JOURNAL OF LAPAROENDOSCOPIC SURGERY Volume 2, Number 6, 1992 Mary Ann Liebert, Inc., Publishers
Comparison of Laparoscopic Cholecystectomy versus Elective Open Cholecystectomy JEFFREY F. SMITH, M.D.,1 DENNIS BOYSEN, M.D.,2 JAMES TSCHIRHART, M.D.,2 TERRI WILLIAMS, M.D.,2 and PETER
VASILENKO, Ph.D.1
ABSTRACT
Laparoscopic cholecystectomy has essentially replaced open cholecystectomy as the procedure of choice for gallbladder disease. This rapid shift to laparoscopic cholecystectomy, however, has resulted more from marketing forces than from prospective clinical trials. To evaluate the safety and efficacy of laparoscopic cholecystectomy, the first 486 laparoscopic cholecystectomies at two institutions were studied. These results were then compared to the results of the last 6 months of elective open cholecystectomy cases prior to the introduction of laparoscopic surgery. The age, sex, height, and weight were similar in both groups. The mean operative time was 78.8 ±1.8 min for laparoscopic cholecystectomy and 62.7 ± 2.6 min for open cholecystectomy (p < 0.01). The mean time for tolerating a regular diet was 1.23 ± 0.04 days in the laparoscopic group versus 2.44 ± 0.07 days in the open group (p < 0.01). Laparoscopic cholecystectomy patients required only oral pain medications by a mean of 1.22 ± 0.03 days postoperatively compared to 2.55 ± 0.07 days postoperatively for those undergoing open cholecystectomy (/; < 0.01). The mean length of hospitalization was 1.58 ± 0.07 days for laparoscopic patients and 3.55 ± 0.11 days for open patients (p < 0.01). Thirty-one patients undergoing laparoscopic cholecystectomy were converted to open cholecystectomy (6.4%). The most common reasons for conversion to open cholecystectomy were acute inflammation, adhesions, and bleeding. For the laparoscopic patients, the morbidity rate was 8.4% and the mortality rate 0.2% (1 death). In the open cholecystectomy group the morbidity rate was 8.0% and there were no deaths. The most troublesome complication in laparoscopic cholecystectomies continues to be bile leaks and bile duct injuries. The authors report two cases of bile duct injuries and one case of bile leak which resulted in the one death in the study. This study reveals some clear advantages of laparoscopic cholecystectomy. However, laparoscopic cholecystectomy remains a major operation with potential for serious morbidity and even death.
'Saginaw Cooperative Hospitals, Inc. and 2Michigan State University-College of Human Medicine, Saginaw, MI. 311
SMITH ET AL.
INTRODUCTION cholecystectomy has open cholecystectomy as the procedure of shift from open to laparoscopic cholecystectomy has not choice for The been the result of prospective clinical trials. Instead, it has resulted from marketing and a public that has come to demand the laparoscopic technique. Yet, as surgeons and scientists, we must critically compare laparoscopic cholecystectomy with the gold standard, open cholecystectomy. The rapid emergence of laparoscopic cholecystectomy has made the performance of prospective trials between the two techniques all but impossible. This study first describes the results of the first 486 laparoscopic cholecystectomies performed by community surgeons in Saginaw, Michigan. The laparoscopic results are then compared to the results of the last 6 months of elective open cholecystectomy cases performed by the same surgeons prior to the introduction of laparoscopic surgery.
essentially replaced disease.9'014 Laparoscopigallbladder c rapid
METHODS The first 486 patients to undergo laparoscopic cholecystectomy from July 1, 1990 through July 31, 1991 included in the study. The chart of each patient was carefully reviewed. Nine general surgeons performed the procedure during the course of the study. The age, sex, height, and weight of each patient was recorded. All preoperative procedures used to confirm biliary tract disease were noted, along with the date of surgery and total operative time. During the operation, the method of dissection, use of intraoperative cholangiogram, and placement of drains were charted. The pain medication requirements of the patients were recorded along with the postoperative diet tolerance for each patient. The patients were carefully followed throughout their hospitalization and monitored as an outpatient for any significant complications. The charts of all patients that had undergone elective open cholecystectomy in the 6 months prior to the first laparoscopic cholecystectomy in Saginaw, MI were carefully reviewed; 124 patients were included in the study. The data obtained was identical to that obtained for the laparoscopic group. Next, the laparoscopic cholecystectomy results and open cholecystectomy results were compared. The comparison of results between the two groups was analyzed using analysis of variance. The 13 months of laparoscopic cholecystectomy cases were also divided into three groups based on the date of the procedure (months 1-4, months 5-8, and months 9-13). Statistical differences between these three groups were analyzed using analysis of variance. The results are expressed as mean ± SEM. were
Patients The mean ages of laparoscopic patients and open cholecystectomy patients were not significantly different, 47.6 ± 0.8 and 47.0 ± 1.4 years respectively. There were 398 females and 88 males in the laparoscopic group and 101 females and 23 males in the open group. The mean weight was 79.0 ± 0.8 kg in the laparoscopic group and 81.6 ± 1.6 kg in the open group. The mean heights were 162.1 ± 0.5 cm and 159.8 ± 1.1 cm in the laparoscopic and open groups respectively.
Operative techniques All patients undergoing laparoscopic cholecystectomy had general anesthesia. All laparoscopic cholecystectomies were performed by two attending surgeons or an attending surgeon and a senior surgery resident. Surgeons used the KTP laser, YAG laser, or electrocautery for dissections. Laser dissection was performed in 199 patients (43.8%). Electrocautery was the primary dissecting method in the remaining 256 patients (56.2%). The cystic duct was always cut with scissors (not cautery or laser). Intraoperative cholangiograms were performed successfully in 59 patients ( 13%) in the laparoscopic group. In 15 patients, cholangiograms were attempted without success; 99 intraoperative cholangiograms (79%) 312
COMPARISON OF LAPAROSCOPIC AND ELECTIVE OPEN were performed in the open cholecystectomy group. Significantly more cholangiograms were performed in the open group compared to the laparoscopic group (p < .01). Drains were placed in 8.1% of patients in the laparoscopic group and in 9.7% of patients in the open cholecystectomy group.
Duration
of procedure
The mean operative time for laparoscopic cholecystectomy was 78.8 ± 1.8 min compared to 62.7 ± 2.6 min for open cholecystectomy. This difference was statistically significant (p < .01). The mean operative time in the first 4 months of the laparoscopic study was 89.4 ± 3.7 min compared to 77.0 ± 1.7 min in the second 4 months of laparoscopic cholecystectomy cases and 74.9 ±5.6 min in the last 5 months. Analysis of variance revealed no significant difference between the operative times of these three groups.
RESULTS Conversion to open
cholecystectomy
Thirty-one of the 486 patients undergoing laparoscopic cholecystectomy were converted to open cholecystectomy for a conversion rate of 6.4%. Ten of these patients required laparotomy for acute inflammation, eight due to adhesions, and seven due to significant bleeding. Four patients were converted to open cholecystectomy because of inability to identify vital anatomy. One obese patient was converted and another was converted because an intraoperative cholangiogram showed common duct stones (Table 1). During the first 4 months of the study, 4.1 % of patients were converted to open cholecystectomy and 3.0% during the second 5 months. However, the conversion rate to open cholecystectomy increased to 11.5% during the final 5 months of the study.
Postoperative
course
The postoperative course of the vast majority of laparoscopic cholecystectomy patients was uneventful. All patients were routinely kept overnight for observation. Nasogastric tubes and Foley catheters were discontinued immediately following surgery. Patients were tolerating a regular diet an average of 1.23 ± 0.04 days postoperatively. Patients were using only oral pain medications by an average of 1.22 ± 0.03 days postoperatively. The mean length of hospitalization in the laparoscopic cholecystectomy patients was 1.58 ± 0.07 days (Table 2). In the open cholecystectomy group, the mean time until a regular diet was tolerated was 2.44 ± 0.07 days. The mean time until only oral pain medications were required was 2.55 ± 0.07 days and the mean hospital stay was 3.55 ± 0.11 days. The differences between the laparoscopic and open cholecystectomy groups in hospital stay, time to regular diet, and time to using only oral pain medications were statistically significant using the student t-test
(p
400 cc blood loss) occurred in nine patients that completed laparoscopic cholecystectomy. Seven additional patients were converted to open cholecystectomy due to bleeding. One patient lost 1100 cc of blood before open cholecystectomy was performed to control the cystic artery hemorrhage. Bleeding originated from the liver bed or cystic artery except in one case where superior epigastric artery bleeding from trocar placement occurred. All patients with excessive bleeding recovered uneventfully. No patient with excessive bleeding required transfusion. The relatively high incidence of bleeding necessitating conversion is not known. None of these patients had any known risk factors for bleeding. Three patients developed postoperative bile leaks. None of these patients had an intraoperative cholangiogram. The first patient was a 45-year-old female who was noted to have dense inflammation of the gallbladder extending down into the cystic duct and common bile duct areas. During initial dissection, a tiny pinpoint perforation was made in the common bile duct near its junction with the cystic duct. The perforation was immediately recognized and a Jackson-Pratt drain placed at the area of the leak. Bile drainage was 500 cc for the first 24 h but decreased daily. By postoperative day 7, the drainage had stopped completely and the drain was removed on postoperative day 8. The patient was discharged without any further problems. The second patient was a thin 35-year-old female who underwent an uneventful laparoscopic cholecystectomy for chronic cholecystitis. She presented 1 week postoperatively complaining of abdominal fullness. At that time, her white blood cell count was normal but her total bilirubin was 5.4 and indirect bilirubin 4.3 mg/dl. Abdominal ultrasound was performed and revealed a small fluid collection near the porta hepatis. Hepatobiliary scan showed a small bile leak near the porta hepa but prompt excretion of radionucleotide into the common bile duct and small bowel. The following day, under CT guidance, two # 10 French nephrostomy catheters were placed for external drainage. The patient had 400-600 cc of bile drainage through the catheters for the next 7 days when she suddenly developed increased abdominal pain, fever, and a white blood cell count of 22,000. ERCP revealed a bile leak at the cystic duct stump. A laparotomy was performed and it was discovered that the clips had fallen off the cystic duct remnant. In addition, an area on the common bile duct near the cystic duct stump was very inflamed. No bile was leaking from the common bile duct. The cystic duct remnant was ligated. A Jackson-Pratt drain was placed near the cystic duct stump and bilious drainage persisted from the drain. It was felt then that there was a small leak near the junction of the cystic duct and common bile duct. Fortunately, using the well established J-P drain tract, a radiologist placed a biliary catheter into the apparent hole in the common bile duct. The catheter was replaced with smaller and smaller catheters until it was finally discontinued 46 days after the initial cystic duct leak. A repeat hepatobiliary scan showed no evidence of obstruction following catheter removal. The patient made a full recovery. Neither patient with a common bile duct injury had an attempted cholangiogram. The third patient with a bile leak was the lone death in this study. The patient was a 69-year-old female admitted for an elective laparoscopic cholecystectomy for chronic cholecystitis. She had symptoms consistent with chronic cholecystitis for 2 months and a recent gallbladder ultrasound revealed two large stones in the gallbladder. Her past medical history was significant for diabetes, coronary artery disease, and hypertension. She also had a questionable history of hepatitis in the past. On admission, the patient was afebrile with only 314
COMPARISON OF LAPAROSCOPIC AND ELECTIVE OPEN mild tenderness in the right upper quadrant and left upper quadrant. Her laboratory tests included a hemoglobin of 13.4, white blood cell count of 7.7, with anormal bilirubin and normal liver enzymes. Her PT was mildly elevated at 14.2 sec (control 12.2). During the operation her liver appeared enlarged and cirrhotic. A tru-cut liver biopsy was done and an uneventful laparoscopic cholecystectomy performed. The cystic duct was double clipped proximally and single clipped distally. Approximately 10 h postoperatively the patient began complaining of increased abdominal pain. Her temperature was 101.0 F and she was mildly confused. Her abdomen was moderately tender throughout with some right upper quadrant guarding. The following morning her fever persisted (101.6F) and her abdomen became distended with no bowel sounds. Her white blood cell count was 16.OK with 58 segs and 34 bands. At this time, a JP drain was inserted under local anesthesia using a lateral trocar site. Some brown bilious fluid was noted. Her clinical status did not improve over the next 12 h so a laparotomy was performed. A large amount of purulent bilious fluid was found in the peritoneal cavity. The clips had fallen off the cystic duct stump. The abdominal cavity was copiously irrigated and the cystic duct stump suture ligated. The patient remained intubated and a Swan-Ganz catheter was placed. Cardiac profile was consistent with sepsis. Despite broad spectrum antibiotics and aggressive fluid and pressure support, the patient continued a downhill course. Blood cultures revealed Staph aureus and Candida. Multi-organ system failure ensued and the patient died on postoperative day 18. A prolonged ileus occurred in five patients. This was defined as an inability to tolerate a regular diet until at least postoperative day 4. Four patients developed postoperative urinary tract infections. One patient each developed parotiditis, a facial rash of unknown origin, superficial thrombophlebitis, and a wound infection. One of the first patients in the study developed a large amount of subcutaneous emphysema on the trunk which resolved spontaneously without sequelae. The patient was discharged home on the third postoperative day. It was felt that during the operation one or more trocars may have been in the subcutaneous space with the dissection of C02 into this space. One patient developed presumed cholangitis from a retained common bile duct stone. The patient was a 45-year-old male admitted for elective laparoscopic cholecystectomy for chronic cholecystitis. His preoperative bilirubin was 1.8. Preoperatively, the treatment options were discussed with the patient. The patient wanted postoperative endoscopie sphincterotomy and ERCP if choledocholithiasis was found. The patient underwent laparoscopic cholecystectomy and cholangiogram revealed common bile duct stones. He had been given Cefotetan preoperatively and this was continued. On postoperative day 2 the patient developed fever to 101.5 and shaking chills. On postoperative day 3 an ERCP and sphincterotomy were performed. The patient recovered with no further fever and was discharged on postoperative day 5 on oral antibiotics for 7 additional days. Open Cholecystectomy. Ten complications occurred in 124 patients for a morbidity rate of 8.0%. No deaths occurred. The most common complication was atelectasis which occurred in five patients (4%). Urinary tract infections occurred in two patients. One patient developed malignant hypertension postoperatively. She required admission to the CCU for 2 days but recovered uneventfully. One patient developed seizures on postoperative day 1. She was treated with phenytoin and no further seizures occurred. A complete neurological work-up was negative. Another patient developed acute urinary retention postoperatively and subsequent urologie work-up revealed benign prostatic hypertrophy. No bile duct injuries or bile leaks occurred in the elective open cholecystectomy group. =
DISCUSSION
Laparoscopic cholecystectomy has become the procedure of choice for gallbladder disease.1214 Critical of this procedure is needed particularly with comparison to open cholecystectomy. In this study, the authors compared the results with laparoscopic cholecystectomy (first 13 months) with the results with elective open cholecystectomy in the 6 months preceding the first laparoscopic cholecystectomy assessment
in the institutions mentioned. In some areas, the laparoscopic cholecystectomy patients did much better than their open cholecystectomy counterparts. The length of hospitalization was much shorter in the laparoscopic patients (1.58 days vs. 3.55
315
SMITH ET AL.
days,/? < .001). The patients postoperative diet tolerance was significantly better (p < .001) and the need for parenteral pain medications significantly less (p < .001) in the laparoscopic group of patients. Patients' decreased discomfort and shortened hospital stay help defray rising medical costs and allow the patients to return to work more quickly.I5 However, a critical assessment of the morbidity and mortality of a new procedure is most important. The overall morbidity rate for these laparoscopic cholecystectomy patients was 8.5% with a major complication rate of 3.0%. In comparison, the overall morbidity rate for elective open cholecystectomy patients was 8.0% with a major complication rate of 1.6%. The laparoscopic morbidity rate '"2"4 Other compares favorably with the 6-21% morbidity rate in the literature for open cholecystectomy. groups have reported morbidity rates for laparoscopic cholecystectomy in the 5-8% range.5"6 Although rare, bile duct leaks and injuries continue to be the most significant and troublesome complication of laparoscopic cholecystectomy. The authors report one bile leak (0.2%) and two bile duct injuries (0.4%) in the first 13 months experience with laparoscopic cholecystectomy. Other series have shown bile duct injury rates of 0.5-1.0% with laparoscopic cholecystectomy and bile leak rates of 0.2-2.0%.5"6 There were no bile duct injuries or bile leaks in the elective open cholecystectomy patients. Historically, the rate of bile duct injury for open cholecystectomy is between 0.1 % and 0.2%.7'8 Morgenstern et al, recently reported a bile duct injury rate of 0.16% in a series of 1200 consecutive open cholecystectomies. ' ' It appears that laparoscopic cholecystectomy continues to have a slightly increased risk of biliary leak and injury compared to open cholecystectomy. One death accounted for the 0.2% mortality rate in the laparoscopic cholecystectomy patients. The overall '' mortality for open cholecystectomy has been reported to be 1.5-2.0%. This 0.2% mortality rate for laparoscopic cholecystectomy appears acceptable. However, this one death was preventable. In this patient, the clips had fallen off the cystic duct stump which led to bile leak, bile peritonitis, sepsis, multi-system organ failure, and death. Diligent effort must be made in each case to prevent bile leak and bile duct injury. The cystic duct should be carefully cut with scissors at least .5-1.0 cm from the proximal clips on the cystic duct. In addition, when finishing the procedure the cystic duct stump and common bile duct areas should be reexamined to check for any bile leaks or injury. If postoperative bile leak does occur, early recognition and immediate surgical
intervention is essential to prevent disaster. The low rate of laparoscopic cholangiography can be explained by two factors. First, these patients represent the first 13 months of laparoscopic cholecystectomies being performed in the authors' community. During this time, the technique and instruments needed for cholangiography were evolving. Secondly, patients that had significant signs, symptoms, or laboratory data indicating possible common bile duct stones had either preoperative or postoperative ERCP or underwent open cholecystectomy. The management of patients suspected of having common bile duct stones remains controversial. Some surgeons advocate preoperative ERCP with removal of stones before laparoscopic cholecystectomy. Others argue that these patients are best managed with laparoscopic cholecystectomy with a mandatory intraoperative cholangiogram. If choledocholithiasis is found, ERCP and sphincterotomy should be performed postoperatively or the patient should be converted to open cholecystectomy with a common bile duct exploration. These various options need to be discussed with patients preoperatively. Then, with the benefits and risks of each option known, the patient can decide which therapy they would prefer. Overall, 6.4% of laparoscopic cholecystectomies were converted to open cholecystectomies. Interestingly, the conversion rate was between 3.0 and 4.1% during the first 8 months of the study, but dramatically increased to 11.6% in the final 5 months of the study. The authors attribute this increase primarily to two factors. First, more difficult cases (i.e. acute cholecystitis, previous upper abdominal surgeries) were being performed in the last 5 months of the study. Obviously, a higher percentage of these cases required conversion to open cholecystectomy. Secondly, as the surgeon gains experience with the laparoscopic technique, he does not hesitate to convert to open cholecystectomy. With experience the surgeon knows the limitations of laparoscopic cholecystectomy and the potential risks of continuing laparoscopically with unclear anatomy. If the crucial anatomy is not seen laparoscopically, conversion to open cholecystectomy is mandatory to prevent bile duct injury. In summary, this study comparing laparoscopic and open cholecystectomy reveals some clear advantages of laparoscopic cholecystectomy. However, laparoscopic cholecystectomy remains a major surgery with potential for serious morbidity and even death. 316
COMPARISON OF LAPAROSCOPIC AND ELECTIVE OPEN
ACKNOWLEDGMENTS The authors thank Sharon for publication.
Sawyers and Leslie Francke for their hours of work in preparing this manuscript REFERENCES
1. Scher KS, Scott-Conner CE:
2.
3.
Complications of biliary surgery. Am Surg 1987;53:16-21. Glenn F, McSherry CK, Dineen P: Morbidity of surgical treatment for nonmalignant biliary tract diseases. Surg Gynecol Obstet 1968;126:15-26. Kent III R: Subcutaneous emphysema and hypercarbia following laparoscopic cholecystectomy. Arch Surg 1991;211(9):1154-1156.
4. 5. 6.
Dowdy Jr GS, Waldron GW: Importance of coexistent factors in biliary tract surgery. Arch Surg 1964;88:314-323. Peters JH, Ellison E, Innés J, et al.: Safety and efficacy of laparoscopic cholecystectomy. Ann Surg 1991;213:3-12. The Southern Surgeons Club: A prospective analysis of 1518 laparoscopic cholecystectomies. N Engl J Med 1991;324:1073-1078.
7. Viikari SJ:
Operative injuries to bile ducts.
8. Raute M,
Schaupp
W:
latrogenic damage
Acta Chir Scand 1960;119:83-92.
of the bile ducts caused
by cholecystectomy. Langenbecks
Arch Chir
1988;373:345-354. 9. Reddick EJ, Olsen DO: Laparoscopic laser Endose 1989;3:131-133. 10. DuBoisF, IchardP, Berthelot G, LevardH:
cholecystectomy:
A
comparison with mini-lap cholecystectomy. Surg
Coelioscopic cholecystectomy: Preliminary report of 36 cases. Ann Surg
1990;211:60-62. 11.
Morgenstern L, Wong
L: Twelve hundred open
cholecystectomies
before the
laparoscopic
era.
Arch
Surg
1992;127:400-403. 12. Stoker ME, Vose J, O'Mara P, Maini BS: operations. Arch Surg 1992;127:589-595.
15.
A clinical and financial
Ballinger JF, Anderson WJ: Appraisal of laparoscopic cholecystectomy.
analysis of 280
Surg 1991;213:655-664. Schirmer BD, Edge SB, Dix J, Hyser MJ, Hanks JB, Jones RS: Laparoscopic cholecystectomy: Treatment of choice for symptomatic cholelithiasis. Ann Surg 1991;213:665-677. Smith J, Boysen D, Tschirhart J, Williams T: Risks and benefits of laparoscopic cholecystectomy in the community hospital setting. J Laparoendosc Surg 1991;1:325-332.
13. Graves HA, 14.
Laparoscopic Cholecystectomy:
Ann
Address
reprint request to: Jeffrey Smith, M.D. Department of Surgery Saginaw Cooperative Hospitals, Inc. 1000 Houghton Avenue Saginaw, Ml 48602
317
Comparison of Laparoscopic Cholecystectomy versus Elective Open Cholecystectomy JEFFREY F. SMITH, M.D.,1 DENNIS BOYSEN, M.D.,2 JAMES TSCHIRHART, M.D.,2 TERRI WILLIAMS, M.D.,2 and PETER
VASILENKO, Ph.D.1
ABSTRACT
Laparoscopic cholecystectomy has essentially replaced open cholecystectomy as the procedure of choice for gallbladder disease. This rapid shift to laparoscopic cholecystectomy, however, has resulted more from marketing forces than from prospective clinical trials. To evaluate the safety and efficacy of laparoscopic cholecystectomy, the first 486 laparoscopic cholecystectomies at two institutions were studied. These results were then compared to the results of the last 6 months of elective open cholecystectomy cases prior to the introduction of laparoscopic surgery. The age, sex, height, and weight were similar in both groups. The mean operative time was 78.8 ±1.8 min for laparoscopic cholecystectomy and 62.7 ± 2.6 min for open cholecystectomy (p < 0.01). The mean time for tolerating a regular diet was 1.23 ± 0.04 days in the laparoscopic group versus 2.44 ± 0.07 days in the open group (p < 0.01). Laparoscopic cholecystectomy patients required only oral pain medications by a mean of 1.22 ± 0.03 days postoperatively compared to 2.55 ± 0.07 days postoperatively for those undergoing open cholecystectomy (/; < 0.01). The mean length of hospitalization was 1.58 ± 0.07 days for laparoscopic patients and 3.55 ± 0.11 days for open patients (p < 0.01). Thirty-one patients undergoing laparoscopic cholecystectomy were converted to open cholecystectomy (6.4%). The most common reasons for conversion to open cholecystectomy were acute inflammation, adhesions, and bleeding. For the laparoscopic patients, the morbidity rate was 8.4% and the mortality rate 0.2% (1 death). In the open cholecystectomy group the morbidity rate was 8.0% and there were no deaths. The most troublesome complication in laparoscopic cholecystectomies continues to be bile leaks and bile duct injuries. The authors report two cases of bile duct injuries and one case of bile leak which resulted in the one death in the study. This study reveals some clear advantages of laparoscopic cholecystectomy. However, laparoscopic cholecystectomy remains a major operation with potential for serious morbidity and even death.
'Saginaw Cooperative Hospitals, Inc. and 2Michigan State University-College of Human Medicine, Saginaw, MI. 311
SMITH ET AL.
INTRODUCTION cholecystectomy has open cholecystectomy as the procedure of shift from open to laparoscopic cholecystectomy has not choice for The been the result of prospective clinical trials. Instead, it has resulted from marketing and a public that has come to demand the laparoscopic technique. Yet, as surgeons and scientists, we must critically compare laparoscopic cholecystectomy with the gold standard, open cholecystectomy. The rapid emergence of laparoscopic cholecystectomy has made the performance of prospective trials between the two techniques all but impossible. This study first describes the results of the first 486 laparoscopic cholecystectomies performed by community surgeons in Saginaw, Michigan. The laparoscopic results are then compared to the results of the last 6 months of elective open cholecystectomy cases performed by the same surgeons prior to the introduction of laparoscopic surgery.
essentially replaced disease.9'014 Laparoscopigallbladder c rapid
METHODS The first 486 patients to undergo laparoscopic cholecystectomy from July 1, 1990 through July 31, 1991 included in the study. The chart of each patient was carefully reviewed. Nine general surgeons performed the procedure during the course of the study. The age, sex, height, and weight of each patient was recorded. All preoperative procedures used to confirm biliary tract disease were noted, along with the date of surgery and total operative time. During the operation, the method of dissection, use of intraoperative cholangiogram, and placement of drains were charted. The pain medication requirements of the patients were recorded along with the postoperative diet tolerance for each patient. The patients were carefully followed throughout their hospitalization and monitored as an outpatient for any significant complications. The charts of all patients that had undergone elective open cholecystectomy in the 6 months prior to the first laparoscopic cholecystectomy in Saginaw, MI were carefully reviewed; 124 patients were included in the study. The data obtained was identical to that obtained for the laparoscopic group. Next, the laparoscopic cholecystectomy results and open cholecystectomy results were compared. The comparison of results between the two groups was analyzed using analysis of variance. The 13 months of laparoscopic cholecystectomy cases were also divided into three groups based on the date of the procedure (months 1-4, months 5-8, and months 9-13). Statistical differences between these three groups were analyzed using analysis of variance. The results are expressed as mean ± SEM. were
Patients The mean ages of laparoscopic patients and open cholecystectomy patients were not significantly different, 47.6 ± 0.8 and 47.0 ± 1.4 years respectively. There were 398 females and 88 males in the laparoscopic group and 101 females and 23 males in the open group. The mean weight was 79.0 ± 0.8 kg in the laparoscopic group and 81.6 ± 1.6 kg in the open group. The mean heights were 162.1 ± 0.5 cm and 159.8 ± 1.1 cm in the laparoscopic and open groups respectively.
Operative techniques All patients undergoing laparoscopic cholecystectomy had general anesthesia. All laparoscopic cholecystectomies were performed by two attending surgeons or an attending surgeon and a senior surgery resident. Surgeons used the KTP laser, YAG laser, or electrocautery for dissections. Laser dissection was performed in 199 patients (43.8%). Electrocautery was the primary dissecting method in the remaining 256 patients (56.2%). The cystic duct was always cut with scissors (not cautery or laser). Intraoperative cholangiograms were performed successfully in 59 patients ( 13%) in the laparoscopic group. In 15 patients, cholangiograms were attempted without success; 99 intraoperative cholangiograms (79%) 312
COMPARISON OF LAPAROSCOPIC AND ELECTIVE OPEN were performed in the open cholecystectomy group. Significantly more cholangiograms were performed in the open group compared to the laparoscopic group (p < .01). Drains were placed in 8.1% of patients in the laparoscopic group and in 9.7% of patients in the open cholecystectomy group.
Duration
of procedure
The mean operative time for laparoscopic cholecystectomy was 78.8 ± 1.8 min compared to 62.7 ± 2.6 min for open cholecystectomy. This difference was statistically significant (p < .01). The mean operative time in the first 4 months of the laparoscopic study was 89.4 ± 3.7 min compared to 77.0 ± 1.7 min in the second 4 months of laparoscopic cholecystectomy cases and 74.9 ±5.6 min in the last 5 months. Analysis of variance revealed no significant difference between the operative times of these three groups.
RESULTS Conversion to open
cholecystectomy
Thirty-one of the 486 patients undergoing laparoscopic cholecystectomy were converted to open cholecystectomy for a conversion rate of 6.4%. Ten of these patients required laparotomy for acute inflammation, eight due to adhesions, and seven due to significant bleeding. Four patients were converted to open cholecystectomy because of inability to identify vital anatomy. One obese patient was converted and another was converted because an intraoperative cholangiogram showed common duct stones (Table 1). During the first 4 months of the study, 4.1 % of patients were converted to open cholecystectomy and 3.0% during the second 5 months. However, the conversion rate to open cholecystectomy increased to 11.5% during the final 5 months of the study.
Postoperative
course
The postoperative course of the vast majority of laparoscopic cholecystectomy patients was uneventful. All patients were routinely kept overnight for observation. Nasogastric tubes and Foley catheters were discontinued immediately following surgery. Patients were tolerating a regular diet an average of 1.23 ± 0.04 days postoperatively. Patients were using only oral pain medications by an average of 1.22 ± 0.03 days postoperatively. The mean length of hospitalization in the laparoscopic cholecystectomy patients was 1.58 ± 0.07 days (Table 2). In the open cholecystectomy group, the mean time until a regular diet was tolerated was 2.44 ± 0.07 days. The mean time until only oral pain medications were required was 2.55 ± 0.07 days and the mean hospital stay was 3.55 ± 0.11 days. The differences between the laparoscopic and open cholecystectomy groups in hospital stay, time to regular diet, and time to using only oral pain medications were statistically significant using the student t-test
(p
400 cc blood loss) occurred in nine patients that completed laparoscopic cholecystectomy. Seven additional patients were converted to open cholecystectomy due to bleeding. One patient lost 1100 cc of blood before open cholecystectomy was performed to control the cystic artery hemorrhage. Bleeding originated from the liver bed or cystic artery except in one case where superior epigastric artery bleeding from trocar placement occurred. All patients with excessive bleeding recovered uneventfully. No patient with excessive bleeding required transfusion. The relatively high incidence of bleeding necessitating conversion is not known. None of these patients had any known risk factors for bleeding. Three patients developed postoperative bile leaks. None of these patients had an intraoperative cholangiogram. The first patient was a 45-year-old female who was noted to have dense inflammation of the gallbladder extending down into the cystic duct and common bile duct areas. During initial dissection, a tiny pinpoint perforation was made in the common bile duct near its junction with the cystic duct. The perforation was immediately recognized and a Jackson-Pratt drain placed at the area of the leak. Bile drainage was 500 cc for the first 24 h but decreased daily. By postoperative day 7, the drainage had stopped completely and the drain was removed on postoperative day 8. The patient was discharged without any further problems. The second patient was a thin 35-year-old female who underwent an uneventful laparoscopic cholecystectomy for chronic cholecystitis. She presented 1 week postoperatively complaining of abdominal fullness. At that time, her white blood cell count was normal but her total bilirubin was 5.4 and indirect bilirubin 4.3 mg/dl. Abdominal ultrasound was performed and revealed a small fluid collection near the porta hepatis. Hepatobiliary scan showed a small bile leak near the porta hepa but prompt excretion of radionucleotide into the common bile duct and small bowel. The following day, under CT guidance, two # 10 French nephrostomy catheters were placed for external drainage. The patient had 400-600 cc of bile drainage through the catheters for the next 7 days when she suddenly developed increased abdominal pain, fever, and a white blood cell count of 22,000. ERCP revealed a bile leak at the cystic duct stump. A laparotomy was performed and it was discovered that the clips had fallen off the cystic duct remnant. In addition, an area on the common bile duct near the cystic duct stump was very inflamed. No bile was leaking from the common bile duct. The cystic duct remnant was ligated. A Jackson-Pratt drain was placed near the cystic duct stump and bilious drainage persisted from the drain. It was felt then that there was a small leak near the junction of the cystic duct and common bile duct. Fortunately, using the well established J-P drain tract, a radiologist placed a biliary catheter into the apparent hole in the common bile duct. The catheter was replaced with smaller and smaller catheters until it was finally discontinued 46 days after the initial cystic duct leak. A repeat hepatobiliary scan showed no evidence of obstruction following catheter removal. The patient made a full recovery. Neither patient with a common bile duct injury had an attempted cholangiogram. The third patient with a bile leak was the lone death in this study. The patient was a 69-year-old female admitted for an elective laparoscopic cholecystectomy for chronic cholecystitis. She had symptoms consistent with chronic cholecystitis for 2 months and a recent gallbladder ultrasound revealed two large stones in the gallbladder. Her past medical history was significant for diabetes, coronary artery disease, and hypertension. She also had a questionable history of hepatitis in the past. On admission, the patient was afebrile with only 314
COMPARISON OF LAPAROSCOPIC AND ELECTIVE OPEN mild tenderness in the right upper quadrant and left upper quadrant. Her laboratory tests included a hemoglobin of 13.4, white blood cell count of 7.7, with anormal bilirubin and normal liver enzymes. Her PT was mildly elevated at 14.2 sec (control 12.2). During the operation her liver appeared enlarged and cirrhotic. A tru-cut liver biopsy was done and an uneventful laparoscopic cholecystectomy performed. The cystic duct was double clipped proximally and single clipped distally. Approximately 10 h postoperatively the patient began complaining of increased abdominal pain. Her temperature was 101.0 F and she was mildly confused. Her abdomen was moderately tender throughout with some right upper quadrant guarding. The following morning her fever persisted (101.6F) and her abdomen became distended with no bowel sounds. Her white blood cell count was 16.OK with 58 segs and 34 bands. At this time, a JP drain was inserted under local anesthesia using a lateral trocar site. Some brown bilious fluid was noted. Her clinical status did not improve over the next 12 h so a laparotomy was performed. A large amount of purulent bilious fluid was found in the peritoneal cavity. The clips had fallen off the cystic duct stump. The abdominal cavity was copiously irrigated and the cystic duct stump suture ligated. The patient remained intubated and a Swan-Ganz catheter was placed. Cardiac profile was consistent with sepsis. Despite broad spectrum antibiotics and aggressive fluid and pressure support, the patient continued a downhill course. Blood cultures revealed Staph aureus and Candida. Multi-organ system failure ensued and the patient died on postoperative day 18. A prolonged ileus occurred in five patients. This was defined as an inability to tolerate a regular diet until at least postoperative day 4. Four patients developed postoperative urinary tract infections. One patient each developed parotiditis, a facial rash of unknown origin, superficial thrombophlebitis, and a wound infection. One of the first patients in the study developed a large amount of subcutaneous emphysema on the trunk which resolved spontaneously without sequelae. The patient was discharged home on the third postoperative day. It was felt that during the operation one or more trocars may have been in the subcutaneous space with the dissection of C02 into this space. One patient developed presumed cholangitis from a retained common bile duct stone. The patient was a 45-year-old male admitted for elective laparoscopic cholecystectomy for chronic cholecystitis. His preoperative bilirubin was 1.8. Preoperatively, the treatment options were discussed with the patient. The patient wanted postoperative endoscopie sphincterotomy and ERCP if choledocholithiasis was found. The patient underwent laparoscopic cholecystectomy and cholangiogram revealed common bile duct stones. He had been given Cefotetan preoperatively and this was continued. On postoperative day 2 the patient developed fever to 101.5 and shaking chills. On postoperative day 3 an ERCP and sphincterotomy were performed. The patient recovered with no further fever and was discharged on postoperative day 5 on oral antibiotics for 7 additional days. Open Cholecystectomy. Ten complications occurred in 124 patients for a morbidity rate of 8.0%. No deaths occurred. The most common complication was atelectasis which occurred in five patients (4%). Urinary tract infections occurred in two patients. One patient developed malignant hypertension postoperatively. She required admission to the CCU for 2 days but recovered uneventfully. One patient developed seizures on postoperative day 1. She was treated with phenytoin and no further seizures occurred. A complete neurological work-up was negative. Another patient developed acute urinary retention postoperatively and subsequent urologie work-up revealed benign prostatic hypertrophy. No bile duct injuries or bile leaks occurred in the elective open cholecystectomy group. =
DISCUSSION
Laparoscopic cholecystectomy has become the procedure of choice for gallbladder disease.1214 Critical of this procedure is needed particularly with comparison to open cholecystectomy. In this study, the authors compared the results with laparoscopic cholecystectomy (first 13 months) with the results with elective open cholecystectomy in the 6 months preceding the first laparoscopic cholecystectomy assessment
in the institutions mentioned. In some areas, the laparoscopic cholecystectomy patients did much better than their open cholecystectomy counterparts. The length of hospitalization was much shorter in the laparoscopic patients (1.58 days vs. 3.55
315
SMITH ET AL.
days,/? < .001). The patients postoperative diet tolerance was significantly better (p < .001) and the need for parenteral pain medications significantly less (p < .001) in the laparoscopic group of patients. Patients' decreased discomfort and shortened hospital stay help defray rising medical costs and allow the patients to return to work more quickly.I5 However, a critical assessment of the morbidity and mortality of a new procedure is most important. The overall morbidity rate for these laparoscopic cholecystectomy patients was 8.5% with a major complication rate of 3.0%. In comparison, the overall morbidity rate for elective open cholecystectomy patients was 8.0% with a major complication rate of 1.6%. The laparoscopic morbidity rate '"2"4 Other compares favorably with the 6-21% morbidity rate in the literature for open cholecystectomy. groups have reported morbidity rates for laparoscopic cholecystectomy in the 5-8% range.5"6 Although rare, bile duct leaks and injuries continue to be the most significant and troublesome complication of laparoscopic cholecystectomy. The authors report one bile leak (0.2%) and two bile duct injuries (0.4%) in the first 13 months experience with laparoscopic cholecystectomy. Other series have shown bile duct injury rates of 0.5-1.0% with laparoscopic cholecystectomy and bile leak rates of 0.2-2.0%.5"6 There were no bile duct injuries or bile leaks in the elective open cholecystectomy patients. Historically, the rate of bile duct injury for open cholecystectomy is between 0.1 % and 0.2%.7'8 Morgenstern et al, recently reported a bile duct injury rate of 0.16% in a series of 1200 consecutive open cholecystectomies. ' ' It appears that laparoscopic cholecystectomy continues to have a slightly increased risk of biliary leak and injury compared to open cholecystectomy. One death accounted for the 0.2% mortality rate in the laparoscopic cholecystectomy patients. The overall '' mortality for open cholecystectomy has been reported to be 1.5-2.0%. This 0.2% mortality rate for laparoscopic cholecystectomy appears acceptable. However, this one death was preventable. In this patient, the clips had fallen off the cystic duct stump which led to bile leak, bile peritonitis, sepsis, multi-system organ failure, and death. Diligent effort must be made in each case to prevent bile leak and bile duct injury. The cystic duct should be carefully cut with scissors at least .5-1.0 cm from the proximal clips on the cystic duct. In addition, when finishing the procedure the cystic duct stump and common bile duct areas should be reexamined to check for any bile leaks or injury. If postoperative bile leak does occur, early recognition and immediate surgical
intervention is essential to prevent disaster. The low rate of laparoscopic cholangiography can be explained by two factors. First, these patients represent the first 13 months of laparoscopic cholecystectomies being performed in the authors' community. During this time, the technique and instruments needed for cholangiography were evolving. Secondly, patients that had significant signs, symptoms, or laboratory data indicating possible common bile duct stones had either preoperative or postoperative ERCP or underwent open cholecystectomy. The management of patients suspected of having common bile duct stones remains controversial. Some surgeons advocate preoperative ERCP with removal of stones before laparoscopic cholecystectomy. Others argue that these patients are best managed with laparoscopic cholecystectomy with a mandatory intraoperative cholangiogram. If choledocholithiasis is found, ERCP and sphincterotomy should be performed postoperatively or the patient should be converted to open cholecystectomy with a common bile duct exploration. These various options need to be discussed with patients preoperatively. Then, with the benefits and risks of each option known, the patient can decide which therapy they would prefer. Overall, 6.4% of laparoscopic cholecystectomies were converted to open cholecystectomies. Interestingly, the conversion rate was between 3.0 and 4.1% during the first 8 months of the study, but dramatically increased to 11.6% in the final 5 months of the study. The authors attribute this increase primarily to two factors. First, more difficult cases (i.e. acute cholecystitis, previous upper abdominal surgeries) were being performed in the last 5 months of the study. Obviously, a higher percentage of these cases required conversion to open cholecystectomy. Secondly, as the surgeon gains experience with the laparoscopic technique, he does not hesitate to convert to open cholecystectomy. With experience the surgeon knows the limitations of laparoscopic cholecystectomy and the potential risks of continuing laparoscopically with unclear anatomy. If the crucial anatomy is not seen laparoscopically, conversion to open cholecystectomy is mandatory to prevent bile duct injury. In summary, this study comparing laparoscopic and open cholecystectomy reveals some clear advantages of laparoscopic cholecystectomy. However, laparoscopic cholecystectomy remains a major surgery with potential for serious morbidity and even death. 316
COMPARISON OF LAPAROSCOPIC AND ELECTIVE OPEN
ACKNOWLEDGMENTS The authors thank Sharon for publication.
Sawyers and Leslie Francke for their hours of work in preparing this manuscript REFERENCES
1. Scher KS, Scott-Conner CE:
2.
3.
Complications of biliary surgery. Am Surg 1987;53:16-21. Glenn F, McSherry CK, Dineen P: Morbidity of surgical treatment for nonmalignant biliary tract diseases. Surg Gynecol Obstet 1968;126:15-26. Kent III R: Subcutaneous emphysema and hypercarbia following laparoscopic cholecystectomy. Arch Surg 1991;211(9):1154-1156.
4. 5. 6.
Dowdy Jr GS, Waldron GW: Importance of coexistent factors in biliary tract surgery. Arch Surg 1964;88:314-323. Peters JH, Ellison E, Innés J, et al.: Safety and efficacy of laparoscopic cholecystectomy. Ann Surg 1991;213:3-12. The Southern Surgeons Club: A prospective analysis of 1518 laparoscopic cholecystectomies. N Engl J Med 1991;324:1073-1078.
7. Viikari SJ:
Operative injuries to bile ducts.
8. Raute M,
Schaupp
W:
latrogenic damage
Acta Chir Scand 1960;119:83-92.
of the bile ducts caused
by cholecystectomy. Langenbecks
Arch Chir
1988;373:345-354. 9. Reddick EJ, Olsen DO: Laparoscopic laser Endose 1989;3:131-133. 10. DuBoisF, IchardP, Berthelot G, LevardH:
cholecystectomy:
A
comparison with mini-lap cholecystectomy. Surg
Coelioscopic cholecystectomy: Preliminary report of 36 cases. Ann Surg
1990;211:60-62. 11.
Morgenstern L, Wong
L: Twelve hundred open
cholecystectomies
before the
laparoscopic
era.
Arch
Surg
1992;127:400-403. 12. Stoker ME, Vose J, O'Mara P, Maini BS: operations. Arch Surg 1992;127:589-595.
15.
A clinical and financial
Ballinger JF, Anderson WJ: Appraisal of laparoscopic cholecystectomy.
analysis of 280
Surg 1991;213:655-664. Schirmer BD, Edge SB, Dix J, Hyser MJ, Hanks JB, Jones RS: Laparoscopic cholecystectomy: Treatment of choice for symptomatic cholelithiasis. Ann Surg 1991;213:665-677. Smith J, Boysen D, Tschirhart J, Williams T: Risks and benefits of laparoscopic cholecystectomy in the community hospital setting. J Laparoendosc Surg 1991;1:325-332.
13. Graves HA, 14.
Laparoscopic Cholecystectomy:
Ann
Address
reprint request to: Jeffrey Smith, M.D. Department of Surgery Saginaw Cooperative Hospitals, Inc. 1000 Houghton Avenue Saginaw, Ml 48602
317
