RESULTS

LEARNING CURVE AND RESULTS OF PERSONAL EXPERIENCE IN LAPAROSCOPIC HYSTERECTOMY (Study I)

The first one hundred laparoscopic hysterectomies were performed by one senior gynecologist. Sixty-four percent of the procedures were carried out because of uterine fibroids and laparoscopic removal of the uterus succeeded in all cases. The weight of the uterus ranged between 70 and 470 grams and the estimated operative blood loss ranged between 50 and 1400 ml. The average hospital stay was 1.3 days and recovery time was 10.9 days. Complications occurred in 10% of cases: intraoperative or postoperative bleeding in 5%, postoperative infection in 4%, bladder injury in 1%. No ureteral, intestinal or major vessel injuries occurred. One bladder perforation and one case of intraoperative heavy bleeding of 1400 ml occurred during the first 20 operations. Vaginal cuff bleeding, inferior epigastric vessel bleeding and pelvic infections occurred throughout study period and did not decrease with experience.

The operating time varied between 45 and 245 minutes, the average being 109 minutes. The operating time shortened consistently with increasing number of operations. The learning curve shows a reduction of the mean operating time from 180 minutes during the first ten procedures to 75 minutes during the last twenty procedures. The mean operating time was reduced to half after 80 operations (Figure I). The results of regression analysis revealed positive correlation between the operating time and the size of the uterus and the weight of the patient. The operating time was unaffected by removal of the adnexa or previous laparotomies. The use of clips or staplers did not shorten the operation time compared with the use of electrocoagulation only during the first 50 operations. After the first 50 laparoscopic hysterectomies the technique was stabilized and also the operating time was stabilized by using three puncture technique, Valtchev uterus manipulator (Conkin Surgical Instruments, Toronto, Canada) and only electrocoagulation and scissors.

NATIONWIDE EXPERIENCE OF THE FIRST TWO YEARS OF LAPAROSCOPIC HYSTERECTOMY (Study II)

According to the hospital records, a total of 1216 laparoscopic hysterectomies were performed in 1993 and 1994. Data was obtained on 1165 procedures (96%) via questionnaires. Most of the physicians were at the beginning of the learning phase, having performed less than ten operations per year during the two study years (Table 6).

Table 6. Gynecologists performing laparoscopic hysterectomies in 1993-1994

Most of the operations were carried out because of uterine fibroids (54%) followed by menorrhagia (27%) and other reasons. In the whole country, the mean operating time was 132 minutes. The mean hospital stay was 3.3 days and the recovery time (written sick leave) ranged between 4 and 42 days (mean 18 days). Complications were reported in 10.2% of cases and major complications in 3.5% of cases (Table 7). Unintended surgery was carried out in 3.0% of laparoscopic hysterectomies because of a complication. The most common complication was infection, accounting for over half of the complications, but the most severe were urinary tract injuries (2.7%). Of the 32 urinary tract injuries, 17 were bladder injuries (1.5%) and 15 were ureteral injuries (1.3%). No fatal complications occurred during the study period.

Table 7. Complications of 1165 laparoscopic hysterectomies in 1993-1994

TISSUE TRAUMA AND CLINICAL OUTCOME IN LAPAROSCOPIC VERSUS ABDOMINAL HYSTERECTOMY (Study III)

Fifty women were recruited into the study, of whom 25 were operated upon laparoscopically and 25 abdominally. There were no differences in patient characteristics, indications for operation or uterine weight between the study groups. The mean operating time (85 min in LH versus 58 min in TAH) and anesthetic time (122 min in LH versus 83 min in TAH) were significantly longer in the laparoscopic group. On the other hand, there was less blood loss and a smaller decrease in blood hemoglobin concentration, and the times of hospital stay (2.1 days in LH versus 3.4 days in TAH) and sick leave (21 days in LH versus 39 days in TAH) were significantly shorter in association with the LH group. There were no significant differences in postoperative complications (24% in LH versus 28% in TAH) between the study groups.

For tissue trauma analysis 18 uncomplicated hysterectomies in both groups were included. Six women with the above-mentioned complications were excluded from the LH group and in addition, one woman with an elevated preoperative CA 125 level (216 kU/L), for no apparent reason. In the TAH group seven women with postoperative complications were excluded. The circulating concentrations of IL-6, CRP, TATI and CA 125 increased significantly in both groups. The increase in IL-6 was greatest on the first postoperative day in both groups, that of CRP on the second postoperative day in both groups, that of TATI on the seventh postoperative day in the LH group and on the second postoperative day in the TAH group and that of CA 125 on the seventh postoperative day in both groups. Serum concentrations of L-6 and CRP were significantly lower in the LH group on the first and second postoperative day, compared with the TAH group. No differences were seen in TATI and CA 125 levels between the groups (Table 8, Figure II).

Table 8. Tissue trauma markers in LH and AH
Variable Day LH (n=18) AH (n=18) P (LH vs AH)
IL-6 (pg/mL) 0 3.7 (0) 4.4 (2.8) NS
1 10.4 (8.8)* 21.6 (15.1)* 0.01
2 5.5 (3.9) 17.0 (18.4)* 0.02
7 4.4 (2.2) 4.8 (3.6) NS
28 3.7 (0) 3.7 (0) NS
CRP (mg/L) 0 1.4 (1.9) 0.7 (0.9) NS
1 12.1 (14.1)* 21.8 (12.0)* 0.03
2 26.5 (21.3)* 55.3 (26.0)* <0.001
7 10.4 (13.2)* 15.0 (15.9)* NS
28 1.8 (2.3) 1.4 (1.3) NS
TATI (ug/L) 0 5.8 (2.9) 5.7 (3.5) NS
1 5.9 (3.6) 8.2 (6.9)* NS
2 7.5 (3.4)* 15.2 (18.7)* NS
7 9.0 (5.9)* 7.7 (4.7) NS
28 7.4 (3.0)* 7.0 (3.5)* NS
CA 125 (kU/L) 0 16.1 (8.9) 15.1 (9.9) NS
1 15.6 (10.4) 11.5 (8.7) NS
2 16.4 (9.7) 13.9 (11.6) NS
7 23.1 (11.2)* 22.8 (17.6)* NS
28 17.1 (10.2) 13.1 (8.6) NS

Values are given as means (standard deviation)
* statistically significant change from preoperative value
paired two-sample Student's t-test
IL-6: LH-0 vs LH-1:P=0.005 / AH-0 vs AH-1:P=0.002 / AH-0 vs AH-2:P=0.01
CRP: LH-0 vs LH-1:P=0.002 / LH-= vs LH-2:P<0.001 / LH-0 vs LH-7:P=0.01

AH-0 vs AH-1:P<0.001 / AH-0 vs AH-2:P<0.001 / AH-0 vs AH-7:P=0.001
TATI: LH-0 vs LH-2:P=0.03 / LH-0 vs LH-7:P0.007 / LH-0 vs LH-28:P=0.002
AH-0 vs AH-1:P=0.05 / AH-0 vs AH-2:P=0.03 / AH-0 vs AH-28:P=0.02
CA 125: LH-0 vs LH-7:P=0.001 / AH-0 vs AH.7:P=0.02

Figure II. Changes in serum IL-6, CRP, TATI and CA 125 concentrations in laparoscopic (___) and abdominal( _ _ _) hysterectomy.

Values are given as means plus or minus standard errors (SE).
I: Before surgery, II: 1st, III: 2nd, IV: 7th, and V: 28th postoperative day.

URINARY TRACT INJURIES AFTER HYSTERECTOMY IN FINLAND (Study IV)

In Finland, 62,379 hysterectomies were carried out from 1990 through 1995. In 1990, 93% of hysterectomies were performed abdominally, and only 7% vaginally. In 1995, the proportion of abdominal hysterectomies had decreased to 75%, vaginal hysterectomies had increased to 11%, and 14% of hysterectomies were carried out laparoscopically.

One hundred and forty-two (0.2%) urinary tract injuries in 136 women were reported: 60 ureteral injuries (0.10%) and 82 bladder injuries (0.13%), and 52 of the 82 bladder injuries resulted in vesicovaginal fistulas (0.08%). The rate of ureteral injury was low after vaginal, supracervical abdominal and total abdominal hysterectomy, but high after laparoscopic hysterectomy, with no significant decrease from 1993 to 1995. Bladder injuries were also rare after vaginal and supracervical hysterectomy, but more common after total abdominal hysterectomy. Again, bladder injuries were most common after laparoscopic hysterectomy (Table 9). Difficulties during an operation with ureteral injury were encountered in 51%, 76%, 100% and 100% of cases and with bladder injury in 53%, 37%, 100% and 0% of cases after laparoscopic, abdominal, supracervical abdominal and vaginal hysterectomy, respectively. Urinary tract injuries greatly prolonged the recovery time. The times taken before the patients were totally recovered were 7.5 months, 6.4 months, 2.7 months and 2.0 months after ureteral injury and 4.1 months, 4.7 months, 2.0 months and 4.0 months after bladder injury following laparoscopic, total abdominal, supracervical abdominal and vaginal hysterectomy, respectively.

Ureteral injuries

Thirty-eight ureteral injuries (1.4%) occurred in 37 patients following laparoscopic hysterectomy (Table 9). Uterine vessels and part of the cardinal ligament were dissected laparoscopically in most of the laparoscopic hysterectomies. The uterine vessels were electrocoagulated in all but four cases, where clips were used. Twenty-two gynecologists were involved in these 38 ureteral injuries. In only two cases was the occurrence of the injury suspected during the primary operation. All ureteral injuries were repaired by urologists, most commonly by ureteroneocystotomy (53% of cases). The failure rate of primary repair was 5%: one patient had to undergo two operations and another had three operations before the injury was repaired.

Eighteen ureteral injuries (0.04%) occurred in 17 patients undergoing total abdominal hysterectomy (Table 9). All uterine vessels were ligated with sutures and each gynecologist encountered ureteral injury only once. None of the ureteral injuries were noticed during the primary operation and all except one were repaired by urologists (ureteroneocystostomy in 67% of cases). The failure rate of primary repair was 12%: one patient had to undergo two operations and the patient with bilateral ureteral injury had four operations.

Table 9. Urinary tract injuries associated with hysterectomy in Finland
Injury 1990 1991 1992 1993 1994 1995 Total
LH
Ureter - - 0 (0) 7 (19.1) 8 (10.0) 23 (14.1) 38 (13.9)
Bladder - - 0 (0) 5 (13.7) 10 (12.5) 3 (1.9) 18 (6.6)
Fistula - - 0 (0) 2 (5.5) 0 (0) 4 (2.6) 6 (2.2)
All - - 0 (0) 14 (38.3) 18 (22.5) 30 (19.2) 62 (22.6)
TAH
Ureter 4 (0.6) 2 (0.3) 1 (0.1) 5 (0.7) 4 (0.6) 2 (0.3) 18 (0.4)
Bladder 1 (0.1) 2 (0.3) 1 (0.1) 2 (0.3) 2 (0.3) 1 (0.1) 9 (0.2)
Fistula 9 (1.5) 9 (1.3) 8 (1.1) 6 (0.8) 5 (0.7) 8 (1.1) 45 (1.0)
All 14 (2.1) 13 (1.8) 10 (1.3) 13 (1.7) 11 (1.5) 11 (1.6) 72 (1.7)
SAH
Ureter 0 (0) 1 (0.5) 0 (0) 2 (1.1) 0 (0) 0 (0) 3 (0.3)
Bladder 0 (0) 0 (0) 1 (0.4) 1 (0.5) 1 (0.6) 0 (0) 3 (0.3)
Fistula 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0)
All 0 (0) 1 (0.5) 1 (0.4) 3 (1.6) 1 (0.6) 0 (0) 6 (0.6)
VH
Ureter 0 (0) 0 (0) 0 (0) 0 (0) 1 (0.9) 0 (0) 1 (0.2)
Bladder 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0)
Fistula 0 (0) 1 (1.2) 0 (0) 0 (0) 0 (0) 0 (0) 1 (0.2)
All 0 (0) 1 (1.2) 0 (0) 0 (0) 1 (0.9) 0 (0) 2 (0.4)
Values are given as numbers (numbers/1000 procedures)

Three ureteral injuries (0.03%) were reported following supracervical abdominal hysterectomy (Table 9). One ureteral lesion was diagnosed during the primary operation and ureteral anastomosis was performed by a surgeon. Primary healing was achieved in all three patients (ureterolysis and stenting, ureteral anastomosis and transureteroureterostomy).

Only one ureteral injury (0.02%) following vaginal hysterectomy was reported (Table 9). The injury was noticed postoperatively and repaired by ureteroneocystostomy.

Bladder injuries

Twenty-four bladder injuries (0.9%) were reported following laparoscopic hysterectomy (Table 9). Of these, 18 were simple bladder perforations repaired with sutures, without any further complications. Six bladder injuries resulted in vesicovaginal fistulas after the primary suturation of bladder injury; the incidence was 0.2%. Fifty-eight percent of bladder perforations were recognized during the primary procedure. Seventeen percent of these bladder injuries were repaired by urologists and 83% by gynecologists. Altogether, 17 simple bladder perforations were repaired with one operation and one perforation was treated only by means of a Foley catheter. Four fistulas were successfully repaired using an abdominal approach. Three of them were operated upon within 6 weeks and one after 18 weeks. One fistula was repaired first vaginally after 8 weeks and again abdominally after almost 5 months. One fistula was treated only by means of a Foley catheter. The failure rates of primary bladder and vesicovaginal fistula repair in the laparoscopy group were 5% and 17%, respectively.

Fifty-four bladder injuries (0.1%) were reported following total abdominal hysterectomy (9 simple bladder perforations and 45 (0.1%) vesicovaginal fistulas) (Table 9). Eight patients with simple bladder perforation recovered with no sequelae after suturing the defect and one perforation was handled with a Foley catheter. Thirty-two vesicovaginal fistulas were repaired in one operation, six fistulas required two operations, three fistulas required three operations and one patient had to undergo four different operations before the fistula was repaired. Three fistulas healed with catheters. The failure rates of bladder and vesicovaginal fistula repair following total abdominal hysterectomy were 18% and 20%, respectively. Twenty-four percent of fistulas were repaired before six weeks, the failure rate being 40%, and the remaining 76% were repaired after six weeks, with a failure rate of 16%. Urologists were involved in 62% of these operations.

Three bladder injuries (0.03%) occurred during supracervical abdominal hysterectomy (Table 9). Two of them were noticed and sutured during the primary operation. No fistulas occurred and all patients recovered without any further complications.

During vaginal hysterectomy, one vesicovaginal fistula (0.02%) occurred (Table 9). Urinary incontinence was detected one month after the vaginal fistula operation and there was no response to a sling procedure or to physiotherapy.

COMPLICATIONS OF LAPAROSCOPY IN FINLAND (Studies V-VI and some unpublished material)

Major complications

A total of 102,812 gynecological laparoscopies were performed from 1990 through 1996 (Table 10). During these seven years, 227 major complications were reported, at an incidence of 2.2/1000 in all laparoscopies; 0.4/1000 in diagnostic laparoscopies, 0.5/1000 in sterilization laparoscopies and 9.0/1000 in operative laparoscopies. The incidence was stable in diagnostic and sterilization laparoscopies but increased from 0/1000 in 1990 to 14.0/1000 in 1996 in operative laparoscopies (Figure III). Eighty-two percent of major complications occurred in operative laparoscopies, and 69% of these occurred in laparoscopic hysterectomy with laparoscopic hysterectomy accounting for 57% of all major complications (Tables 10 and 11). Operative laparoscopies other than hysterectomies also became more difficult during the study years and major complications increased in these procedures from 0/1000 in 1990 to 6.9/1000 in 1996 (Figure III).

Table 10. Major complicationsof laparoscopy in Finland in 1990-1996
Diagnostic Sterilization Operative Total
Procedures (n) 25.059 56.989 20.764 102.812
Intestinal 6 (0.2) 23 (0.4) 38 (1.8) 67 (0.7)
Bladder 1 (0.04) 1 (0.02) 40 (1.9) 42 (0.4)
Ureteral 0 (0) 2 (0.04) 74 (3.6) 76 (0.7)
Major vascular 1 (0.04) 2 (0.04) 8 (0.4) 11 (0.1)
Incisional hernia 0 (0) 0 (0) 8 (0.4) 8 (0.08)
Other 1 (0.04) 3 (0.05) 18 (0.9) 22 (0.2)
Death 0 (0) 0 (0) 1 (0.05) 1 (0.01)
All 9 (0.4) 31 (0.5) 187 (9.0) 227 (2.2)
Values given as numbers and numbers/1000 procedures

Table 11. Laparoscopic procedures where major complications occurred in Finland

Entry-related complications (unpublished data)

Twenty-two percent (51 of 227) of all major complications were caused by a Veress needle or a trocar, at an average incidence of 0.5/1000 procedures: 0.4/1000 in diagnostic and sterilization laparoscopies and 0.9/1000 in operative laparoscopies (2=7.37, P=0.007). The total incidence increased from 0.4/1000 in 1990 to 0.8/1000 in 1996 but the change was not statistically significant. In 1990, 83% of major complications were entry-related injuries and in 1996, only 18% (2=13.6, P <0.001). Altogether, 18% of entry-related complications occurred during diagnostic laparoscopies, 45% during sterilisation laparoscopies and 37% during operative laparoscopies. Difficulties in entering the abdominal cavity were encountered in only 10% of cases when injuries occurred, previous abdominal surgery was reported in 38% of cases and laparotomy was performed in 92% of cases to repair the injury (Table 12).

Table 12. Entry-related laparoscopic injuries in Finland in 1990-1996

Gastrointestinal injuries

The incidence of gastrointestinal injury was 0.7/1000 in all laparoscopies: lowest in diagnostic and highest in operative laparoscopies (Table 10). The incidence increased until 1994 and decreased after that in operative laparoscopies (Figure IV). Gastrointestinal injuries were the most common major complications in operative laparoscopies excluding hysterectomies (Figure V). Forty-three percent of gastrointestinal injuries were related to laparoscopic entry (unpublished data).

Damage to the small bowel was the most common gastrointestinal complication (58%), followed by damage to the large bowel (34%), rectum (5%) and stomach (3%). Only 21% of these injuries were diagnosed and managed during the primary operation. The time from injury to diagnosis was longer after perforation as a result of electrocoagulation (range 0-38 days) than after perforation with a Veress needle or trocar (range 0-5 days). All gastrointestinal injuries but one were treated by means of laparotomy; one stomach injury healed without any procedure. Sixty-six percent of injuries were primarily repaired by suturation of the bowel, followed in numbers by resection and anastomosis of the bowel, and enterostomy. Of 67 women, eleven (16%) had to undergo two operations, one woman underwent three operations and one woman six operations before the bowel injury was repaired.

Urinary tract injuries

The incidence of urinary tract injury was 1.1/1000 in all laparoscopies, again being most common in operative laparoscopies (Table 10). The risk of ureteral injury increased significantly in the 1990s in operative laparoscopies, but after subtracting the number of laparoscopic hysterectomies, the incidence rate was stable (Figures IV and V). Eighty-eight percent (67 of 76) of all ureteral injuries took place in laparoscopic hysterectomies and none of them were entry-related injuries. Bladder injuries also increased after 1992 but the incidence was stable in operative laparoscopies other than hysterectomies (Figures IV and V). Seven percent of bladder injuries were entry-related (unpublished data).

Seventy-six ureteral injuries occurred and 84% were caused by electrocoagulation, followed by injuries caused by clips, scissors, staplers and sutures. Three bilateral injuries occurred after laparoscopic hysterectomy. Ureteroneocystostomy was the most common method of repair, followed by ureteral stenting, end-to-end anastomosis, nephrectomy, pyelostomy, transureteroureterostomy and splitting the stricture by ureteroscopy. Two patients were operated on twice.

Forty-two bladder injuries were reported. Thirty of them were simple bladder perforations which were sutured with no further delay. Bladder perforations were sutured by means of laparotomy in thirteen cases, laparoscopically in twelve cases, vaginally in three cases and two patients were successfully treated by use of a Foley catheter. Twelve vesicovaginal fistulas occurred after suturation of bladder perforation and all of them occurred in laparoscopic hysterectomies. One fistula was treated with a Foley catheter, eight fistulas were repaired in one procedure and three patients had to undergo two operations before the fistula was closed.

Major vascular injuries

Major vascular injuries were rare, at an incidence of 0.1/1000 in all laparoscopies (Table 10). Half of these injuries occurred during laparoscopic entry (unpublished data) and they were also rare in laparoscopic hysterectomies. Altogether, eleven major vascular injuries were reported: nine to the iliac vessels and two to the aorta. All were sutured by way of laparotomy and all cases except one recovered without further complications. Three years after the primary operation, a fistula between the iliac artery and vein was repaired.

Incisional hernias

Only eight hernias associated with trocar incisions were reported, at an incidence of 0.08/1000 in all laparoscopies and 0.4/1000 in operative laparoscopies where secondary trocars are usually used (Table 10). Diagnosis was made between the 2nd and 14th postoperative day. The size of the trocars ranged from 5 to 12 mm (one 5 mm, five 10 mm, and two 12 mm), and 5-30 cm of small intestine were prolapsed into the hernia. In three cases bowel resection was carried out and in the others the intestine was merely repositioned.

Other injuries

The incidence of other injuries was 0.2/1000 in all laparoscopies (Table 10). Twenty-two injuries were reported: nine hemorrhages from small vessels (epigastric, mesenteric, uterine, vaginal cuff, umbilical trocar incision, and retroperitoneal small vessels), six cases of paresis of the brachial plexus or peroneal nerve, three laparotomies were carried out because of peritoneal abscesses, there were two cases of deep venous thrombosis, one case of persistently elevated human chorionic gonadotropin concentration after ectopic pregnancy, leading to laparotomy, and one case of persistent pain after laparoscopic clip sterilization, leading to supracervical abdominal hysterectomy.

Mortality

During the seven study years the mortality rate was 1/100,000, consisting of one fatal pulmonary embolism following laparoscopic supracervical hysterectomy. The patient had received no anticoagulant medication and after her death a positive family history of thromboembolism was reported.

COMPLICATIONS OF LAPAROSCOPIC HYSTERECTOMY IN FINLAND (Studies II, IV-VI and some unpublished material)

From 1992 through 1998, 10,998 laparoscopic hysterectomies were performed in Finland. The proportion of abdominal hysterectomies declined from 93% in 1990 to 45% in 1998. At the same time the proportions of laparoscopic and vaginal hysterectomies increased to 28% and 27% in 1998, respectively (Table 13). Laparoscopic hysterectomies were the main source of laparoscopic complications in the 1990s but the major complications of laparoscopic hysterectomies decreased significantly from 4.9% in 1993 to 1.4% in 1998 (Table 14).

Table 13. The number of hysterectomies in Finland
Year AH VH LH All
1990 8474 (93%) 621 (7%) 0 (0%) 9095
1991 9291 (92%) 815 (8%) 0 (0%) 10106
1992 9676 (92%) 822 (8%) 11 (0.1%) 10508
1993 9371 (87%) 1030 (10%) 366 (3%) 10767
1994 8841 (81%) 1110 (10%) 799 (9%) 10750
1995 8350 (75%) 1238 (11%) 1565 (14%) 11153
1996 5875 (58%) 1801 (18%) 2424 (24%) 10100
1997 5401 (51%) 2242 (21%) 2887 (28%) 10531
1998 4804 (45%) 2873 (27%) 2946 (28%) 10623

Table 14. Major complications of laparoscopic hysterectomy in Finland

Table 15. Ureteral injuries of laparoscopic hysterectomy in different Finnish hospitals

Altogether, 175 urinary tract injuries were reported during these seven years. One hundred and eleven ureteral injuries occurred and they accounted for 50% of all major complications in laparoscopic hysterectomies. The average risk of ureteral injury during these seven years was 1.0% and the incidence decreased from 1.9% in 1993 to 0.7% in 1998. Ureteral injuries were least common in university hospitals and most common in local hospitals (Table 15). Fifty-five percent of operating gynecologists were involved once and 31% twice in ureteral injuries. The ureteral complications were associated with ligation of uterine vessels in all but one cases. In that case, ureter was injured when suturing a large bladder perforation vaginally. In 107 cases the uterine vessels were ligated laparoscopically most often with electrocoagulation (96 cases), followed by clips (6 cases), scissors (3 cases), staplers (1 case) and an ultrasonic scalpel (1 case). The uterine vessels were ligated vaginally in only three cases when ureteral injury occurred. The injury was bilateral in five cases (5%). As a result of 111 ureteral injuries, three nephrectomies were performed but most often the injury was repaired with ureteroneocystostomy (59% of cases). Seven percent of women had to undergo several operations before ureteral injury was repaired.

Thirty bladder injuries took place during a laparoscopic part of the operation and were sutured by way of laparotomy in 17 cases, by way of laparoscopy in 12 cases and with a Foley catheter in one case. Nine bladder injuries occurred during a vaginal part of the operation and were sutured vaginally during the primary operation. Twenty vesicovaginal fistulas occurred during a laparoscopic part and five during a vaginal part of the operation. Three of them were treated with a Foley catheter and the rest needed a repair.

Of 18 intestinal injuries, seven were small bowel, five were large bowel and six were rectum perforations. Fifteen injuries were caused by electrocoagulation, two were entry-related and one was caused by scissors. Small bowel perforations were sutured in five cases and bowel resection was done in two cases. Large bowel perforations were treated by suturation (2 cases), resection (2 cases) and enterotomy (1 case). One of the three rectovaginal fistulas healed spontaneously and two required enterotomy. Three rectum perforations were treated by way of enterotomy.

Major vascular injuries (two iliac arteries) and incisional hernias were rare complications throughout the study period. Other injuries consisted of twelve hemorrhages of small vessels in abdominal wall, rectoperitoneum, vaginal cuff or uterine vessels as well as five nerve paresis, two abdominal severe infections, two deep venous thrombosis, one non fatal pulmonary embolism and one rupture of vaginal sutures. One death occurred because of a massive pulmonary embolism.