Laparoscopic Management of Bleeding After Laparoscopic or Vaginal Hysterectomy

Abstract

Objective:

To assess the results and contributions of laparoscopy in the management of postoperative bleeding following laparoscopic (LH) or vaginal hysterectomy (VH).

Methods:

A retrospective study of a 5-year period was carried out on 1167 women who underwent laparoscopic or vaginal hysterectomy. Ten women with postoperative bleeding following laparoscopic or vaginal hysterectomy were identified.

Results:

The overall incidence of bleeding after laparoscopic or vaginal hysterectomy was 0.85% (10 of 1167). Over the 5-year study period, the incidence fluctuated between 1.1% and 0.4%. Surgical revision was primarily vaginal in 1 woman, followed by laparoscopic control. In 6 patients, laparoscopy was performed immediately. The patients profited from the prompt laparoscopic treatment, because intraabdominal hemorrhage was found and stopped. Of 6 cases of intraperitoneal bleeding, 1 resulted from a blood disorder. The collagen-fibrin agent TachoComb was applied locally, and the patient was postoperatively treated with blood products and coagulation factors. Only bipolar coagulation, TachoComb, and Foley catheter were used to achieve local hemostasis during laparoscopy. The remaining 3 cases where the vaginal cuff was bleeding were managed by vaginal repair and packing without laparoscopy.

Conclusion:

The laparoscopic approach to postoperative bleeding following laparoscopic or vaginal hysterectomy is an attractive alternative to the abdominal surgical approach. Bleeding following laparoscopic or vaginal hysterectomy can be managed by laparoscopy in the majority of patients. Because the abdominal incision is avoided, the recovery time is reduced.

INTRODUCTION

Bleeding after abdominal or vaginal hysterectomy is traditionally treated by laparotomy or the transvaginal approach. Laparotomy is recommended in cases of intraperitoneal bleeding or unsuccessful conservative transvaginal treatment. Laparoscopy was successfully used for the treatment of bleeding associated with vaginal or laparoscopically assisted vaginal hysterectomy.^{1, 2} Hemostasis can be more easily obtained in laparoscopic surgery because of magnification, close inspection, routine use of suction irrigation, and bipolar coagulation.

The aim of this retrospective study was to evaluate the role of laparoscopy in the management of postoperative bleeding after vaginal or laparoscopic hysterectomy.

METHODS

The study was approved and reviewed by the Local Ethics Committee of Kladno Hospital. The 1226 patients who underwent vaginal or laparoscopic hysterectomy for benign or malignant female pelvic diseases between January 1997 and December 2001 were analyzed retrospectively. We excluded 59 patients with laparoscopic hysterectomy and lymphadenectomy performed for uterine malignancy. Ten women with postoperative bleeding following vaginal (3) or laparoscopic hysterectomy (7) were identified from the group of 1167 selected patients. We began to perform laparoscopic hysterectomies in April 1994. Five experienced surgeons were proficient in advanced laparoscopy or classic vaginal hysterectomy. The choice of the route of access for hysterectomy depended on the clinical parameters in each individual case after appropriate assessment of patients. A vaginal approach was used if the uterus was considered mobile, there was free vaginal access, and an enlarged uterus with more than 14 weeks of gestation was absent. We opted for a combination laparoscopy and vaginal surgery in the patients with a history of pelvic or abdominal surgery, patients suspected of an adhesive process after the pelvic inflammation, patients who had fixation of a retroflexed uterus, patients with a uterus size more than 14 weeks, patients with adnexal pathologies, or patients in whom difficulties in removal of adnexa were expected. The following 3 variants of laparoscopic hysterectomy (LH) were used: the laparoscopy assisted vaginal hysterectomy with vaginal colpotomy (LAVH-VC), laparoscopy assisted vaginal hysterectomy with laparoscopic colpotomy (LAVH-LC), and total laparoscopic hysterectomy (TLH).

Surgical revision for postoperative bleeding was performed transvaginally, laparoscopically, or both. Only bipolar coagulation, Foley catheter, or a collagen-fibrin agent (TachoComb) was used to achieve local hemostasis during laparoscopy. We retrospectively studied the following parameters: history of previous surgery, the time lag between hysterectomy and the occurrence of bleeding symptoms, the duration of surgery, blood loss, hospital stay, and convalescence. Statistical analysis was carried out with the Fisher exact t test. A P value <0.05 was considered significant.

RESULTS

The median age and weight of the 10 studied patients were 49.8 years (range, 32 to 72) and 67.5 kg (range, 48 to 82), respectively. The indications for vaginal or laparoscopic hysterectomy are listed in Table 1. Surgically, revision was primarily vaginal in 1 woman, followed by laparoscopic control. In this case, laparoscopy was of no further benefit because the bleeding had been stopped transvaginally. In 6 patients, laparoscopy was performed immediately because intraabdominal bleeding was suspected. Five of the 6 patients profited from the prompt laparoscopic treatment because the intraabdominal hemorrhage was found and stopped. The bleeding sources were the ovarian stump vessels in the 2 patients and the vaginal cuff in the other 2 patients. In the remaining case, the bleeding source was detected in the abdominal wall. The port-site bleeding was successfully treated with a Foley catheter only.

Table 1.

Indications for Surgery (N=1167)^*

Indications for Surgery	VH (n=355)	VH+BSO (n=63)	LH (n=137)	LH+BSO (n=612)
Benign uterine pathology	143	61	134	398
Adnexal mass	0	0	0	67
Ovarian cystadenomas	0	0	0	145
CIN III†	5	1	3	0
Uterine prolapse	197	1	0	2

^*VH = vaginal hysterectomy; BSO = bilateral salpingo-oophorectomy; LH = laparoscopic hysterectomy.

^†CIN = cervical intraepithelial neoplasia grade III.

Of 6 cases of intraabdominal bleeding, one resulted from a blood disorder (disseminated intravascular coagulation), and no source of bleeding was found in the other. At laparoscopy, we were unable to locate the source of bleeding. It was generalized oozing from the pelvic cavity. The collagen-fibrin agent TachoComb (Hafslund Nycomed Pharma AG, Linz, Austria) was applied locally to the area of bleeding, and the patient was postoperatively treated with blood products and coagulation factors. The 3 cases where the vaginal cuff was bleeding were managed by vaginal repair and packing without laparoscopic reexploration or transfusion. In 2 patients from the 3 mentioned cases, the bleeding vaginal cuff was found 2 weeks after previous total laparoscopic hysterectomy.

Table 2 shows the incidence of postoperative bleeding per year following laparoscopic or vaginal hysterectomy. The 278 patients (36.7%) who were treated laparoscopically had had previous surgery. The incidence of postoperative bleeding after LH or VH was 0.85% over a period of 5 years. During the same period, the incidence of postoperative bleeding ranged from 2/183 (1.1%) in 1997 to 1/231 (0.43%) in 2001. The frequency of bleeding was 0.93% following laparoscopic hysterectomy and 0.71% following vaginal hysterectomy. The difference was statistically insignificant (P>0.05). Eighty percent (8/10) of postoperative bleeding was diagnosed during the first 24 hours after surgery. The time lag between primary hysterectomy and revision surgery ranged from 120 minutes to 2 weeks. Median operating time for revision was 57 minutes (range, 20 to 95). Five women had relaparoscopy after LH for excessive bleeding from the Redon's drain. Six patients had to be transfused. The median hospital stay was 5.4 days (range, 3 to 18) following reoperation. The primary revision approach had influence on the postoperative course in 1 case. In this case, laparotomy was necessary for the pelvic abscess. The patient did not receive antibiotics during the laparoscopic revision after vaginal hysterectomy.

Table 2.

Postoperative Bleeding per Year Following Laparoscopic or Vaginal Hysterectomy^*

Procedure†	1997	1998	1999	2000	2001	Total
LH	19	27 (1)	39 (1)	27 (1)	25	137 (3)
LH+BSO	93 (1)	120	148 (2)	145 (1)	106	612 (4)
VH	60 (1)	62 (1)	65	83	85 (1)	355 (3)
VH+BSO	11	11	12	14	15	63
Total	183 (2)	220 (2)	264 (3)	269 (2)	231(1)	1167(10)

^*The number of bleeding complications is in parentheses.

^†LH = laparoscopic hysterectomy; BSO = bilateral salpingo-oophorectomy; VH = vaginal hysterectomy.

DISCUSSION

Peri- and early postoperative complications after hysterectomy for pelvic disorders are not rare events. Makinen et al³ reported in a large-scale observational study of 10 110 hysterectomies that the most severe type of hemorrhagic events occurred in 2.1%, 3.1%, and 2.7% in the abdominal, vaginal, and laparoscopic group, respectively. Some authors suppose that the postoperative bleeding occurs more frequently after abdominal or laparoscopic than after vaginal hysterectomy.^{4, 5} The incidence of bleeding requiring reoperation after a hysterectomy varies from 0.2% to 2.0%.^2,6–9

The frequency of major complications from laparoscopic surgery is reported to be 0.6% to 2.5%.¹⁰ Hemorrhage accounts for half of the complications, ranging from persistent venous oozing to massive blood loss from injury to retroperitoneal vessels.¹¹ Delayed hemorrhagic complications also occur; the true frequency is unknown, but the consequences can be particularly significant in patients undergoing outpatient surgery.¹² Possible reasons for delayed hemorrhage are a bleeding vessel missed at the end of the procedure, effects of pneumoperitoneum, Trendelenburg position, low intraoperative pressure, wearing off the effect of vasopressin, subacute infection, postoperative analgesia, and bleeding disorders.¹²

Nezhat and colleagues¹³ reported 17 complications involving postoperative bleeding. Only 3 patients required laparotomy. Location of the bleeding included 9 from the anterior abdominal wall (7 from the suprapubic incision and 2 from the umbilical incision), 3 from the vaginal cuff after laparoscopic and laparoscopic-assisted vaginal hysterectomy, and 5 cases of intraabdominal bleeding. Postoperative hematomas were frequent with early use of the Multifire Endo GIA (US Surgical Corporation, Nowalk, CT) for the upper uterine pedicle during hysterectomy and oophorectomy.¹⁴ We had no experience of postoperative hematoma. This fact can be related to Redon's drain, which was routinely introduced during closure of the vaginal cuff in our patients with laparoscopic hysterectomy. Abdominal wall vessel injury occurs with increasing frequency, as the practice of laparoscopic surgery becomes wider and trocars become sharper. Aharoni et al¹⁵ recommend that in most cases a Foley catheter should be used to stop bleeding of the abdominal wall. Wilke et al² illustrated that laparoscopic surgery in postoperative bleeding after hysterectomy is feasible and may be recommended if the source of bleeding cannot be clearly identified by vaginal examination.

Our study results demonstrate that hemorrhage after laparoscopic or vaginal hysterectomy can be successfully arrested and controlled using a laparoscope for bipolar coagulation, or to introduce a Foley catheter in the port-site bleeding, or to apply a fibrin glue agent. The TachoComb was successfully applied in 1 case of postoperative disseminate intravascular coagulopathy. Ochiai et al¹⁶ evaluated the utility of TO-193 (TachoComb), a new sheet form of fibrin adhesive, in gynecologic open surgery. In their surgeries, 45 patients were treated with the combined agent during obstetric and gynecologic operations to stop oozing hemorrhages after primary hemostatic treatment, and the efficacy rate was 97.8%. Holub and Kliment¹⁷ reported successful treatment of hemorrhage from damaged tissue near important pelvic structures using the laparoscope to apply collagen fleece combined with fibrin glue. At present, 5 patients have been effectively treated using electrosurgery during laparoscopic surgery. The 0.43% to 1.13% frequency of posthysterectomy bleeding in our study is in accordance with rates reported in the recent literature.^{2, 9, 18} The difference in postoperative bleeding between laparoscopic and vaginal hysterectomy was statistically insignificant (P>0.05).

CONCLUSION

Our study illustrates that the laparoscopic approach to postoperative bleeding after laparoscopic or vaginal hysterectomy is feasible and may be recommended for both surgical procedures. However, we think that the selection of patients for laparoscopy with vaginal bleeding only should be performed with reference to optimal usefulness and safety. The prophylactic antibiotics should be administered intraoperatively. We support the idea reported by Sagiv et al¹⁹ that with the increased expertise of laparoscopists, length of surgery as an argument against laparoscopy in hemodynamically unstable women can be challenged successfully.