Angiographic embolization in the management of hemorrhagic complications of pregnancy William M. Gilbert, MD: Thomas R. Moore, MD: Robert Resnik, MD: John Doemeny, MD; Homer Chin, MD: and Joseph J. Bookstein, MDc San Diego, California Obstetric hemorrhage continues to be a major cause of maternal mortality and morbidity. Recent developments in percutaneous angiographic embolization techniques have afforded the ability to control persistent bleeding from pelvic vessels while avoiding the morbidity of surgical exploration. We report the use of angiographic embolization in 10 cases of pregnancy-related hemorrhage, including persistent postcesarean bleeding (three cases), vaginal wall hematomas (four cases), cervical ectopic pregnancies (two cases), and postpartum bleeding as a secondary complication of uterine myomas (one case). The embolization procedures were successful in all cases. Nine of 10 patients experienced postprocedural fever with eight cases resolving with antibiotic therapy alone and one patient requiring vaginal drainage of the hematoma-abscess. The mean length of time for the procedure was 167 minutes (range 70 to 270). The average length of hospitalization was 8 days (range 2 to 13). These data indicate that angiographic embolization is effective in treating hemorrhagic complications of pregnancy in hemodynamically stable patients and is preferable to surgery in selected cases. (AM J OSSTET GVNECOL 1992;166:493-7.)
Key words: Angiographic embolization, hemorrhage, pregnancy Obstetric hemorrhage remains one of the three main causes of maternal death in spite of modern blood banks, anesthesia, and pharmacologic agents to treat such complications. Effective management requires prompt identification of the cause and usually responds to one of several targeted conservative modalities, such as oxytoxics, blood component therapy, and careful surgical technique. However, persistent bleeding may occur after the evacuation of a vaginal hematoma or cesarean hysterectomy and may also result from postpartum uterine atony associated with myomas. Cervical ectopic pregnancies, although rare, represent a particularly challenging problem of hemostasis, particularly if unrecognized before curettage. Hypogastric artery ligation has traditionally been the surgical approach when exploration and arterial ligation prove unsuccessful. With the development of radiographic arterial embolization techniques, an alternative to exploratory surgery or hypogastric artery ligation has become available. The use of angiographic embolization to control hemorrhage has been well described in patients with gynecologic cancer and postpartum vaginal wall hematomas.'·6 However, in spite of From the Division of Maternal Fetal Medicine, Department of Reproductive Medicine: and the Department of Radiology,' University of California at San Diego, and the Department of Radiology,' Mercy Hospital. Received for publication April 10, 1991; revised June 6, 1991; accepted June 26,1991. Reprint requests: William M. Gilbert, MD, 225 Dickinson St., H813, San Diego, CA 92103. 611 132019
the availability of this technique in many hospitals throughout the United States, it is not widely used on most obstetric services. In this study we report our experience with angiographic embolization in various clinical hemorrhagic settings to emphasize its use as compared with hypogastric artery ligation.
Material and methods Between July 1984 and December 1990, 10 patients at the University of California at San Diego Medical Center and Mercy Hospital and Medical Center required angiographic embolization to control obstetric hemorrhage. All hospital records were reviewed and detailed to collect pertinent clinical data pertaining to the specific leading complication, clinical status, blood replacement requirements, length of embolization procedure, and complications. Two of these cases (7 and 8, Table I) have been reported previously.' Angiographic consultation and intensive care facilities are available on a 24-hour basis at both of these hospitals. Several patients were referred to our institutions from outside hospitals where these embolization techniques were not available. All patients with hemorrhagic complications unresponsive to conservative management were stabilized and taken to the radiology suite for the angiogram and embolization. By the percutaneous approach with the patient under local anesthesia, an aortogram was performed for identification of the pelvic arterial anatomy and bleeding sites. A variety of catheter shapes can then be used to enable selective catheterization of the bleeding artery, but one
494 Gilbert et al.
February 1992 Am J Obstet Gyneco1
Fig. 1. A, Preembolization arteriogram demonstrates free bleeding in area of vaginal cuff after cesarean hysterectomy for uterine atony (case 6). This free bleeding is described as "blush" (arrows) and demonstrates significant bleeding. B, Postembolization film demonstrates complete blockage of hypogastric artery on right after the Gelfoam injection.
Fig. 2. A, Preembolization angiogram demonstrates outline of vaginal hematoma (arrows) from case 10. B, Postembolization film demonstrates complete blockage of left vaginal branch and pudendal branch of hypogastric artery.
of us (j.J.B.) strongly recommends the long-reversedcurved catheter of Cordis (Cordis Inc., Miami). Subselective embolization of the bleeding vessel was then performed, usually with small pieces (2 x 2 x 2 mm) of Gelfoam (Upjohn, Co., Kalamazoo), through the catheter. Gelfoam is a gelatin sponge that is slowly absorbed and ultimately enables recanalization of the embolized vessel. Hemostasis was documented by fluoroscopic monitoring. If bleeding continued, further embolization was performed until the bleeding stopped. The catheter was then switched to the opposite hypo-
gastric artery and the same procedure as above was performed to provide hemostasis if needed. A final aortogram was performed to document overall hemostasis and to exclude the possibility of continued bleeding from collateral sources. If necessary, coli aterals themselves were then embolized. Results
A summary of the 10 patients requiring embolization is shown in Table I. The average length of the angiographic procedure was 167 minutes (range 70 to 270).
Angiographic embolization in pregnancy
Volume 166 Number 2
495
Table I. Ten cases of angiographic embolization Case No.
Gestational age (wh)
7
Diagnosis
Length of procedure (min)
Hematocrit (%)
Blood
Complications
Fever, disseminated intravascular coagulation Fever, disseminated intravascular coagulation Fibroids, pregnancy-induced hypertension Fever, disseminated intravascular coagulation Fever
2S.3
Cesarean hysterectomy for atony
21
Cervical ectopic
155
9 units PRBC
2
40
Uterine atony
165
6 units PRBC
3
40
Uterine atony
150
Refused
4
12
Cervical ectopic
70
2 units PRBC
5
3S
Vaginal hematoma
165
7 units PRBC
6
35
Vaginal bleeder
240
IS units PRBC
Bleeding
210
Persistent vaginal bleeding Fever, pelvic cellulitis
IS.1
22.5
29.3
19.4
7
40
Vaginal hematoma
120
Refused
S
40
Vaginal hematoma
270
33 units PRBC
IS
9
39
Bleeding Uterine atony
Fever, liver dysfunction shock, dissemina ted intravascular coagulation
??
Refused
Fever, chronic pelvic pain
15
10
39
Vaginal hematoma
90
4 units PRBC
Fever, drainage of hematoma
21.6
13
Procedure
Bilateral hypogastric artery embolization Bilateral hypogastric artery embolization Bilateral hypogastric artery embolization Bilateral hypogastric artery embolization Bilateral hypogastric artery embolization Bilateral hypogastric artery embolization Right ovarian Right hypogastric artery embolization Bilateral hypogastric artery embolization Right ovarian Bilateral hypogastric artery embolization Left hypogastric artery embolization
Length of stay (days)
S
S
3
2
6
12
13
11
6
11
PRBC, Packed red blood cells. Cases 6 and S had two angiographic embolizations performed on separate days.
The time of the procedure is recorded from that time the patient was sent to the angiographic unit until her return to the obstetric or intensive care unit. However, the time required to control bleeding usually is much shorter (but less well documented) and was only 25 minutes from start of procedure to control of bleeding in case 10. The primary complication included postprocedural fever in nine of 10 cases, eight of nine responding to antibiotics alone, whereas in case 7 vaginal drainage of a paravaginal abscess cavity was required 8 days after radiographic embolization. Seven of 10 patients required blood transfusions before or after the procedure, and the other three patients refused blood products (cases 3, 7, and 9).
In case 6 the patient had undergone a cesarean hysterectomy because of uterine atony associated with a triplet delivery. The estimated blood loss during and after surgery was 3000 ml, and the patient was transfused with 18 units of packed red blood cells and multiple units of fresh-frozen plasma. The patient continued to experience vaginal bleeding after surgery and was taken to the radiology suite where angiography revealed bleeding in the area of the vaginal cuff, as demonstrated by the dye "blush" outlined by the arrows in Fig. 1, A. Bilateral hypogastric artery embolization was accomplished (Fig. 1, B), and adequate hemostasis was felt to have been obtained. However, the patient continued to have bleeding from drains and underwent
496
Gilbert et al.
a second angiogram the following day to determine whether the ovarian arteries could be a possible source for the continued bleeding. Embolization of the right ovarian artery was successful in obtaining hemostasis. In case 3 the patient underwent a normal vaginal delivery and was found to have continued bleeding in spite of a firm, well contracted uterus and postpartum curettage. The patient was known to have myomas. myomas. She underwent angiograms which revealed uterine bleeding. Embolization of the anterior division of both hypogastric arteries was successful in stopping the hemorrhage. The patient described in case 10 was transferred after a traumatic forceps delivery, which resulted in an expanding left vaginal hematoma. hematoma. In Fig. Fig. 2, A, the hematoma is outlined by the arrows. Embolization was directed into the left hypogastric artery, and follow-up x-ray films (Fig. 2, B) reveal that the bleeding has stopped. Two days later the patient was taken to the operating room, where the hematoma was uneventfully drained. The patient did quite well and went home on the eleventh hospital day. Case 8 involved a patient who presented with signs of severe pregnancy-induced hypertension and possible fatty liver of pregnancy, including liver failure and disseminated intravascular coagulation. After a forceps delivery, an expanding hematoma that was noted on the left vaginal wall did not respond to conservative exploration and packing. The patient was subsequently transferred to our institution where additional conservative treatment, including two additional attempts at vaginal packing, was unsuccessful. The patient then underwent embolization of the left and right hypogastric arteries, which initially controlled the bleeding. However, because of recurrent hemorrhage into the paravaginal space and the question of a retroperitoneal hematoma, she had a second procedure that embolized the remaining vessels of the left hypogastric artery and right ovarian artery.
Comment These cases illustrate the role of angiographic embolization as an alternative to surgery for the treatment of obstetric hemorrhage from a variety of causes. Angiographic embolization prevented the need for a general anesthetic, allowed the preservation of reproductive capability, and prevented reexploration in those patients who had undergone surgery. The complications related to the procedure were minimal and included postprocedural fever, fever, which responded to antibiotic therapy, and transient buttock pain, which resolved over time. Angiographic embolization has been reported effective at stopping arterial hemorrhage in >90% of cases of gynecologic hemorrhage because of
February 1992 Am J Obstet Gynecol
the ability to place the embolic material within the hypogastric artery.7. 8 In our cases, success was achieved after the first attempt at embolization in eight of 10 cases, whereas an additional procedure to produce hemostasis was necessary in two cases. Hypogastric artery ligation is often recommended as the primary conservative surgical procedure to control hemorrhage in similar complications. The primary goals of hypogastric artery ligation are to preserve reproductive function and to offer less morbidity than more definitive surgery. Clark et al.,9 al.,9 however, found that hypogastric artery ligation was successful in only eight of 19 (42%) patients with obstetric hemorrhage. Bleeding failed to stop in the remaining 11 II patients, and hysterectomy was required. 9 Furthermore, those patients in that series who underwent hysterectomy as the primary procedure had substantially less morbidity than those who underwent hypogastric artery ligation and hysterectomy. Hypogastric artery ligation decreases arterial" pulse pressure by 85 85% % distal to the ligation but decreases blood flow in these distal vessels by only 48%.10 The extensive collateral circulation in the pelvis maintains a significant blood flow in spite of the ligation. This may well account for the low success rate of hypogastric ligation in providing adequate hemostasis. Embolization, however, enables specific obstruction of the bleeding vessel and produces obstruction beyond entry of large collaterals. Angiographic embolization must be considered before hypogastric artery ligation, because once the artery is ligated, embolization cannot be performed. Obstetric hemorrhage can often be extremely rapid, with exsanguination and death occurring within minutes. In these emergency cases the patient must be stabilized and the bleeding stopped by the most rapid and complete method possible. possible. If conservative methods such as applied pressure or packing are sufficient to provide temporary hemostasis, the patient may be stabilized with intravenous fluids, blood replacement, and clotting factors, if indicated. Resuscitation status is monitored by measuring the arterial blood gas and base excess, hematocrit, and urine output. If a coagulopathy is present, it should be corrected before or during embolization. The angiographic team should be consulted early in the course of a hemorrhagic complication to allow time for the team to arrive and prepare to treat the patient. patient. Complications associated with angiographic embolization include embolization of an inappropriate vessel or ischemia distal to the embolization. These complications are rare because of the rich collateral blood supply in the pelvis. In our cases fever occurred in nine of 10 cases. All but one responded to antibiotic therapy alone, whereas one case required vaginal drainage of an abscess 8 days after the embolization. Angiographic embolization embolization should also be con sid-
Angiographic embolization in pregnancy
Volume 166 Number 2
ered in cervical ectopic pregnancy or if bleeding ensues when this complication has gone unrecognized. Under optimal circumstances, consideration should be given first to the use of new treatment modalities, such as methotrexate, to treat such ectopic gestation. I I. 12 Our experience demonstrates that angiographic embolization is a safe, effective, and relatively rapid procedure that should be considered early in the course of obstetric hemorrhage that is unresponsive to conservative management. If the patient is hemodynamically stable, the procedure should be performed before hypogastric artery ligation or hysterectomy.
REFERENCES 1. Chin HG, Scott DR, Resnik R, Davis GB, Lurie AL. Angiographic embolization of intractable puerperal hematomas. AM J OBSTET GVNECOL 1989; 160:434-8. 2. Heffner LJ, Mennuti MT, Rudoff JC, McLean GK. Primary management of postpartum vulvovaginal hematomas by angiographic embolization. Am J Perinatol 1985;2:204-7. 3. Smith DC, Wyatt JF. Embolization of the hypogastric arteries in the control of massive vaginal hemorrhage. Obstet GynecoI1977;49:317-22.
4. Rosenthal DM, Cola pinto R. Angiographic arterial embolization in the management of postoperative vaginal hemorrhage. AM J ORTHOD DENTOFAC ORTHOP 1985;151:227-31. 5. Pais SO, Glickman M, Schwartz P, Pingoud E, Berkowitz R. Embolization of pelvic arteries for control of postpartum hemorrhage. Obstet Gynecol 1980;55:754-8. 6. Greenwood LH, Glickman MG, Schwartz PE, Morse SS, Denny DF. Obstetric and nonmalignant gynecologic bleeding: treatment with angiographic embolization. Radiology 1987;164:155-9. 7. Brown BJ, Heaston DK, Poulson AM, Gabert HA, Mineau DE, Miller FJ. Uncontrollable postpartum bleeding: a new approach to hemostasis through angiographic arterial embolization. Obstet Gynecol 1979;54:361-8. 8. J ander HP, Russinovich NAE. Transcatheter Gelfoam embolization in abdominal, retroperitoneal, and pelvic hemorrhage. Radiology 1980; 136:337-44. 9. Clark SL, Phelan JP, Yeh SY, Bruce SR, Paul RH. Hypogastric artery ligation for obstetric hemorrhage. Obstet Gynecol 1985;66:353-6. 10. Burchell RC. Physiology of internal iliac artery ligation. J Obstet Gynaecol Br Commonw 1968;75:642. 11. Kaplan BR, Brandt T, Javaheri G, Scommegna A. Nonsurgical treatment of a viable cervical pregnancy with intra-amniotic methotrexate. Fertil Steril 1990;53:941-3. 12. Palti Z, Rosenn B, Goshen R, Ben-Chitrit A, Yagel S. Successful treatment of a viable cervical pregnancy with methotrexate. AM J OBSTET GVNECOL 1989; 161: 1147-8.
Breech extraction of low-birth-weight second twins: Can cesarean section be justified? Lisa Davison, MD,. Thomas R. Easterling, MD," J. Craig Jackson, MD,b and Thomas J. Benedetti, MD"
Seattle, Washington The outcomes of 54 breech-extracted second twins weighing between 750 and 2000 gm were compared with the outcomes of their siblings and of 43 sets of twins delivered by cesarean section for malpresentation. Vaginally delivered first twins had fewer days of mechanical ventilation and oxygen therapy than their breech-extracted siblings (p = 0.004). There were no significant differences in any measures of neonatal outcome when breech-extracted twins were compared with second twins delivered by cesarean section. We conclude that routine cesarean section is not justified for nonvertex second twins expected to weigh
Key words: Angiographic embolization, hemorrhage, pregnancy Obstetric hemorrhage remains one of the three main causes of maternal death in spite of modern blood banks, anesthesia, and pharmacologic agents to treat such complications. Effective management requires prompt identification of the cause and usually responds to one of several targeted conservative modalities, such as oxytoxics, blood component therapy, and careful surgical technique. However, persistent bleeding may occur after the evacuation of a vaginal hematoma or cesarean hysterectomy and may also result from postpartum uterine atony associated with myomas. Cervical ectopic pregnancies, although rare, represent a particularly challenging problem of hemostasis, particularly if unrecognized before curettage. Hypogastric artery ligation has traditionally been the surgical approach when exploration and arterial ligation prove unsuccessful. With the development of radiographic arterial embolization techniques, an alternative to exploratory surgery or hypogastric artery ligation has become available. The use of angiographic embolization to control hemorrhage has been well described in patients with gynecologic cancer and postpartum vaginal wall hematomas.'·6 However, in spite of From the Division of Maternal Fetal Medicine, Department of Reproductive Medicine: and the Department of Radiology,' University of California at San Diego, and the Department of Radiology,' Mercy Hospital. Received for publication April 10, 1991; revised June 6, 1991; accepted June 26,1991. Reprint requests: William M. Gilbert, MD, 225 Dickinson St., H813, San Diego, CA 92103. 611 132019
the availability of this technique in many hospitals throughout the United States, it is not widely used on most obstetric services. In this study we report our experience with angiographic embolization in various clinical hemorrhagic settings to emphasize its use as compared with hypogastric artery ligation.
Material and methods Between July 1984 and December 1990, 10 patients at the University of California at San Diego Medical Center and Mercy Hospital and Medical Center required angiographic embolization to control obstetric hemorrhage. All hospital records were reviewed and detailed to collect pertinent clinical data pertaining to the specific leading complication, clinical status, blood replacement requirements, length of embolization procedure, and complications. Two of these cases (7 and 8, Table I) have been reported previously.' Angiographic consultation and intensive care facilities are available on a 24-hour basis at both of these hospitals. Several patients were referred to our institutions from outside hospitals where these embolization techniques were not available. All patients with hemorrhagic complications unresponsive to conservative management were stabilized and taken to the radiology suite for the angiogram and embolization. By the percutaneous approach with the patient under local anesthesia, an aortogram was performed for identification of the pelvic arterial anatomy and bleeding sites. A variety of catheter shapes can then be used to enable selective catheterization of the bleeding artery, but one
494 Gilbert et al.
February 1992 Am J Obstet Gyneco1
Fig. 1. A, Preembolization arteriogram demonstrates free bleeding in area of vaginal cuff after cesarean hysterectomy for uterine atony (case 6). This free bleeding is described as "blush" (arrows) and demonstrates significant bleeding. B, Postembolization film demonstrates complete blockage of hypogastric artery on right after the Gelfoam injection.
Fig. 2. A, Preembolization angiogram demonstrates outline of vaginal hematoma (arrows) from case 10. B, Postembolization film demonstrates complete blockage of left vaginal branch and pudendal branch of hypogastric artery.
of us (j.J.B.) strongly recommends the long-reversedcurved catheter of Cordis (Cordis Inc., Miami). Subselective embolization of the bleeding vessel was then performed, usually with small pieces (2 x 2 x 2 mm) of Gelfoam (Upjohn, Co., Kalamazoo), through the catheter. Gelfoam is a gelatin sponge that is slowly absorbed and ultimately enables recanalization of the embolized vessel. Hemostasis was documented by fluoroscopic monitoring. If bleeding continued, further embolization was performed until the bleeding stopped. The catheter was then switched to the opposite hypo-
gastric artery and the same procedure as above was performed to provide hemostasis if needed. A final aortogram was performed to document overall hemostasis and to exclude the possibility of continued bleeding from collateral sources. If necessary, coli aterals themselves were then embolized. Results
A summary of the 10 patients requiring embolization is shown in Table I. The average length of the angiographic procedure was 167 minutes (range 70 to 270).
Angiographic embolization in pregnancy
Volume 166 Number 2
495
Table I. Ten cases of angiographic embolization Case No.
Gestational age (wh)
7
Diagnosis
Length of procedure (min)
Hematocrit (%)
Blood
Complications
Fever, disseminated intravascular coagulation Fever, disseminated intravascular coagulation Fibroids, pregnancy-induced hypertension Fever, disseminated intravascular coagulation Fever
2S.3
Cesarean hysterectomy for atony
21
Cervical ectopic
155
9 units PRBC
2
40
Uterine atony
165
6 units PRBC
3
40
Uterine atony
150
Refused
4
12
Cervical ectopic
70
2 units PRBC
5
3S
Vaginal hematoma
165
7 units PRBC
6
35
Vaginal bleeder
240
IS units PRBC
Bleeding
210
Persistent vaginal bleeding Fever, pelvic cellulitis
IS.1
22.5
29.3
19.4
7
40
Vaginal hematoma
120
Refused
S
40
Vaginal hematoma
270
33 units PRBC
IS
9
39
Bleeding Uterine atony
Fever, liver dysfunction shock, dissemina ted intravascular coagulation
??
Refused
Fever, chronic pelvic pain
15
10
39
Vaginal hematoma
90
4 units PRBC
Fever, drainage of hematoma
21.6
13
Procedure
Bilateral hypogastric artery embolization Bilateral hypogastric artery embolization Bilateral hypogastric artery embolization Bilateral hypogastric artery embolization Bilateral hypogastric artery embolization Bilateral hypogastric artery embolization Right ovarian Right hypogastric artery embolization Bilateral hypogastric artery embolization Right ovarian Bilateral hypogastric artery embolization Left hypogastric artery embolization
Length of stay (days)
S
S
3
2
6
12
13
11
6
11
PRBC, Packed red blood cells. Cases 6 and S had two angiographic embolizations performed on separate days.
The time of the procedure is recorded from that time the patient was sent to the angiographic unit until her return to the obstetric or intensive care unit. However, the time required to control bleeding usually is much shorter (but less well documented) and was only 25 minutes from start of procedure to control of bleeding in case 10. The primary complication included postprocedural fever in nine of 10 cases, eight of nine responding to antibiotics alone, whereas in case 7 vaginal drainage of a paravaginal abscess cavity was required 8 days after radiographic embolization. Seven of 10 patients required blood transfusions before or after the procedure, and the other three patients refused blood products (cases 3, 7, and 9).
In case 6 the patient had undergone a cesarean hysterectomy because of uterine atony associated with a triplet delivery. The estimated blood loss during and after surgery was 3000 ml, and the patient was transfused with 18 units of packed red blood cells and multiple units of fresh-frozen plasma. The patient continued to experience vaginal bleeding after surgery and was taken to the radiology suite where angiography revealed bleeding in the area of the vaginal cuff, as demonstrated by the dye "blush" outlined by the arrows in Fig. 1, A. Bilateral hypogastric artery embolization was accomplished (Fig. 1, B), and adequate hemostasis was felt to have been obtained. However, the patient continued to have bleeding from drains and underwent
496
Gilbert et al.
a second angiogram the following day to determine whether the ovarian arteries could be a possible source for the continued bleeding. Embolization of the right ovarian artery was successful in obtaining hemostasis. In case 3 the patient underwent a normal vaginal delivery and was found to have continued bleeding in spite of a firm, well contracted uterus and postpartum curettage. The patient was known to have myomas. myomas. She underwent angiograms which revealed uterine bleeding. Embolization of the anterior division of both hypogastric arteries was successful in stopping the hemorrhage. The patient described in case 10 was transferred after a traumatic forceps delivery, which resulted in an expanding left vaginal hematoma. hematoma. In Fig. Fig. 2, A, the hematoma is outlined by the arrows. Embolization was directed into the left hypogastric artery, and follow-up x-ray films (Fig. 2, B) reveal that the bleeding has stopped. Two days later the patient was taken to the operating room, where the hematoma was uneventfully drained. The patient did quite well and went home on the eleventh hospital day. Case 8 involved a patient who presented with signs of severe pregnancy-induced hypertension and possible fatty liver of pregnancy, including liver failure and disseminated intravascular coagulation. After a forceps delivery, an expanding hematoma that was noted on the left vaginal wall did not respond to conservative exploration and packing. The patient was subsequently transferred to our institution where additional conservative treatment, including two additional attempts at vaginal packing, was unsuccessful. The patient then underwent embolization of the left and right hypogastric arteries, which initially controlled the bleeding. However, because of recurrent hemorrhage into the paravaginal space and the question of a retroperitoneal hematoma, she had a second procedure that embolized the remaining vessels of the left hypogastric artery and right ovarian artery.
Comment These cases illustrate the role of angiographic embolization as an alternative to surgery for the treatment of obstetric hemorrhage from a variety of causes. Angiographic embolization prevented the need for a general anesthetic, allowed the preservation of reproductive capability, and prevented reexploration in those patients who had undergone surgery. The complications related to the procedure were minimal and included postprocedural fever, fever, which responded to antibiotic therapy, and transient buttock pain, which resolved over time. Angiographic embolization has been reported effective at stopping arterial hemorrhage in >90% of cases of gynecologic hemorrhage because of
February 1992 Am J Obstet Gynecol
the ability to place the embolic material within the hypogastric artery.7. 8 In our cases, success was achieved after the first attempt at embolization in eight of 10 cases, whereas an additional procedure to produce hemostasis was necessary in two cases. Hypogastric artery ligation is often recommended as the primary conservative surgical procedure to control hemorrhage in similar complications. The primary goals of hypogastric artery ligation are to preserve reproductive function and to offer less morbidity than more definitive surgery. Clark et al.,9 al.,9 however, found that hypogastric artery ligation was successful in only eight of 19 (42%) patients with obstetric hemorrhage. Bleeding failed to stop in the remaining 11 II patients, and hysterectomy was required. 9 Furthermore, those patients in that series who underwent hysterectomy as the primary procedure had substantially less morbidity than those who underwent hypogastric artery ligation and hysterectomy. Hypogastric artery ligation decreases arterial" pulse pressure by 85 85% % distal to the ligation but decreases blood flow in these distal vessels by only 48%.10 The extensive collateral circulation in the pelvis maintains a significant blood flow in spite of the ligation. This may well account for the low success rate of hypogastric ligation in providing adequate hemostasis. Embolization, however, enables specific obstruction of the bleeding vessel and produces obstruction beyond entry of large collaterals. Angiographic embolization must be considered before hypogastric artery ligation, because once the artery is ligated, embolization cannot be performed. Obstetric hemorrhage can often be extremely rapid, with exsanguination and death occurring within minutes. In these emergency cases the patient must be stabilized and the bleeding stopped by the most rapid and complete method possible. possible. If conservative methods such as applied pressure or packing are sufficient to provide temporary hemostasis, the patient may be stabilized with intravenous fluids, blood replacement, and clotting factors, if indicated. Resuscitation status is monitored by measuring the arterial blood gas and base excess, hematocrit, and urine output. If a coagulopathy is present, it should be corrected before or during embolization. The angiographic team should be consulted early in the course of a hemorrhagic complication to allow time for the team to arrive and prepare to treat the patient. patient. Complications associated with angiographic embolization include embolization of an inappropriate vessel or ischemia distal to the embolization. These complications are rare because of the rich collateral blood supply in the pelvis. In our cases fever occurred in nine of 10 cases. All but one responded to antibiotic therapy alone, whereas one case required vaginal drainage of an abscess 8 days after the embolization. Angiographic embolization embolization should also be con sid-
Angiographic embolization in pregnancy
Volume 166 Number 2
ered in cervical ectopic pregnancy or if bleeding ensues when this complication has gone unrecognized. Under optimal circumstances, consideration should be given first to the use of new treatment modalities, such as methotrexate, to treat such ectopic gestation. I I. 12 Our experience demonstrates that angiographic embolization is a safe, effective, and relatively rapid procedure that should be considered early in the course of obstetric hemorrhage that is unresponsive to conservative management. If the patient is hemodynamically stable, the procedure should be performed before hypogastric artery ligation or hysterectomy.
REFERENCES 1. Chin HG, Scott DR, Resnik R, Davis GB, Lurie AL. Angiographic embolization of intractable puerperal hematomas. AM J OBSTET GVNECOL 1989; 160:434-8. 2. Heffner LJ, Mennuti MT, Rudoff JC, McLean GK. Primary management of postpartum vulvovaginal hematomas by angiographic embolization. Am J Perinatol 1985;2:204-7. 3. Smith DC, Wyatt JF. Embolization of the hypogastric arteries in the control of massive vaginal hemorrhage. Obstet GynecoI1977;49:317-22.
4. Rosenthal DM, Cola pinto R. Angiographic arterial embolization in the management of postoperative vaginal hemorrhage. AM J ORTHOD DENTOFAC ORTHOP 1985;151:227-31. 5. Pais SO, Glickman M, Schwartz P, Pingoud E, Berkowitz R. Embolization of pelvic arteries for control of postpartum hemorrhage. Obstet Gynecol 1980;55:754-8. 6. Greenwood LH, Glickman MG, Schwartz PE, Morse SS, Denny DF. Obstetric and nonmalignant gynecologic bleeding: treatment with angiographic embolization. Radiology 1987;164:155-9. 7. Brown BJ, Heaston DK, Poulson AM, Gabert HA, Mineau DE, Miller FJ. Uncontrollable postpartum bleeding: a new approach to hemostasis through angiographic arterial embolization. Obstet Gynecol 1979;54:361-8. 8. J ander HP, Russinovich NAE. Transcatheter Gelfoam embolization in abdominal, retroperitoneal, and pelvic hemorrhage. Radiology 1980; 136:337-44. 9. Clark SL, Phelan JP, Yeh SY, Bruce SR, Paul RH. Hypogastric artery ligation for obstetric hemorrhage. Obstet Gynecol 1985;66:353-6. 10. Burchell RC. Physiology of internal iliac artery ligation. J Obstet Gynaecol Br Commonw 1968;75:642. 11. Kaplan BR, Brandt T, Javaheri G, Scommegna A. Nonsurgical treatment of a viable cervical pregnancy with intra-amniotic methotrexate. Fertil Steril 1990;53:941-3. 12. Palti Z, Rosenn B, Goshen R, Ben-Chitrit A, Yagel S. Successful treatment of a viable cervical pregnancy with methotrexate. AM J OBSTET GVNECOL 1989; 161: 1147-8.
Breech extraction of low-birth-weight second twins: Can cesarean section be justified? Lisa Davison, MD,. Thomas R. Easterling, MD," J. Craig Jackson, MD,b and Thomas J. Benedetti, MD"
Seattle, Washington The outcomes of 54 breech-extracted second twins weighing between 750 and 2000 gm were compared with the outcomes of their siblings and of 43 sets of twins delivered by cesarean section for malpresentation. Vaginally delivered first twins had fewer days of mechanical ventilation and oxygen therapy than their breech-extracted siblings (p = 0.004). There were no significant differences in any measures of neonatal outcome when breech-extracted twins were compared with second twins delivered by cesarean section. We conclude that routine cesarean section is not justified for nonvertex second twins expected to weigh
