% as fractured and 7% as discolored teeth. The most frequently involved teeth were the maxillary primary central incisors. Fused primary teeth were observed in two children. There were no congenitally missing primary teeth and only one supernumerary maxillary primary lateral incisor was observed.

Buprenorphine is an opioid agonist-antagonist that has emerged as an option for postoperative analgesia. We compared the postoperative hospital course of patients undergoing open cholecystectomy who received buprenorphine hydrochloride with those who received morphine sulfate. Patients in both groups administered the analgesic using a patient-controlled analgesia infusion device. Comparison of the two groups demonstrated no difference with respect to clinical indicators of intestinal motility, visual analog pain scores and hospitalization period. Postoperative nausea occurred more frequently in the buprenorphine group, but the difference was not significant. We concluded that the patient-controlled analgesia device is a valuable tool for comparing different analgesics. Both analgesics tested provide adequate analgesia with a similar postoperative course.

Percutaneous endoscopic gastrostomy (PEG) is clearly better than operative gastrostomy performed under general anesthesia. Whether or not PEG offers any significant advantage over operative gastrostomy performed using local anesthesia remains to be proved. Operative gastrotomy performed with local anesthesia seems comparable to nonoperative techniques.

Previous studies have shown that patient-controlled analgesia (PCA) provides effective pain control in the postoperative patient. To determine the impact of PCA technology on the overall hospital course, we designed a randomized controlled study comparing patients receiving analgesia using PCA infusion (Abbott Lifecare, Abbott Laboratories; Chicago, IL) with patients receiving analgesia by traditional intramuscular or intravenous methods. All patients had undergone elective cholecystectomy. Sixty-nine patients completed the study, 35 received traditional postoperative analgesia, and 34 received analgesia using the PCA infuser. Comparison of both groups demonstrated no significant difference in postoperative bowel activity with both groups receiving liquids on the first postoperative day. There was no significant difference between the two groups with respect to postoperative length of stay (3.4 days for PCA vs 3.6 days for traditional). Patients demonstrated a wide range of analgesic requirement in the first 24 hours but the average of the total analgesic required was higher in the PCA group (average, 29.5 mg) than the traditional group (22.8 mg). Urinary complications occurred more commonly in the group of patients receiving traditional analgesia than in the group of patients receiving analgesia with the PCA device. When compared with patients receiving analgesia by traditional methods, patients receiving the PCA infusion required more analgesia with fewer urinary complications and similar postoperative length of stay.

The indications for highly selective vagotomy have expanded in recent years, with the technique being applied to selected cases of perforation and bleeding. Its use in obstruction is controversial, but two options are available for managing the stenotic pylorus or duodenum: dilatation or duodenoplasty. The latter choice requires that the stenosis be located in the postbulbar area. Since 1981, we have managed 15 patients with postbulbar stenosis by means of highly selective vagotomy and duodenoplasty. All patients had a previous history of ulcer disease, and vomiting was a consistent symptom. All patients were referred for surgery, 10 by a gastroenterologist. There was no operative mortality or procedure-related morbidity. Two patients have been lost to follow-up. Both were classified as Visick I and had normal endoscopic results at their last visit. The remaining 13 patients have all been followed very recently. Twelve patients (92%) are currently classified as Visick I or II. One patient (Visick IV), who was essentially asymptomatic, was found to have a recurrent ulcer on endoscopy. Endoscopic (11 patients) or radiographic (1 patient) patency of the duodenoplasty has been demonstrated in 12 patients. Highly selective vagotomy and duodenoplasty should be a surgical consideration when the pathologic anatomy of the duodenum lends itself to that choice.

Numerous investigators have reported the safety, cost effectiveness, and low morbidity and mortality of percutaneous endoscopic gastrostomy (PEG) but not studies compare the results of gastroenterologists and surgeons performing PEG. In a retrospective review of PEG performed at our institution, morbidity and mortality were compared between these two groups. The procedure was performed by either the surgical service (n = 49) or gastroenterology service (n = 51). One hundred PEGs were successfully placed in 92 patients. Three placement failures occurred in the gastroenterology group. Major complications were defined as conditions requiring operative intervention or resulting in death. Minor complications, using a defined list (13 different complications), were those not serious or life-threatening, which were managed medically or resolved without treatment. Twenty-nine minor (17 patients) complications and 3 major (3 patients) complications occurred in the gastroenterology group. Thirteen minor (11 patients) complications and 4 major (4 patients) complications occurred in the surgery group. Overall 30-day mortality was 14 per cent (13 patients), two of which were probably procedure-related in the gastroenterology group. In both groups, there was no difference in the numbers of patients who had complications (P greater than 0.05). The complication rate (numbers of complications/number of patients) doubled in the gastroenterology group compared with the surgery group for minor (P less than 0.04) and total complications (minor and major--P less than 0.06). Since PEG is a procedure frequently performed by gastroenterologists, it is imperative that surgeons, particularly those not performing endoscopy, be aware of the potential morbidity associated with PEG.

The effect of administration time on the serum concentrations of cefazolin and cefoxitin achieved during surgery was studied. Patients about to undergo elective surgery were randomly selected to receive one of four antibiotic prophylaxis regimens. Two of the regimens consisted of cefazolin 20 mg/kg i.v. (as the sodium salt) administered either "on call" to the operating room or with anesthesia in the operating room. The other two regimens were identical except that cefoxitin 30 mg/kg i.v. (as the sodium salt) was substituted for cefazolin. Intraoperative blood samples for analysis of serum concentrations were obtained at the time of surgical incision, one hour after incision, and at wound closure. Forty patients completed the study (10 patients received each treatment). Mean projected cefazolin concentrations at the time of incision and at wound closure for both two-hour and three-hour operations were significantly higher after administration with anesthesia than after "on-call" administration. For cefoxitin, differences in projected concentrations for "on-call" administration versus administration with anesthesia were significant at the time of incision and at wound closure in a two-hour operation. At times corresponding to initial incision, projected free-cefoxitin concentrations were higher than those of cefazolin; however, projected free-cefazolin concentrations were higher than free-cefoxitin concentrations at wound closure for two-hour and three-hour operations. To ensure that antibiotic is present in serum throughout a two- or three-hour surgical procedure, cefoxitin should probably be administered at the induction of anesthesia. Cefazolin administered "on call" appears to be acceptable for maintaining serum concentrations throughout surgery of this duration.

Twenty-five per cent of the authors' total upper endoscopy experience since 1974 has been in patients who have had upper gastrointestinal tract surgery. The observations from 617 examinations in 400 of these patients is reviewed. Pain or nausea and vomiting was a common presenting symptom. Multiple complaints were frequent. Gastritis was the most common endoscopic finding present in 127 patients (32%). Biopsy yielded an 89% histologic confirmation of the endoscopic perception. Coexisting mucosal pathology was common, with 39% of the patients having two or more abnormalities. X-ray in 190 patients had only a 30% accuracy rate and a frequent occurrence of false negatives (46%). Bezoars and intraluminal sutures were seen commonly and endoscopy provided a therapeutic choice for enzyme injection of the bezoar and removal of the sutures. Endoscopy provided a method of evaluation of our highly selective vagotomy technique; 96% of our patients with ulceration at the time of surgery were healed by endoscopy and 93% had active antral-pyloric function.

Early classification of lower gastrointestinal bleeding as occult, minor overt, or major overt allows a practical approach to evaluation and management. Localization of the bleeding site is the next step. In occult and minor overt bleeding, the cause can usually be determined from results of conventional diagnostic tests; however, angiography and even intraoperative endoscopy may be necessary in some particularly difficult cases. In contrast, major overt bleeding may only allow time for angiographic localization before surgery. Using this approach, the primary care physician can successfully manage most patients with lower gastrointestinal bleeding.

A four year experience with the adaptation of the flexible fiberoptic endoscope to the intraoperative environment is presented in 30 patients. The technique of intraoperative endoscopy was utilized in a wide variety of difficult gastrointestinal surgical problems to include the location of the site and cause of bleeding of obscure etiology; resolution of intraoperative dilemmas without the necessity of opening abdominal viscera; resection of lesions during operations conducted for other pathological processes; and enhancement of diagnosis at laparotomy. There were no complications from the use of intraoperative endoscopy and the technique was beneficial in 28 of the 30 patients (93.3%). Limiting factors in the full utilization of the endoscope at celiotomy were dense adhesions with a shortened mesentery and massive hemorrhage with blood obscuring the intestinal lumen.

For the patient and the clinician, it is well that small-bowel diseases are unusual, as our ability to access this area remains limited. Frequently, all of the diagnostic and therapeutic modalities of radiology, endoscopy, and surgery are required for successful resolution of a given problem. Because management of bleeding from a small-bowel source usually will involve the surgeon at some point, it is mandatory that the best "road map" be obtained prior to exploration. If a small-bowel source is suspected after a negative endoscopic evaluation of the esophagus, stomach, duodenum, and colon, then the clinician must decide which radiographic and endoscopic examination is most appropriate. If bleeding is slow or intermittent, push-type enteroscopy to evaluate the proximal jejunum will have an expected diagnostic discovery rate of about 30 per cent. A stiffening over-tube or internal cable should result in deeper passage of the instrument and a potentially greater yield. Retrograde ileoscopy should be a part of every colonoscopy done for occult bleeding. If endoscopy does not identify a bleeding source, then a detailed barium study of the small bowel using an enteroclysis double-contrast technique will discover more pathology than a standard small-bowel-follow-through. Because sonde-type enteroscopes are not readily available, the clinician must decide at this point whether to refer the patient to an enteroscopist or consider surgery and intraoperative endoscopy. If a bleeding source has been found, then intraoperative endoscopy can localize lesions for specific resection. If the pathology remains obscure, intraoperative endoscopy will have a discovery rate of about 70 per cent. For more active hemorrhage, a bleeding scan with 99mTc-labeled red blood cells can confirm that blood loss is continuing and also will guide the angiographer toward a more directed study, thus decreasing the contrast material load for the patient. If a bleeding source can be identified angiographically, a short course of vasopressin infusion to convert the need for surgical intervention to a more elective situation would be beneficial to the patient. Intraoperative endoscopy under urgent conditions is more difficult, because luminal blood must be lavaged or cleared for a proper examination. Many times, however, intraoperative endoscopy can "surround" a segment of intestine by identifying areas that are clearly normal.

Seventeen patients had rectal bleeding and intestinal vascular ectasia. Actual bleeding from the ectasia was seen in five patients. The average age of the 17 patients was 60.2 years, but 41% were under the age of 60. Melena was common initially. Both colonoscopy and angiography had a high degree of diagnostic accuracy, but angiography failed to demonstrate some lesions seen by colonoscopy. We consider these examinations complementary. Intraoperative endoscopy with transillumination was of value of determining the distribution of those lesions that had not been completely demonstrated by colonoscopy and angiography. We have used intraoperative endoscopy to guide the eventual amount of bowel resection, and intraoperative marking with this technic has resulted in a more specific pathologic examination. There has been no operative mortality. Six patients have re-bled from sources other than vascular ectasia.

We determined the intraoperative serum and wound-muscle concentrations of cefazolin and cefoxitin in 40 patients who were undergoing cholecystectomies. The study employed an open-label design in which all of the patients randomly received cefazolin sodium (20 mg/kg) or cefoxitin sodium (30 mg/kg) intravenously while the patient was in the ward ("on call") or with the induction of anesthesia. Multiple blood and wound-muscle samples were collected intraoperatively and assayed for their cephalosporin concentrations. Considerable differences in intraoperative serum and tissue concentrations between antibiotics were apparent; there were usually higher levels of cefazolin. In all of the patients who received cefazolin sodium, the antimicrobial was detectable in wound tissue at wound closure, while it was detectable in 86% and 38% of patients who received cefoxitin sodium with anesthesia and on call, respectively. Because cefoxitin has a much shorter elimination half-life than cefazolin it seems prudent to administer the agent as close to the start of the operation as possible, and readminister the agent every two to three hours until the wound is closed. For cefazolin, on-call administration appears to be acceptable, with readministration not required for at least four hours.

The physiology, solution composition, indications, efficacy, and safety of whole-bowel irrigation (WBI) for mechanical bowel cleansing are reviewed. WBI with isotonic electrolyte solutions produces diarrhea when the infusion rate exceeds the capacity of the intestine to distend and absorb the solution. A number of solutions are used for WBI, including 0.9% sodium chloride, balanced electrolyte solutions, lactated Ringer's, mannitol, and electrolyte solutions containing polyethylene glycol 3350 (PEG). WBI solution administration rates vary from 15-90 mL/min, by oral ingestion or nasogastric tube, with total volumes ranging from 1 to 20 L. The onset of diarrhea occurs as soon as 20 minutes with clearing of the effluent as early as 90 minutes. Faster administration rates appear to shorten overall cleansing time. Two PEG-electrolyte lavage solutions (ELSs) have recently gained FDA approval. The recommended dosage rate is 1.2-1.8 L/hr orally or by nasogastric tube until rectal effluent is clear. In most patients, this requires a maximum of 4-6 L. Initial data indicate that PEG-ELSs are safe for elderly patients and for patients who have an increased risk of fluid overload, but these solutions have not been evaluated in children, pregnant women, or patients with inflammatory bowel disease. WBI is an effective alternative to other regimens for removing fecal material and reducing bowel lumen bacterial counts before colonoscopy and colorectal surgery. Retention of bacterial counts before colonoscopy and colorectal surgery. Retention of excess WBI solution may interfere with the quality of barium enema radiographs; this can be minimized by completing the irrigation the evening before the examination. Gastrointestinal side effects occur in about one third of the patients following WBI, but do not generally require discontinuing the irrigation. Solutions containing PEG with sodium sulfate as the primary electrolyte result in the least net water and electrolyte movement and are preferred over other solutions.

We have reported a case of intraduodenal duct disruption and duodenal hematoma associated with ERCP. The location of the extravasated contrast material created a confusing picture, which preoperatively was thought to be a pseudocyst, but was found at operation to be a duodenal hematoma. In retrospect, the upper gastrointestinal series done immediately after ERCP was very helpful. We believe that an upper GI series done immediately after ERCP will help determine the location of unusual collections of contrast material seen at ERCP.

Highly selective vagotomy (HSV) is an accepted choice for the treatment of uncomplicated duodenal ulcer. Its use in patients with gastric outlet obstruction, however, remains quite controversial. Since 1980, 69 patients have undergone HSV at the Medical College of Georgia Hospitals. Of these, 20 (29%) underwent either dilatation (14) or duodenoplasty (6) for accompanying outlet obstruction. The obstruction was graded as severe in 17 (85%) and moderate in three (15%). Follow-up evaluation has included Visick grading and endoscopy. There have been two deaths (38 and 54 months following surgery). Both patients were Visick I. Of the 18 patients available for review to date, 12 (67%) are Visick I and four (22%) are Visick II. There have been two failures (11%), discovered only by endoscopic follow-up in asymptomatic patients. No patients have required reoperation. HSV with dilatation or duodenoplasty is a reasonable surgical alternative for the treatment of obstructing duodenal ulcer disease.

Enteral nutrition is best delivered via a small bore feeding tube whose tip lies in the proximal jejunum. A major obstacle to tube placement is the lack of a reliable means of assuring passage through the pylorus. A simple, quick method of tube placement using endoscopic assistance that was successful in 18 of 20 (90%) attempts is described.

The proper duration of antimicrobial use for the prevention of postoperative surgical infection has been a subject of controversy. Currently, more than 40 published clinical trials are available in which the efficacy of single dose surgical prophylaxis with parenteral antimicrobials has been studied. These studies have compared single doses versus multiple doses of the same agent, single doses of antimicrobial versus placebo, single doses of various antimicrobials, and a single dose of one agent versus multiple doses of another agent. In all trials in which single dose regimens were compared with multidose regimens, the single dose regimens resulted in a similar frequency of postoperative wound infections. Single antimicrobial doses, usually cephalosporins given immediately before operation, are effective in preventing wound infections in gastric, biliary, and transurethral operations, hysterectomies and cesarean sections. For colorectal operations, the value of single parenteral doses of various agents has been established; however, it is not clear if there is an added benefit when oral antimicrobials are also used. For open heart operations or those in which prosthetic materials are implanted, the value of single dose regimens has not been established.

It is accepted that the use of oral neomycin sulfate and erythromycin base before colon surgery results in decreased numbers of intestinal bacteria. Intraluminal levels of these agents are reported to be very high, but systemic availability is still debated. The systemic levels were studied in 8 patients undergoing colon surgery. Each patient received neomycin sulfate and erythromycin base, 1 g each, 19, 18 and 9 hours preoperatively. Twelve samples from serum, one from wound muscle and one from the intestinal wall were obtained from each patient in the 26 hours after the initial dose. Considerable variation was observed among levels. The following means were calculated: peak serum levels were 3.4 and 0.59 micrograms/ml, muscle levels were 1.68 and 0.23 micrograms/g and intestinal wall levels were 6.4 and 12.9 micrograms/g for erythromycin and neomycin respectively. Observed times to peak levels were 19 and 12 hours after the initial dose for erythromycin and neomycin respectively. The detectable systemic concentrations that result when these agents are given orally for bowel preparation before colon surgery may contribute to the drugs' efficacy.

One hundred and ninety-five patients undergoing abdominal surgical procedures completed a multicentre, randomized, open-label study comparing the safety and efficacy of cefmetazole and cefoxitin for the prevention of postoperative wound infection. Cefmetazole was administered iv in a single 2 g dose given within 90 min of the operation. Cefoxitin was administered in a single 2 g, similarly timed, preoperative dose and two additional doses given at 6 h intervals after surgery. For operations that exceeded 2-4 h duration an additional dose of each agent was administered. Patients undergoing colorectal operations received oral neomycin and erythromycin as bowel preparation. Colorectal operations were performed most frequently (49% of patients) followed by cholecystectomies (26%) and gastroduodenal procedures (21%). The operative site infection rate was 6.5% for cefmetazole and 7.7% for cefoxitin (P greater than 0.05). Serious drug related adverse effects were not observed. This study demonstrates that administration of single-dose cefmetazole is as effective as a standard three dose regimen of cefoxitin for prophylaxis with abdominal operations.

Cephalosporin antibiotics are the most frequently used agents for surgical prophylaxis. Within this class are considerable pharmacokinetic variations that could have significant implications. We used a computer simulation of cephalosporin serum levels to describe concentrations achieved and maintained intraoperatively when the agents are given intravenously "on call" to the operating room or with induction of anesthesia. Intraoperative serum concentrations fall below 1 microgram/ml if an operation lasts longer than 2.3, 2.7, 3.8, or 4.0 hours when cephalothin, cephapirin, cefamandole, or cefoxitin, respectively, is given in usual doses upon induction of anesthesia. When the same agents are given intravenously on call to the operating room, intraoperative serum concentrations fall below 1 microgram/ml for operations lasting longer than 1.1, 1.5, 2.6, or 2.8 hours, respectively. If cephalothin, cephapirin, cefamandole, or cefoxitin is used, it should be given at induction of anesthesia to provide maximal intraoperative serum concentrations. The longer half-life of cefazolin, ceforanide, cefonicid, and cefuroxime is a potential advantage because serum concentrations of these agents are well above 1 microgram/ml for as long as eight to 22 hours even after on-call administration.

The introduction of gentamicin almost 20 years ago provided an effective option for the treatment of gram-negative bacillary infections. During the past few years, the availability of aztreonam (a monobactam), imipenem (a carbapenem), and newer cephalosporins within vitro activities comparable with aminoglycosides against many gram-negative bacilli, has stimulated a reassessment of the role of aminoglycosides in treating these infections. When determining the role of new antimicrobials as potential replacements for more established agents, the clinical focus should be on three factors: comparative efficacy, safety, and cost. Consideration of cost is relevant only when efficacy and safety are equivalent. Other factors, such as comparative in vitro antimicrobial activity, pharmacokinetics, and effect on normal flora can also influence the selection of an antimicrobial regimen. A new class of antimicrobials, the monobactams, is the focus of this review. The only member of this class currently in clinical use is aztreonam. A comparison with aminoglycosides is particularly relevant because aztreonam is active against aerobic gram-negative bacilli. This review will discuss the acknowledged concerns with aminoglycoside use and compare the characteristics of aztreonam and currently marketed aminoglycosides.

Over an 18-year period 16 patients with pseudomembraneous enterocolitis were treated by restoration of floral homeostasis by administration of fecal enemas when standard accepted forms of therapy had failed to resolve the disease. Thirteen of the patients responded dramatically, with decreases in diarrhea, temperature, white blood cell counts, and a rapid convalesence. Of the three patients who died, two did not have the pseudomembrane at death, and one had involvement of the small bowel. No ill effects from the fecal enemas were noted. The observation of the pseudomembrane involving only the right half of the colon in a patient with a diverting transverse colostomy indicates that the disease is a contact mucositis.

Parenteral prophylactic cephalosporins used in surgery were compared in 17 published studies. Examination of these studies reveals little justification for preference of one cephalosporin over another. For gastrointestinal, obstetrical-gynecologic, or cardiac operations, newer cephalosporins did not result in substantial decreases in adverse postoperative clinical events (eg, wound infections, intra-abdominal and pelvic infections, and endocarditis) when compared with older cephalosporins. There is no evidence that second- or third-generation cephalosporins result in postoperative infection rates lower than with first-generation cephalosporins.

We determined the intraoperative concentrations of cefmetazole and cefoxitin in serum and muscle from the wound of 30 patients who were undergoing cholecystectomies. The study employed an open-label design in which all patients randomly received cefoxitin sodium (30 mg/kg) or cefmetazole sodium (15 or 30 mg/kg) intravenously with the induction of anesthesia. Total serum and wound-muscle concentrations achieved with cefmetazole 30 mg/kg were significantly greater than those achieved with a similar dose of cefoxitin. Cefmetazole in a 15 mg/kg dose was comparable with cefoxitin 30 mg/kg in achieved concentrations. The elimination half-life for cefoxitin was much shorter than that for cefmetazole (41 min v. 64-68 min, respectively) and this relates to a shorter duration of action for the former. The choice of agent for surgical prophylaxis should incorporate factors relating to drug pharmacokinetic properties as well as microbiological factors.

Intraoperative gastrointestinal endoscopy is an evolving technique that has received limited attention in the literature through case and small series reports. This literature does contain some papers in which the method was employed instead of conventional endoscopy, but, in general, intraoperative endoscopy has been effectively used as an aid to the surgeon to locate pathology not otherwise detectable and to enhance diagnosis and therapy at laparotomy. The procedure will not be required often, but when used, has the potential to be of as much service to the surgeon as a scalpel or retractor. As more surgeons are being trained in gastrointestinal endoscopy, the indications and uses will surely expand.

The routine use of intraoperative cholangiography has vastly improved the results of common duct exploration by reducing the number of negative explorations. The controversy surrounding the use of routine versus selective intraoperative cholangiography has centered on the incidence of unsuspected common-duct stones. A prospective study was designed to examine both preoperative clinical data and intraoperative anatomical information to determine criteria that would identify patients who would not require cholangiography. One hundred consecutive patients undergoing cholecystectomy were included in the study. When considering patients without preoperative clinical data suspicious for common duct stones three anatomical conditions were identified in which common duct stones would not be present: cystic duct less than 3 mm; smallest stone size greater than 6 mm; a single stone. Using intraoperative criteria as a basis for cholangiography, 44 per cent of patients without clinical suspicion of common duct stones would be spared an intraoperative cholangiogram. It is the authors' opinion that the addition of anatomic findings to preoperative clinical data can further reduce or eliminate the risk of unsuspected stones while sparing a large number of patients the risk and expense of routine intraoperative cholangiography and possible negative duct exploration.

The use of polyethylene glycol electrolyte lavage solution (PEG-ELS) as a whole-bowel irrigation for colorectal surgery is retrospectively evaluated in 65 consecutive patients. Sixty-three (97%) of the patients were able to complete the lavage. Fifty-five (85%) of these tolerated the preparation without event. Unpleasant reactions of nausea, vomiting, or fullness resolved in most patients by decreasing the rate of ingestion. Twenty-seven patients had associated medical illnesses that may have been aggravated by volume overload, but none of these experienced any adverse effects from the lavage. Most patients completed the total lavage ingestion of 4 L in 4 hours, which resulted in a 1-day preparation prior to operation. There were no wound or septic complications in any of the lavage patients. This clinical review indicates that PEG-ELS is a safe, rapid, and effective mechanical bowel preparation that should be able to reduce in-hospital preparation time for most patients undergoing colorectal surgery.