- Email: [email protected]
Laparoscopic versus traditional appendectomy for suspected appendicitis
Laparoscopic Versus Traditional Appendectomy for Suspected Appendicitis Bruce D. Schirmer, MD, Robert E. Schmieg, Jr., MD, Janet Dix, PA-C, Stephen B. Edge, Mo, John B. Hanks, MD, Charlottesville, Virginia
We compared the results of concurrently performed laparoscopic versus open appendectomy as treatments for suspected acute appendicitis. The 68 laparoseopic procedures resulted in 62 appendectomies, 47 by the laparoscopie (LA) technique and I5 by the open (LO) technique. Another 54 patients underwent open appendectomy (OA). Significantly more females underwent laparoscopy (LA and LO: 52% versus OA: 33%, p = 0 . 0 4 7 ) . Operative duration was shortest for OA (81 • 3 minutes), which was shorter than for LO (108 4- 7 minutes), but not different than LA (86 4- 6 minues). The postoperative length of stay was not different for LA (3.5 4- 0.5 days) compared with OA (5.9 4- 1.6 days) or LO (4.8 4- 1.3 days). One death occurred in the OA group. Wound complication rates were not significantly different for LA (4.3%) compared with OA (9.4%) and LO (13.3%). Overall complication rates were lower for LA (10.6%) and OA (18.9%) compared with LO (46.7%, p < 0 . 0 1 ) . Median hospital cost for LO ($10,425) was higher (p < 0 . 0 2 ) than for either LA ($5,899) or OA ( $ 5 , 2 2 0 ) . When appendicitis was not present, definitive confirmation of pathology was achieved in 9 of 18 patients undergoing LA versus 4 of 14 patients having OA (p = not significant). We conclude that when laparoseopy and laparoscopie appendectomy can be performed, the procedure is safe and produces results comparable with those of open appendectomy without significant overall cost differences.
L
aparoscopic appendectomy was first described by Semm [1] in the early 1980s. Initial acceptance was very limited due to a low prevalence of laparoscopic skills among general surgeons treating appendicitis. The value of diagnostic laparoscopy in confirming suspected acute appendicitis has been documented [2,3]. However, it was not until the emergence of laparoscopic cholecystectomy during the last 2 to 3 years that most practicing general surgeons considered performing appendectomy using a laparoscopic approach. Laparoscopic cholecystectomy has improved the treatment of cholelithiasis by decreasing postoperative pain, reducing recovery time (both hospital and return to work), and improving cosmesis [4-8]. General surgeons in the United States now perform an estimated 80% or more of cholecystectomies using a laparoscopic technique. As therapeutic laparoscopy for cholelithiasis became more commonplace, an expansion in the use of a laparoscopic approach to other general surgical procedures such as appendectomy occurred [9]. In order to compare the safety, efficacy, and treatment results of laparoscopic appendectomy to those of traditional open appendectomy, we reviewed our series of concurrently performed open and laparoscopic appendectomies.
PATIENTS AND METHODS Patient records were reviewed for patients taken to the operating room for right lower quadrant pain and the presumptive diagnosis of acute appendicitis during the period from December 1989 (the initiation of laparoscopy to confirm this diagnosis) to February 1993. Only those adult patients cared for by three attending surgeons (BDS, SBE, JBH) were included in this series. All procedures were performed at the University of Virginia Health Sciences Center with residents in training involved in every case. Patient data were recorded and stored on a computer database program (Foxpro version 1.02, Microsoft Corp., Seattle, WA). Parameters recorded included patient age, sex, preoperative symptoms, findings of physical examination, preoperative laboratory values, duration of operation, method of appendectomy (open versus laparoscopic), intraoperative diagnosis, histologic report, length of postoperative hospitalization, postoperative complications if present, mortality, and costs of hospitalization. Complications were defined as From the Departmentof Surgery,Universityof VirginiaHealth Sci- any adverse event that delayed hospital discharge by 1 or more days. Wound complications were specifically idenences Center.Charlottesville,Virginia. Requests for reprints should be addressedto BruceD. Schirmer, tiffed and analyzed. MD, Box 181,Departmentof Surgery,Universityof VirginiaHealth The technique of open appendectomy involved use of SciencesCenter,Charlottesville,Virginia22901. a standard transverse or McBurney right lower quadrant Presented as a poster paper at the 33rd Annual Meetingof the Societyfor Surgeryof the AlimentaryTract, San Francisco,California, incision, with extension as needed in cases of perforation. May 11-13. 1992. Ligation of the appendiceal stump was performed with an
670
THE AMERICAN JOURNAL OF SURGERY VOLUME165 JUNE 1993
LAPAROSCOPIC APPENDECTOMY
absorbable suture. Inversion of the appendiceal stump was also performed (in some but not all cases) using a purse-string suture in the cecum. The technique of laparoscopic appendectomy involved initial creation of a pneumoperitoneum using a Veress needle placed through the umbilical ring. A 10- or 11-mm trocar inserted through the umbilical ring was then used to allow laparoscopy with a 0 ~ telescope. A Hasson trocar with cutdown into the peritoneum under direct visualization was used if the patient had a previous abdominal incision. Additional trocars were placed as noted in Rgme 1. The left lower quadrant port was usually a larger trocar, 12 mm in patients in whom the Endo-GIA stapler (U.S. Surgical Corp., Norwalk, CT) was used (22 of 27 patients), or 11 mm in patients in whom it was not used. The appendiceal mesentery was divided and ligated using either the stapler or Roeder loop ligatures placed endoscopically or, on occasion, clips from the endoscopic clip applier (U.S. Surgical Corp.). The appendix was divided using either the stapler or a double ligature with Roeder loops. Appendix extraction was performed using the 12mm trocar sleeve to protect the wound from contamination during removal. In female patients undergoing laparoscopy for suspected acute appendicitis, a uterine sound and a tenaculum for manipulating the uterus during the procedure were placed prior to preparing and draping the patient. A combination of this device, table positioning, and at times another port allowed adequate inspection of the pelvic organs. Data are expressed as mean 4- SEM or as group percentages. Statistical analysis of data was performed using a standard computer statistical software package (NCSS version 5.03, Dr. Jerry L. Hinteze, Kaysville, UT). Analysis of variance (GLM-ANOVA with Fisher's LSD post-hoc testing) or 9(2 analysis was used where appropriate, with statistical significance defined as p <0.05. RESULTS Patients were divided into four groups based on their surgical procedure. Fifty-four patients underwent open appendectomy (OA). Sixty-eight patients had laparoscopy performed as the initial procedure or part of their operative treatment. In six patients, diagnostic laparoscopy (DL) was the only operative therapy performed. In 47 of the 68 patients, appendectomy was also performed using laparoscopic techniques (LA), whereas in 15 patients an open incision was used to perform appendectomy (LO). This last group included patients in whom an attempt to perform LA was either not attempted or abandoned in favor of an open technique due to consideration of the anatomy or pathology (frequently appendicitis with perforation or severe scarring). DL alone was performed in only six patients, all of them female. In all DL patients, acute appendicitis was not found, and incidental appendectomy was not performed. A definitive diagnosis was established in four of the six patients using laparoscopy; in the other two pa-
;.;~: ~ ..~.- Umbilicus x"~!. ~ 11 mm sheath / ;~< j ~ - ~ : (camera) 5mm, Sheaths
Figure 1. Position of placement of the abdominal trocar sheaths for the performance of laparoscopic appendectomy. The two 5mm sheaths are generally used for retraction, whereas the 12ram sheath is used by the surgeon for dissecting and stapling instruments.
TABLE I
Patient Population Patient
Characteristics
LA
LO
OA
DL
No. of patients Age(y) Sex Male (%) Female (%) Previous abdominal surgery (%)
47 28.6--2.0
15 31.2+4.3
54 30.8_+2.0
6 37.7-+5.8
24 (51) 23 (49) 10 (21)
6 (40) 9 (60) 3 (20)
36 {67)* 18 (33)* 8 (15)
0 6 (100) 1 (20)
LA = laparoseopic appendectomy; LO = laparoscoDie and open appendec~ tomy; OA = open appendectomy; DL = diagnostic laparoscopy. *p < 0.05 for OA versus LA and LO.
tients, no intra-abdominal pathology was observed. The average duration of operation for the DL group was 63.0 4- 10.1 minutes. There were no deaths or complications of any type (including wound complications) in this group. The average length of postoperative stay was 2.00 4- 0.86 days. The DL group was not compared statistically with the other three groups due to the small number of patients in this group. Table I lists the patient population characteristics of the four groups. Patient population characteristics did not differ between the LA, LO, and OA groups for age or
THE AMERICAN JOURNAL OF SURGERY
VOLUME 165
JUNE 1993
671
SCHIRMER ET AL
TABLE II Treatment Data Operative Characteristics
LA
Duration of operation (mini Perforated appendici-
LO
86.0 • 5.6
108.3 • 7.3*
OA 81.4 • 3.3
6.4
66.7t
18.7
38.3 78.7
6.7t 86.7
25.9 74.1
0 10.6 4.3
0 46.7 t 13.3
1.9 18.9 9.4
4.79 • 1.31
5.87 • 1.58
tis (%) Normal appendix (%) Diagnosis established in operating room Mortality (%) Complications (%) Wound complications
(%) Postoperative length of 3.47 • 0.49 stay (d) Abbreviationsas in TableI. *p <0.05 for LO versusOA. tp < 0.05 for LO versus LAand OA.
TABLE III Costs Data
Cost of operating room ($) Mean Median Cost of medical supplies ($) Mean Median Overall hospital cost Mean • SEM Median
LA
LO
1,752 • 84 1,664
2,007 • 105" 1,998
1,484 • 156 1,341
2,028 • 656 1,166
OA
1,626 • 79 1,570
845 • 102t 664
$7,043 _+ 862 $15,231 -+ 5,061 $ $7,453 --+ 1,066 $5,899 $10,425 $5,220
Abbreviationsas in Table L *p <0.05 for LO comparedwith OA. tp < 0.05 for OA comparedwith LAor LO. ~p < 0.05 for LO comparedwith LA or OA.
incidence of previous abdominal surgery. There was a significantly higher percentage of females in the two groups of patients undergoing laparoscopy (LA and LO) than in the group of patients undergoing OA. The clinical presentations of patients in the various groups were generally comparable. Abdominal pain was the most common symptom and was present in virtually all patients in all groups. Most patients also complained of nausea, vomiting, and anorexia. Diarrhea was present in a minority of cases (LA: 9.3%, LO: 28.6%, OA: 23.1%). Fever was present less frequently in LA patients (25.5%) than in LO (60.0%) plus OA (54.8%) patients (p
<0.05). Physical examination demonstrated abdominal tenderness in virtually all patients. Rebound was thought to be present in about half the patients (LA: 53%, LO: 40%, 672
THE AMERICAN JOURNAL OF SURGERY
OA: 50%). Cervical motion tenderness was present more frequently in the LO group (55.6%) than in the LA (5.3%) or OA (11.8%) groups (p <0.01). The white blood cell count on admission was recorded in all patients, and all groups showed leukocytosis (LA: 13.8 4- 0.7, LO: 14.7 4- 1.3, OA: 13.8 4- 0.6, DL: 12.5 41.7 white blood cells X 103/mL), with no significant differences between groups. Treatment data are summarized in Table lI. The duration of operation for LO patients was significantly longer than for either LA or OA patients. There was no significant difference in the average duration of operation between LA and OA patients. Acute appendicitis was present clinically in 61.7% of LA patients, 86.7% of LO patients, and 74.1% of OA patients, with the incidence of perforation among the groups being significantly higher for LO patients when compared with either LA or OA patients. The incidence of normal appendix was not statistically different among groups. A definitive diagnosis was established at the time of surgery in most patients. In the LA group, this occurred in 37 of 47 patients. Appendicitis was correctly diagnosed in 27 of 29 LA patients in whom pathologic findings confirmed acute appendicitis; in 2 patients, the surgeon did not believe appendicitis was present but performed appendectomy. Of the remaining 18 LA patients in whom appendicitis was thought not to be present, a definitive intraoperative diagnosis confirmed by pathologic examination was established in 9 of 18 patients. In the OA group, acute appendicitis was present in 40 of 54 patients, with 36 of 40 intraoperative diagnoses of acute appendicitis confirmed by pathologic examination. Of the 14 patients in whom appendicitis was not present, a definitive intraoperative diagnosis was confirmed in only 4. The comparison between LA and OA in patients in whom acute appendicitis was not present but in whom a definitive diagnosis was made intraoperatively was not statistically significant. In the LO group, there were two patients in whom acute appendicitis was not present. OA was performed after laparoscopy ruled out the presence of other significant intra-abdominal pathology in those cases. Pathologic specimen confirmation of the intraoperative diagnosis occurred in most patients in all groups (LA: 89.4%; LO: 93.3%; OA: 88.9%). There was one death in the series. This occurred in an 80-year-old man who underwent OA and had significant intra-abdominal contamination. He subsequently developed myocardial infarction, sepsis, and multiorgan failure and died. Complication rates were 10.6% in the LA group, 46.7% in the LO group, 18.9% in the OA group, and 0% in the DL group. There were no significant differences in complication rates between the LA, LO, and OA groups. However, the incidence of complications for the LO group compared with the combined LA and OA groups was significant. Wound complication rates were not significantly different between groups nor was the postoperative length of stay (Table II). Available data concerning hospital costs are given in Table III for the LA (n = 27), LO (n = 14), and OA (n =
VOLUME 165
JUNE 1993
LAPAROSCOPIC APPENDECTOMY
41) groups. Itemized costs for both hospital operating room time charges and medical supply charges are given, as well as overall hospital costs. Both mean and median values are listed. The latter are thought to be more representative of the groups as a whole since each group had one or a few patients with excessive costs. The cost of medical supplies was lower for the OA group when compared with either the LA or the LO group. Cost of operating room time was lower for the OA group when compared with the LO group but not with the LA group. Overall hospital costs were lower for the LO group compared with either the LA or the OA group. COMMENTS Laparoscopic surgery is increasing in popularity among general surgeons in the United States. In 1989 and 1990, controversy over the safety and efficacy of laparoscopic cholecystectomy developed. Reports showed the procedure could be performed with an acceptable degree of safety [4-8]. Since laparoscopic cholecystectomy conveyed a substantial benefit to patients in terms of decreased postoperative pain, length of hospitalization, size of incision and scars, and return to normal activity, its rapid gain in popularity was predictable. Despite these well-documented benefits, the issue of an increased incidence of bile duct injuries during laparoscopic cholecystectomy has been raised [10-12]. The experience with laparoscopic cholecystectomy suggests careful review of any new similar laparoscopic procedure is indicated before its widespread application. Having the ability to perform a procedure using a laparoscopic technique does not necessarily mean that it is the best method of accomplishing the procedure. Safety and efficacy need to be demonstrated for each additional procedure to which a laparoscopic approach is being initiated. For these reasons, we report our results with laparoscopic appendectomy. Prior experiences regarding laparoscopic appendectomy have been reported mainly from Europe. Semm [1] first described the ability to successfully perform the procedure. Large studies of children [13] and adults [14] have been reported. The largest experience reported in adults is by Pier et al [14], a series of 625 laparoscopic appendectomies performed for 678 patients with suspected appendicitis. In this series, only 6% of the procedures were performed using a traditional open technique, and 2% of procedures were converted to an open technique after attempted laparoscopic removal. The authors report an average operative time of 15 to 20 minutes, with few significant complications (three cases of bleeding and three of abscess formation). The incidence of normal appendix removal for the series was 14%, with histologic confirmation of inflammation in all other specimens. Duration of operation in our series is considerably longer than that reported by Pier et al [14]. Factors contributing to this finding likely include the learning curve of our initial experience as well as the performance of most of these procedures in a resident teaching manner. Surgical residents performed the role of first assistant or surgeon in all cases and stood in the "surgeon's"
position in most cases. The operative duration of 89 minutes for LA is similar to two reports on LA in the United States [15,16]. Resident involvement was reported in all cases in one of those reports [15]. In the other, the operative duration of LA was 119 minutes [16]. Apelgren et al [16] also reported an increased duration of operation for LA compared with OA (average duration: 68.5 minutes). In our series, OA took nearly as long as LA. One factor that may have contributed to the relatively long duration of OA in our series is the fact that first-year resident participation was more common in OA, whereas LA was performed exclusively by senior or chief residents. These data confirm the previously reported finding that laparoscopy is a valuable tool in the management of acute appendicitis [2,3]. The technique allows the surgeon to accurately make that diagnosis. In our series, in only one patient (early in the experience) was an appendix removed using a laparoscopic technique that the surgeon believed probably had inflammation and in whom histologic confirmation was lacking. In one other case, adhesions around the appendix and the histologic diagnosis of periappendicitis were found when intraoperative presumption of appendicitis occurred. The incidence of normal appendix removal (negative appendectomy) reported by us is higher than that reported by Pier et al [14], being about 29% of appendectomies performed for the overall series. The incidence of cases in which acute appendicitis was not found at operation for our series was 32% overall, a figure that is comparable to several large series of OAs previously reported in the literature [17-20]. Negative appendectomy is not accomplished without some morbidity, with wound complication rates of 2% to 4% [21,22] and overall complication rates of 10% to 15% [19,21] reported in the literature. LA may lower these rates. The two wound infections that occurred in the LA group in our series were both the result of clear errors in laparoscopic technique: in both cases, the acutely inflamed appendix was pulled through a wound without adequate serosal protection, resulting in wound contamination. Avoidance of such obvious errors would have likely resulted in a zero wound complication rate for the LA group. Lowering wound complications alone will lower the overall complication rate. However, it would seem likely that other complications such as postoperative small bowel obstruction, incisional hernia, and morbidity from an undiagnosed source of intra-abdominal inflammation may all be lowered using the laparoscopic approach in diagnosing and treating this patient population. Confirmation of a correct diagnosis in patients without acute appendicitis was not statistically higher in the LA group (50%) than in the OA group (29%), but the numbers suggest that these differences may become significant with a larger patient sample. LA may offer patients the additional benefit of decreased hospital stay. There was not a significant difference for the LA group when compared with the OA group in this series, but the lower average length of hospi-
THE AMERICAN JOURNAL OF SURGERY
VOLUME 165
JUNE 1993
673
SCHIRMER ET AL
talization of the LA group may become significant with increasing numbers of patients. It was interesting to observe a lack of significant overall hospitalization costs when comparing the LA group with the OA group. The mean cost is actually numerically higher for the OA group, but median data show the LA group to be numerically higher (neither comparison reaches statistical significance). These data reflect the fact that one or several patients in the OA group had very high hospitalization costs (one was the patient who died). As might be predicted, the cost of medical supplies is significantly higher for both LA and LO when compared with OA, probably reflecting the increased cost of laparoscopic instrumentation. However, the cost of operating time is comparable for OA and LA, and the fact that the overall cost of hospitalization was also similar for these groups suggests that some items, perhaps numerically lower charges for shorter hospital stay in the LA group (although this parameter was not significantly different by statistical analysis), may account for these observations. In all categories, the LO group had higher charges, as would be predicted from their longer average duration of operation, the need to use both laparoscopic and regular instrumentation, and the high incidence of postoperative complications for the group. The LO group was composed largely of patients early in the series in whom the appendix was believed to be too inflamed to remove safely using laparoscopie techniques. Therefore, the composition of the LO group was biased toward patients with slightly worse disease, as manifested by the increased hospitalization cost and perforation rate for this group. Preoperative parameters were generally not helpful in differentiating the LO group from the LA and OA groups, with cervical motion tenderness on physical examination being the only finding significantly higher in the LO group. During the last 10 months of the series, there have been only two patients added to the LO group, which suggests that, with increasing experience, more difficult cases of acute appendicitis, including some with perforation, can be successfully treated using a laparoscopic approach. Population and preoperative characteristics of the LA and OA groups were also very comparable except for sex distribution. This difference in sex distribution likely reflects the greater difficulty in correctly diagnosing acute appendicitis preoperatively in women compared with men. The surgeon was therefore more likely to use laparoscopy as an additional initial tool to confirm the diagnosis of acute appendicitis. The criticism can be raised regarding the separate analysis of LO patients from the LA group, since the intent to treat using laparoscopy was the same in both groups. If such analysis were to be done, any tendencies seen toward higher complications for OA versus LA would be negated, suggesting that, in the early experience with LA, little improvement in clinical outcome may be seen, whereas cost differences could still be higher for the overall laparoscopic group. Surgeons now have the option of offering their pa674
tients the possibility of incidental appendectomy using a laparoscopic technique in the event no appendicitis is found. Our experience suggests that this can be done with acceptably low morbidity, since there was no significant morbidity in performing appendectomy in the setting of no acute appendicitis. More data are needed to confirm the suspected decreased morbidity of LA versus OA in this clinical situation. Appendectomy has traditionally been, in academic teaching centers, an operation during which junior residents can gain valuable operative experience. The same is true for appendectomy in the array of laparoscopic procedures. As minimally invasive surgery becomes more commonly practiced in the years ahead, the use of LA as a procedure in which residents may learn the basics of laparoscopic surgery is very likely. Obviously, the supervision and performance of LA by a general surgeon skilled and experienced in the procedure are required in these circumstances. In summary, based on the data presented and our experience to date, laparoscopy and LA are safe and effective additions to the successful diagnosis and treatment of patients with suspected acute appendicitis. These procedures may potentially supply greater information to the surgeon at the time of operation. When successfully used, they resulted in similar complication rates, length of hospitalization, and overall costs of hospitalization when compared with OA. With increasing experience, these procedures can be performed with increasing reliability for a larger percentage of patients with all forms of acute appendicitis or the clinical picture suggesting that disease. The data produced by this study do not confirm a superior role of either LA or OA in the diagnosis and treatment of suspected acute appendicitis. These data are evidence that a controlled randomized trial comparing these two treatments is indicated. Such a trial should ideally include large enough numbers of patients to allow conclusions to be drawn. It should be conducted by surgeons sufficiently skilled in LA so that biases potentially introduced by the learning process are avoided. REFERENCES 1. Semm K. Endoscopic appendectomy. Endoscopy 1983; 15: 59-64. 2. Leape LL, RamenofskyMI. Laparoscopyfor questionable appendicitis. Ann Surg 1980; 191: 410-3. 3. WhitworthCM, WhitworthPW, SanfillipoJ, PolkHC. Valueof diagnosticlaparoscopyin youngwomenwith possibleappendicitis. Surg GynecolObstet 1988; 167: 187-90. 4. Dubois F, Icard P, BerthelotG, Levard H. Coelioscopiccholecystectomy:a preliminaryreport of 36 cases. Ann Surg 1990; 211: 60-2. 5. CushieriA, Dubois F, MouielJ, et al. The Europeanexperience with laparoscopiccholecystectomy.Am J Surg 1991; 161: 385-7. 6. MeyersWC, AghazarianSG, AlbertsonDA, et al. A prospective analysis of 1518 laparoscopic cholecystectomiesperformed by southern U.S. surgeons. N Engl J Med 1991; 324: 1075-8. 7. Schirmer BD, Edge SB, Dix J, Hyser MJ, Hanks JB, Jones RS. Laparoscopiccholecystectomy:treatmentof choicefor symptomatic cholelithiasis.Ann Surg 1991; 213: 665-77.
THE AMERICAN JOURNAL OF SURGERY VOLUME165 JUNE 1993
LAPAROSCOPICAPPENDECTOMY
8. Reddick E J, Olsen D, Spaw A, et al. Safe performance of difficult laparoscopic cholecystectomies. Am J Surg 1991; 161: 377-81. 9. Reddick E J, Saye WB. Laparoscopic appendectomy. In: Zucker KA, editor. Surgical laparoscopy. St. Louis: Quality Medical Publishing, 1991: 227-39. 10. Davidoff AM, Pappas TN, Murray EN, et al. Mechanisms of major biliary injury during laparoscopic cholecystectomy. Ann Surg 1992; 215: 196-202. 11. Moosa AR, Easter DW, VonSonnenberg E, Casola G, D'Agostino H. Laparoscopic injuries to the bile duct. Ann Surg 1992; 215: 203-8. 12. Meyer H. Tighter rules urged on new gallbladder surgery. American Medical News. June 1, 1992. 13. Valla JS, Limonne B, Valla V, et al. Laparoscopic appendectomy in children: report of 465 cases. Surg Laparosc Endosc 1991; 1: 166-72. 14. Pier A, Gotz F, Bacher C. Laparoscopic appendectomy in 625 cases: from innovation to routine. Surg Laparosc Endosc 1991; 1: 8-13.
15. Scott-Conner C, Hall T J, Anglin B, Muakkassa F. Laparoscopic appendectomy: initial experience in a teaching program. Ann Surg 1992; 215: 660-8. 16. Apelgren KN, Molnar RG, Kisala JM. Is laparoscopic better than open appendectomy? Surg Endosc 1992; 6: 298-301. 17. VanWay CW III, Murphy JR, Dunn EL, Elerding SC. A feasibility study of computer aided diagnosis in appendicitis. Surg Gynecol Obstet 1982; 155: 685-8. 18. Teieher I, Landa B, Cohen M, Kabnick LS, Wise L. Scoring system to aid in diagnoses of appendicitis. Ann Surg 1983; 198: 553-9. 19. Chang FC, Hogle HH, Welling BA. The fate of the negative appendix. Am J Surg 1973; 126: 752-4. 20. Mason KB, Deyden WE. Primary appendectomy. Am Surg 1976; 42: 239-43. 21. Lewis FR, Holcroft JW, Boey J, Dunphy JE. Appendicitis. A critical review of diagnosis and treatment in 1,000 cases. Arch Surg 1975; 110: 677-84. 22. Berry J, Malt RA. Appendicitis near its centenary. Ann Surg 1984; 200: 567-75.
THE AMERICAN JOURNAL OF SURGERY VOLUME 165 JUNE 1993 675
We compared the results of concurrently performed laparoscopic versus open appendectomy as treatments for suspected acute appendicitis. The 68 laparoseopic procedures resulted in 62 appendectomies, 47 by the laparoscopie (LA) technique and I5 by the open (LO) technique. Another 54 patients underwent open appendectomy (OA). Significantly more females underwent laparoscopy (LA and LO: 52% versus OA: 33%, p = 0 . 0 4 7 ) . Operative duration was shortest for OA (81 • 3 minutes), which was shorter than for LO (108 4- 7 minutes), but not different than LA (86 4- 6 minues). The postoperative length of stay was not different for LA (3.5 4- 0.5 days) compared with OA (5.9 4- 1.6 days) or LO (4.8 4- 1.3 days). One death occurred in the OA group. Wound complication rates were not significantly different for LA (4.3%) compared with OA (9.4%) and LO (13.3%). Overall complication rates were lower for LA (10.6%) and OA (18.9%) compared with LO (46.7%, p < 0 . 0 1 ) . Median hospital cost for LO ($10,425) was higher (p < 0 . 0 2 ) than for either LA ($5,899) or OA ( $ 5 , 2 2 0 ) . When appendicitis was not present, definitive confirmation of pathology was achieved in 9 of 18 patients undergoing LA versus 4 of 14 patients having OA (p = not significant). We conclude that when laparoseopy and laparoscopie appendectomy can be performed, the procedure is safe and produces results comparable with those of open appendectomy without significant overall cost differences.
L
aparoscopic appendectomy was first described by Semm [1] in the early 1980s. Initial acceptance was very limited due to a low prevalence of laparoscopic skills among general surgeons treating appendicitis. The value of diagnostic laparoscopy in confirming suspected acute appendicitis has been documented [2,3]. However, it was not until the emergence of laparoscopic cholecystectomy during the last 2 to 3 years that most practicing general surgeons considered performing appendectomy using a laparoscopic approach. Laparoscopic cholecystectomy has improved the treatment of cholelithiasis by decreasing postoperative pain, reducing recovery time (both hospital and return to work), and improving cosmesis [4-8]. General surgeons in the United States now perform an estimated 80% or more of cholecystectomies using a laparoscopic technique. As therapeutic laparoscopy for cholelithiasis became more commonplace, an expansion in the use of a laparoscopic approach to other general surgical procedures such as appendectomy occurred [9]. In order to compare the safety, efficacy, and treatment results of laparoscopic appendectomy to those of traditional open appendectomy, we reviewed our series of concurrently performed open and laparoscopic appendectomies.
PATIENTS AND METHODS Patient records were reviewed for patients taken to the operating room for right lower quadrant pain and the presumptive diagnosis of acute appendicitis during the period from December 1989 (the initiation of laparoscopy to confirm this diagnosis) to February 1993. Only those adult patients cared for by three attending surgeons (BDS, SBE, JBH) were included in this series. All procedures were performed at the University of Virginia Health Sciences Center with residents in training involved in every case. Patient data were recorded and stored on a computer database program (Foxpro version 1.02, Microsoft Corp., Seattle, WA). Parameters recorded included patient age, sex, preoperative symptoms, findings of physical examination, preoperative laboratory values, duration of operation, method of appendectomy (open versus laparoscopic), intraoperative diagnosis, histologic report, length of postoperative hospitalization, postoperative complications if present, mortality, and costs of hospitalization. Complications were defined as From the Departmentof Surgery,Universityof VirginiaHealth Sci- any adverse event that delayed hospital discharge by 1 or more days. Wound complications were specifically idenences Center.Charlottesville,Virginia. Requests for reprints should be addressedto BruceD. Schirmer, tiffed and analyzed. MD, Box 181,Departmentof Surgery,Universityof VirginiaHealth The technique of open appendectomy involved use of SciencesCenter,Charlottesville,Virginia22901. a standard transverse or McBurney right lower quadrant Presented as a poster paper at the 33rd Annual Meetingof the Societyfor Surgeryof the AlimentaryTract, San Francisco,California, incision, with extension as needed in cases of perforation. May 11-13. 1992. Ligation of the appendiceal stump was performed with an
670
THE AMERICAN JOURNAL OF SURGERY VOLUME165 JUNE 1993
LAPAROSCOPIC APPENDECTOMY
absorbable suture. Inversion of the appendiceal stump was also performed (in some but not all cases) using a purse-string suture in the cecum. The technique of laparoscopic appendectomy involved initial creation of a pneumoperitoneum using a Veress needle placed through the umbilical ring. A 10- or 11-mm trocar inserted through the umbilical ring was then used to allow laparoscopy with a 0 ~ telescope. A Hasson trocar with cutdown into the peritoneum under direct visualization was used if the patient had a previous abdominal incision. Additional trocars were placed as noted in Rgme 1. The left lower quadrant port was usually a larger trocar, 12 mm in patients in whom the Endo-GIA stapler (U.S. Surgical Corp., Norwalk, CT) was used (22 of 27 patients), or 11 mm in patients in whom it was not used. The appendiceal mesentery was divided and ligated using either the stapler or Roeder loop ligatures placed endoscopically or, on occasion, clips from the endoscopic clip applier (U.S. Surgical Corp.). The appendix was divided using either the stapler or a double ligature with Roeder loops. Appendix extraction was performed using the 12mm trocar sleeve to protect the wound from contamination during removal. In female patients undergoing laparoscopy for suspected acute appendicitis, a uterine sound and a tenaculum for manipulating the uterus during the procedure were placed prior to preparing and draping the patient. A combination of this device, table positioning, and at times another port allowed adequate inspection of the pelvic organs. Data are expressed as mean 4- SEM or as group percentages. Statistical analysis of data was performed using a standard computer statistical software package (NCSS version 5.03, Dr. Jerry L. Hinteze, Kaysville, UT). Analysis of variance (GLM-ANOVA with Fisher's LSD post-hoc testing) or 9(2 analysis was used where appropriate, with statistical significance defined as p <0.05. RESULTS Patients were divided into four groups based on their surgical procedure. Fifty-four patients underwent open appendectomy (OA). Sixty-eight patients had laparoscopy performed as the initial procedure or part of their operative treatment. In six patients, diagnostic laparoscopy (DL) was the only operative therapy performed. In 47 of the 68 patients, appendectomy was also performed using laparoscopic techniques (LA), whereas in 15 patients an open incision was used to perform appendectomy (LO). This last group included patients in whom an attempt to perform LA was either not attempted or abandoned in favor of an open technique due to consideration of the anatomy or pathology (frequently appendicitis with perforation or severe scarring). DL alone was performed in only six patients, all of them female. In all DL patients, acute appendicitis was not found, and incidental appendectomy was not performed. A definitive diagnosis was established in four of the six patients using laparoscopy; in the other two pa-
;.;~: ~ ..~.- Umbilicus x"~!. ~ 11 mm sheath / ;~< j ~ - ~ : (camera) 5mm, Sheaths
Figure 1. Position of placement of the abdominal trocar sheaths for the performance of laparoscopic appendectomy. The two 5mm sheaths are generally used for retraction, whereas the 12ram sheath is used by the surgeon for dissecting and stapling instruments.
TABLE I
Patient Population Patient
Characteristics
LA
LO
OA
DL
No. of patients Age(y) Sex Male (%) Female (%) Previous abdominal surgery (%)
47 28.6--2.0
15 31.2+4.3
54 30.8_+2.0
6 37.7-+5.8
24 (51) 23 (49) 10 (21)
6 (40) 9 (60) 3 (20)
36 {67)* 18 (33)* 8 (15)
0 6 (100) 1 (20)
LA = laparoseopic appendectomy; LO = laparoscoDie and open appendec~ tomy; OA = open appendectomy; DL = diagnostic laparoscopy. *p < 0.05 for OA versus LA and LO.
tients, no intra-abdominal pathology was observed. The average duration of operation for the DL group was 63.0 4- 10.1 minutes. There were no deaths or complications of any type (including wound complications) in this group. The average length of postoperative stay was 2.00 4- 0.86 days. The DL group was not compared statistically with the other three groups due to the small number of patients in this group. Table I lists the patient population characteristics of the four groups. Patient population characteristics did not differ between the LA, LO, and OA groups for age or
THE AMERICAN JOURNAL OF SURGERY
VOLUME 165
JUNE 1993
671
SCHIRMER ET AL
TABLE II Treatment Data Operative Characteristics
LA
Duration of operation (mini Perforated appendici-
LO
86.0 • 5.6
108.3 • 7.3*
OA 81.4 • 3.3
6.4
66.7t
18.7
38.3 78.7
6.7t 86.7
25.9 74.1
0 10.6 4.3
0 46.7 t 13.3
1.9 18.9 9.4
4.79 • 1.31
5.87 • 1.58
tis (%) Normal appendix (%) Diagnosis established in operating room Mortality (%) Complications (%) Wound complications
(%) Postoperative length of 3.47 • 0.49 stay (d) Abbreviationsas in TableI. *p <0.05 for LO versusOA. tp < 0.05 for LO versus LAand OA.
TABLE III Costs Data
Cost of operating room ($) Mean Median Cost of medical supplies ($) Mean Median Overall hospital cost Mean • SEM Median
LA
LO
1,752 • 84 1,664
2,007 • 105" 1,998
1,484 • 156 1,341
2,028 • 656 1,166
OA
1,626 • 79 1,570
845 • 102t 664
$7,043 _+ 862 $15,231 -+ 5,061 $ $7,453 --+ 1,066 $5,899 $10,425 $5,220
Abbreviationsas in Table L *p <0.05 for LO comparedwith OA. tp < 0.05 for OA comparedwith LAor LO. ~p < 0.05 for LO comparedwith LA or OA.
incidence of previous abdominal surgery. There was a significantly higher percentage of females in the two groups of patients undergoing laparoscopy (LA and LO) than in the group of patients undergoing OA. The clinical presentations of patients in the various groups were generally comparable. Abdominal pain was the most common symptom and was present in virtually all patients in all groups. Most patients also complained of nausea, vomiting, and anorexia. Diarrhea was present in a minority of cases (LA: 9.3%, LO: 28.6%, OA: 23.1%). Fever was present less frequently in LA patients (25.5%) than in LO (60.0%) plus OA (54.8%) patients (p
<0.05). Physical examination demonstrated abdominal tenderness in virtually all patients. Rebound was thought to be present in about half the patients (LA: 53%, LO: 40%, 672
THE AMERICAN JOURNAL OF SURGERY
OA: 50%). Cervical motion tenderness was present more frequently in the LO group (55.6%) than in the LA (5.3%) or OA (11.8%) groups (p <0.01). The white blood cell count on admission was recorded in all patients, and all groups showed leukocytosis (LA: 13.8 4- 0.7, LO: 14.7 4- 1.3, OA: 13.8 4- 0.6, DL: 12.5 41.7 white blood cells X 103/mL), with no significant differences between groups. Treatment data are summarized in Table lI. The duration of operation for LO patients was significantly longer than for either LA or OA patients. There was no significant difference in the average duration of operation between LA and OA patients. Acute appendicitis was present clinically in 61.7% of LA patients, 86.7% of LO patients, and 74.1% of OA patients, with the incidence of perforation among the groups being significantly higher for LO patients when compared with either LA or OA patients. The incidence of normal appendix was not statistically different among groups. A definitive diagnosis was established at the time of surgery in most patients. In the LA group, this occurred in 37 of 47 patients. Appendicitis was correctly diagnosed in 27 of 29 LA patients in whom pathologic findings confirmed acute appendicitis; in 2 patients, the surgeon did not believe appendicitis was present but performed appendectomy. Of the remaining 18 LA patients in whom appendicitis was thought not to be present, a definitive intraoperative diagnosis confirmed by pathologic examination was established in 9 of 18 patients. In the OA group, acute appendicitis was present in 40 of 54 patients, with 36 of 40 intraoperative diagnoses of acute appendicitis confirmed by pathologic examination. Of the 14 patients in whom appendicitis was not present, a definitive intraoperative diagnosis was confirmed in only 4. The comparison between LA and OA in patients in whom acute appendicitis was not present but in whom a definitive diagnosis was made intraoperatively was not statistically significant. In the LO group, there were two patients in whom acute appendicitis was not present. OA was performed after laparoscopy ruled out the presence of other significant intra-abdominal pathology in those cases. Pathologic specimen confirmation of the intraoperative diagnosis occurred in most patients in all groups (LA: 89.4%; LO: 93.3%; OA: 88.9%). There was one death in the series. This occurred in an 80-year-old man who underwent OA and had significant intra-abdominal contamination. He subsequently developed myocardial infarction, sepsis, and multiorgan failure and died. Complication rates were 10.6% in the LA group, 46.7% in the LO group, 18.9% in the OA group, and 0% in the DL group. There were no significant differences in complication rates between the LA, LO, and OA groups. However, the incidence of complications for the LO group compared with the combined LA and OA groups was significant. Wound complication rates were not significantly different between groups nor was the postoperative length of stay (Table II). Available data concerning hospital costs are given in Table III for the LA (n = 27), LO (n = 14), and OA (n =
VOLUME 165
JUNE 1993
LAPAROSCOPIC APPENDECTOMY
41) groups. Itemized costs for both hospital operating room time charges and medical supply charges are given, as well as overall hospital costs. Both mean and median values are listed. The latter are thought to be more representative of the groups as a whole since each group had one or a few patients with excessive costs. The cost of medical supplies was lower for the OA group when compared with either the LA or the LO group. Cost of operating room time was lower for the OA group when compared with the LO group but not with the LA group. Overall hospital costs were lower for the LO group compared with either the LA or the OA group. COMMENTS Laparoscopic surgery is increasing in popularity among general surgeons in the United States. In 1989 and 1990, controversy over the safety and efficacy of laparoscopic cholecystectomy developed. Reports showed the procedure could be performed with an acceptable degree of safety [4-8]. Since laparoscopic cholecystectomy conveyed a substantial benefit to patients in terms of decreased postoperative pain, length of hospitalization, size of incision and scars, and return to normal activity, its rapid gain in popularity was predictable. Despite these well-documented benefits, the issue of an increased incidence of bile duct injuries during laparoscopic cholecystectomy has been raised [10-12]. The experience with laparoscopic cholecystectomy suggests careful review of any new similar laparoscopic procedure is indicated before its widespread application. Having the ability to perform a procedure using a laparoscopic technique does not necessarily mean that it is the best method of accomplishing the procedure. Safety and efficacy need to be demonstrated for each additional procedure to which a laparoscopic approach is being initiated. For these reasons, we report our results with laparoscopic appendectomy. Prior experiences regarding laparoscopic appendectomy have been reported mainly from Europe. Semm [1] first described the ability to successfully perform the procedure. Large studies of children [13] and adults [14] have been reported. The largest experience reported in adults is by Pier et al [14], a series of 625 laparoscopic appendectomies performed for 678 patients with suspected appendicitis. In this series, only 6% of the procedures were performed using a traditional open technique, and 2% of procedures were converted to an open technique after attempted laparoscopic removal. The authors report an average operative time of 15 to 20 minutes, with few significant complications (three cases of bleeding and three of abscess formation). The incidence of normal appendix removal for the series was 14%, with histologic confirmation of inflammation in all other specimens. Duration of operation in our series is considerably longer than that reported by Pier et al [14]. Factors contributing to this finding likely include the learning curve of our initial experience as well as the performance of most of these procedures in a resident teaching manner. Surgical residents performed the role of first assistant or surgeon in all cases and stood in the "surgeon's"
position in most cases. The operative duration of 89 minutes for LA is similar to two reports on LA in the United States [15,16]. Resident involvement was reported in all cases in one of those reports [15]. In the other, the operative duration of LA was 119 minutes [16]. Apelgren et al [16] also reported an increased duration of operation for LA compared with OA (average duration: 68.5 minutes). In our series, OA took nearly as long as LA. One factor that may have contributed to the relatively long duration of OA in our series is the fact that first-year resident participation was more common in OA, whereas LA was performed exclusively by senior or chief residents. These data confirm the previously reported finding that laparoscopy is a valuable tool in the management of acute appendicitis [2,3]. The technique allows the surgeon to accurately make that diagnosis. In our series, in only one patient (early in the experience) was an appendix removed using a laparoscopic technique that the surgeon believed probably had inflammation and in whom histologic confirmation was lacking. In one other case, adhesions around the appendix and the histologic diagnosis of periappendicitis were found when intraoperative presumption of appendicitis occurred. The incidence of normal appendix removal (negative appendectomy) reported by us is higher than that reported by Pier et al [14], being about 29% of appendectomies performed for the overall series. The incidence of cases in which acute appendicitis was not found at operation for our series was 32% overall, a figure that is comparable to several large series of OAs previously reported in the literature [17-20]. Negative appendectomy is not accomplished without some morbidity, with wound complication rates of 2% to 4% [21,22] and overall complication rates of 10% to 15% [19,21] reported in the literature. LA may lower these rates. The two wound infections that occurred in the LA group in our series were both the result of clear errors in laparoscopic technique: in both cases, the acutely inflamed appendix was pulled through a wound without adequate serosal protection, resulting in wound contamination. Avoidance of such obvious errors would have likely resulted in a zero wound complication rate for the LA group. Lowering wound complications alone will lower the overall complication rate. However, it would seem likely that other complications such as postoperative small bowel obstruction, incisional hernia, and morbidity from an undiagnosed source of intra-abdominal inflammation may all be lowered using the laparoscopic approach in diagnosing and treating this patient population. Confirmation of a correct diagnosis in patients without acute appendicitis was not statistically higher in the LA group (50%) than in the OA group (29%), but the numbers suggest that these differences may become significant with a larger patient sample. LA may offer patients the additional benefit of decreased hospital stay. There was not a significant difference for the LA group when compared with the OA group in this series, but the lower average length of hospi-
THE AMERICAN JOURNAL OF SURGERY
VOLUME 165
JUNE 1993
673
SCHIRMER ET AL
talization of the LA group may become significant with increasing numbers of patients. It was interesting to observe a lack of significant overall hospitalization costs when comparing the LA group with the OA group. The mean cost is actually numerically higher for the OA group, but median data show the LA group to be numerically higher (neither comparison reaches statistical significance). These data reflect the fact that one or several patients in the OA group had very high hospitalization costs (one was the patient who died). As might be predicted, the cost of medical supplies is significantly higher for both LA and LO when compared with OA, probably reflecting the increased cost of laparoscopic instrumentation. However, the cost of operating time is comparable for OA and LA, and the fact that the overall cost of hospitalization was also similar for these groups suggests that some items, perhaps numerically lower charges for shorter hospital stay in the LA group (although this parameter was not significantly different by statistical analysis), may account for these observations. In all categories, the LO group had higher charges, as would be predicted from their longer average duration of operation, the need to use both laparoscopic and regular instrumentation, and the high incidence of postoperative complications for the group. The LO group was composed largely of patients early in the series in whom the appendix was believed to be too inflamed to remove safely using laparoscopie techniques. Therefore, the composition of the LO group was biased toward patients with slightly worse disease, as manifested by the increased hospitalization cost and perforation rate for this group. Preoperative parameters were generally not helpful in differentiating the LO group from the LA and OA groups, with cervical motion tenderness on physical examination being the only finding significantly higher in the LO group. During the last 10 months of the series, there have been only two patients added to the LO group, which suggests that, with increasing experience, more difficult cases of acute appendicitis, including some with perforation, can be successfully treated using a laparoscopic approach. Population and preoperative characteristics of the LA and OA groups were also very comparable except for sex distribution. This difference in sex distribution likely reflects the greater difficulty in correctly diagnosing acute appendicitis preoperatively in women compared with men. The surgeon was therefore more likely to use laparoscopy as an additional initial tool to confirm the diagnosis of acute appendicitis. The criticism can be raised regarding the separate analysis of LO patients from the LA group, since the intent to treat using laparoscopy was the same in both groups. If such analysis were to be done, any tendencies seen toward higher complications for OA versus LA would be negated, suggesting that, in the early experience with LA, little improvement in clinical outcome may be seen, whereas cost differences could still be higher for the overall laparoscopic group. Surgeons now have the option of offering their pa674
tients the possibility of incidental appendectomy using a laparoscopic technique in the event no appendicitis is found. Our experience suggests that this can be done with acceptably low morbidity, since there was no significant morbidity in performing appendectomy in the setting of no acute appendicitis. More data are needed to confirm the suspected decreased morbidity of LA versus OA in this clinical situation. Appendectomy has traditionally been, in academic teaching centers, an operation during which junior residents can gain valuable operative experience. The same is true for appendectomy in the array of laparoscopic procedures. As minimally invasive surgery becomes more commonly practiced in the years ahead, the use of LA as a procedure in which residents may learn the basics of laparoscopic surgery is very likely. Obviously, the supervision and performance of LA by a general surgeon skilled and experienced in the procedure are required in these circumstances. In summary, based on the data presented and our experience to date, laparoscopy and LA are safe and effective additions to the successful diagnosis and treatment of patients with suspected acute appendicitis. These procedures may potentially supply greater information to the surgeon at the time of operation. When successfully used, they resulted in similar complication rates, length of hospitalization, and overall costs of hospitalization when compared with OA. With increasing experience, these procedures can be performed with increasing reliability for a larger percentage of patients with all forms of acute appendicitis or the clinical picture suggesting that disease. The data produced by this study do not confirm a superior role of either LA or OA in the diagnosis and treatment of suspected acute appendicitis. These data are evidence that a controlled randomized trial comparing these two treatments is indicated. Such a trial should ideally include large enough numbers of patients to allow conclusions to be drawn. It should be conducted by surgeons sufficiently skilled in LA so that biases potentially introduced by the learning process are avoided. REFERENCES 1. Semm K. Endoscopic appendectomy. Endoscopy 1983; 15: 59-64. 2. Leape LL, RamenofskyMI. Laparoscopyfor questionable appendicitis. Ann Surg 1980; 191: 410-3. 3. WhitworthCM, WhitworthPW, SanfillipoJ, PolkHC. Valueof diagnosticlaparoscopyin youngwomenwith possibleappendicitis. Surg GynecolObstet 1988; 167: 187-90. 4. Dubois F, Icard P, BerthelotG, Levard H. Coelioscopiccholecystectomy:a preliminaryreport of 36 cases. Ann Surg 1990; 211: 60-2. 5. CushieriA, Dubois F, MouielJ, et al. The Europeanexperience with laparoscopiccholecystectomy.Am J Surg 1991; 161: 385-7. 6. MeyersWC, AghazarianSG, AlbertsonDA, et al. A prospective analysis of 1518 laparoscopic cholecystectomiesperformed by southern U.S. surgeons. N Engl J Med 1991; 324: 1075-8. 7. Schirmer BD, Edge SB, Dix J, Hyser MJ, Hanks JB, Jones RS. Laparoscopiccholecystectomy:treatmentof choicefor symptomatic cholelithiasis.Ann Surg 1991; 213: 665-77.
THE AMERICAN JOURNAL OF SURGERY VOLUME165 JUNE 1993
LAPAROSCOPICAPPENDECTOMY
8. Reddick E J, Olsen D, Spaw A, et al. Safe performance of difficult laparoscopic cholecystectomies. Am J Surg 1991; 161: 377-81. 9. Reddick E J, Saye WB. Laparoscopic appendectomy. In: Zucker KA, editor. Surgical laparoscopy. St. Louis: Quality Medical Publishing, 1991: 227-39. 10. Davidoff AM, Pappas TN, Murray EN, et al. Mechanisms of major biliary injury during laparoscopic cholecystectomy. Ann Surg 1992; 215: 196-202. 11. Moosa AR, Easter DW, VonSonnenberg E, Casola G, D'Agostino H. Laparoscopic injuries to the bile duct. Ann Surg 1992; 215: 203-8. 12. Meyer H. Tighter rules urged on new gallbladder surgery. American Medical News. June 1, 1992. 13. Valla JS, Limonne B, Valla V, et al. Laparoscopic appendectomy in children: report of 465 cases. Surg Laparosc Endosc 1991; 1: 166-72. 14. Pier A, Gotz F, Bacher C. Laparoscopic appendectomy in 625 cases: from innovation to routine. Surg Laparosc Endosc 1991; 1: 8-13.
15. Scott-Conner C, Hall T J, Anglin B, Muakkassa F. Laparoscopic appendectomy: initial experience in a teaching program. Ann Surg 1992; 215: 660-8. 16. Apelgren KN, Molnar RG, Kisala JM. Is laparoscopic better than open appendectomy? Surg Endosc 1992; 6: 298-301. 17. VanWay CW III, Murphy JR, Dunn EL, Elerding SC. A feasibility study of computer aided diagnosis in appendicitis. Surg Gynecol Obstet 1982; 155: 685-8. 18. Teieher I, Landa B, Cohen M, Kabnick LS, Wise L. Scoring system to aid in diagnoses of appendicitis. Ann Surg 1983; 198: 553-9. 19. Chang FC, Hogle HH, Welling BA. The fate of the negative appendix. Am J Surg 1973; 126: 752-4. 20. Mason KB, Deyden WE. Primary appendectomy. Am Surg 1976; 42: 239-43. 21. Lewis FR, Holcroft JW, Boey J, Dunphy JE. Appendicitis. A critical review of diagnosis and treatment in 1,000 cases. Arch Surg 1975; 110: 677-84. 22. Berry J, Malt RA. Appendicitis near its centenary. Ann Surg 1984; 200: 567-75.
THE AMERICAN JOURNAL OF SURGERY VOLUME 165 JUNE 1993 675