FERTILITY AND STERILITY

Vol. 58, No.6, December 1992

Printed on acid-free paper in U.S.A.

Copyright" 1992 The American Fertility Society

Transcervical cannulation of the fallopian tube for the management of ectopic pregnancy: prospective multicenter study

Francisco Risquez, M.D.*t:t: Robert Forman, M.D.§ Fred Maleika, M.D.II Herve Foulot, M.D.1f

John Reidy, M.D.§ Michel Chapman, M.D.§ Jean Rene Zorn, M.D.*

Clinique Universitaire Baudelocque, Hopital Cochin, and Maternite Port Royal Hospital, Paris, France; Guys Hospital, London, England; and Lessingstrasse Clinic, Stuttgart, Germany

Objective: To determine the efficacy of transcervical tubal cannulation and intraluminal methotrexate injection for the management of tubal ectopic pregnancy (EP). Design: Prospective multicenter study of 33 patients with tubal pregnancies. Setting: Four university-based gynecology and radiology departments in three different countries: France, England, and Germany. Patients: Thirty-three patients who presented with a clinical diagnosis of EP. Interventions: Patients underwent transcervical tubal cannulation under fluoroscopic or ultrasound control and local injection of methotrexate (up to 50 mg). Main Outcome Measures: We evaluate the feasibility of transcervical tubal cannulation for the management of tubal pregnancy. Results: Two patients elected to withdraw from the protocol. In the remaining 31 patients there was complete resolution of the EP in 27 (87%). Surgery was performed in 4 patients. Seventeen patients, 14 of whom desired pregnancies, were available for follow-up to assess the return of reproductive potential. Seven of 7 patients who subsequently underwent hysterosalpingography had patency of the affected tube. Five patients later had an intrauterine pregnancy. One patient had an early miscarriage, two have given birth, and two singleton pregnancies are still ongoing. The remaining patients are symptom free. Conclusions: This study demonstrates that transcervical tubal catheterization in patients with tubal pregnancies is feasible and can be performed without anesthesia or analgesia in most cases. Intraluminal methotrexate per se is capable of causing regression of the EP. This approach offers a new alternative for the treatment of selected patients with tubal EP. Fertil SterilI992;58:1131-5 Key Words: Transcervical tubal cannulation, ectopic pregnancy, methotrexate injection

The reported incidence of tubal ectopic pregnancy (EP) is rising. Recent figures suggest an incidence of 1 in 50 to 1 in 79 live births (1). Some Received December 12, 1991; revised and accepted August 18, 1992. * Clinique Universitaire Baudelocque. t Hopital Cochin. :j: Reprint requests and permanent address: Francisco Risquez, 69, Ave pplla Castellana, Caracas, Venezuela. § Guys Hospital. II Lessingstrasse Clinic. 11 Maternite Port Royal. Vol. 58, No.6, December 1992

of this increase may reflect the rise in the number of cases of sexually transmitted disease (2), but diagnostic advances, notably transvaginal ultrasound (US) and rapid quantitative human chorionic gonadotropin (heG) assays, have significantly improved our ability to identify early possibly asymptomatic tubal pregnancies. Over the last few years there has been great interest in more conservative management of tubal pregnancies. Systemic methotrexate (MTX) was first used for an interstitial EP in 1982 (3), but because of the high incidence of side effects, direct administraRisquez et al.

Tubal cannulation in ectopic pregnancy

1131

tion of MTX into the gestation sac was proposed. The drug can be administered at laparoscopy (4) or using vaginal US (5). Drug administration is usually preceded by aspiration of the contents of the gestational sac. Other drugs have also been locally administered. These include potassium chloride (6), prostaglandin derivatives (7,8), and hyperosmoloar glucose (9). All treatment modalities depend on the preliminary diagnosis of the tubal pregnancy. Laparoscopy is the most accurate diagnostic tool but is invasive. At the simplest level the diagnosis can be suspected by elevated hCG levels in the absence of an intrauterine pregnancy on transvaginal US. Recently a new nonsurgical technique has been proposed for both the diagnosis and treatment of EP. Transcervical tubal cannulation is combined with selective salpingography to diagnose the ectopic gestation and then MTX is administered into the tubal lumen. In a preliminary report Risquez et al. (10) described the use of this technique in 4 patients. The present expanded series reports the outcome of a multicenter study of 33 patients with EP managed by transcervical tubal cannulation.

ditions the cervix was fixed using a single-toothed tenaculum forceps and the introducer advanced through the cervical os to the fundus. On withdrawing the metal mandril, the introducer's tip oriented toward the selected cornua. The inner coaxial catheter was then advance through the introducer and the interstitial portion of the fallopian tube was cannulated using tactile impression. In 18 cases, the correct placement of the catheter was verified using US. In the remainder, fluoroscopy was performed and contrast medium was injected through the inner catheter. After visualization of the tubal pregnancy, the soft-tipped metallic guidewire was advanced to the implantation site and the position was confirmed by fluoroscopy. The coaxial catheter was advanced over the guidewire and the latter was withdrawn. Methotrexate was then injected through the coaxial to the implantation site. All patients were followed up using serial hCG measurements until baseline levels of hCG were recorded or a further intervention became necessary. They were advised to wait for 3 months before attempting conception. RESULTS

MATERIALS AND METHODS

The series consists of 33 patients who presented with a clinical diagnosis of ectopic pregnancy. Some patients were symptomatic. In others, the diagnosis was made after 6 weeks amenorrhea in fertility patients undergoing active treatment such as intrauterine insemination (lUI) or in vitro fertilization (IVF). Pregnancy was confirmed by elevated hCG levels and intrauterine pregnancy was excluded by transvaginal US using a 5- or 7.5-MHz probe. Eighteen of the 33 patients had estradiol levels >2,000 IU /L. This level is highly discriminatory for the presence of an intrauterine pregnancy (11). The remaining patients had US evidence of a mass or gestation sac lateral to the uterus and/or had sufficient symptoms to warrant laparoscopy if tubal cannulation had not been available. The following inclusion criteria were adopted: patients all desired future fertility, there was no clinical evidence of tubal rupture, and patients were hemodynamically stable. Informed written consent was obtained by all patients. The procedure was performed as previously described using the Risquez-Zorn catheter system (Nycomed, Paris, France) (12). This consists of an introducer with a curved tip that is straightened using a metal obturator, two coaxial inner catheters, and a soft tip metal guidewire. Under aseptic con1132

Risquez et al.

Tubal cannulation in ectopic pregnancy

Thirty-three patients were treated by transcervical tubal cannulation and MTX. Their ages ranged from 25 to 41 years (32.0 ± 4.5 years; mean ± SD). The mean (±SD) gestational age at presentation was 6.6 ± 1.7 weeks (range 5 to 12 weeks). Only 8 patients were asymptomatic. Eighteen patients presented with abdominal pain and 16 had vaginal bleeding. Transvaginal US was performed in all patients and no intrauterine pregnancies were recorded. In 21 patients a mass was documented lateral to the uterus. A fetal sac was observed in 5 of these but fetal heart activity was only present in one case. Free peritoneal fluid was documented in 17 patients «100 mL). On the day of transcervical tubal cannulation the mean hCG level was 2,250 ± 2,361 mIU /mL (range 217 to 10,600 mIU/mL). There was no significant correlation between the gestational age and the level of hCG (r = 0.0374, P = 0.856). The dose of the MTX administered varied. Twenty-two patients received 50 mg, 10 patients received 20 to 50 mg, and 1 patient received 5 mg. After transcervical tubal cannulation and MTX administration, 2 of the 33 patients elected to withdraw from the protocol and have operative treatment. In 27 of the remaining 31 patients (87%), there was complete resolution of the EP (as shown by reFertility and Sterility

duction ofhCG to baseline levels) (Table 1). Surgery was performed in the remaining 4 patients. The first patient had MTX when the hCG was 10,600 mIU /mL. Despite reduction in the hCG level over the next 13 days to 2,225 mIU /mL the EP ruptured and an emergency laparotomy was required. A second patient had MTX when the hCG was 7,250 mIU /mL. Human chorionic gonadotropin continued to rise over the 7 days after treatment to 10,200 mIU /mL and surgery was electively performed. The third patient required a laparoscopy for pain associated with intra-abdominal bleeding 23 days after MTX. The hCG had fallen from 1,935 to 24 mIU /mL at the time of surgery. The fourth patient developed clinical symptoms of acute salpingitis 4 days after transcervical tubal cannulation and MTX administration with severe lower abdominal pain associated with a fever of 40°C. She underwent salpingectomy. At surgery there was no evidence of pelvic infection. Examination of the removed fallopian tubes of those patients who underwent salpingectomy revealed chronic salpingitis in 2 patients, tubal bleeding in 1 patient, and evidence of tubal rupture in 1 patient. The mucosa was replaced by fragments of decidual tissue and fibrous placental villi. There was no pathological evidence of acute necrosis within the fallopian tubes. In the 27 patients successfully treated, the time taken to achieve baseline levels of hCG was related to the initial hCG level on the day of MTX. The mean (±SD) number of days to achieve resolution was 18.0 ± 9.8, 25.8 ± 8.6, and 37.0 ± 12.7 when initial hCG level was 5,000 mIU/mL, respectively. None of these patients experienced any side effects. Treatment was successful in all 11 patients with hCG 5,000 mIU /mL. Seventeen patients were available for follow-up to examine the return of reproductive potential after the procedure. Ofthese 17 patients, 14 desired pregnancies as soon as possible. Seven of 7 patients who subsequently underwent hysterosalpingography had a normal uterine configuration and patency of the affected tube. Five patients later had an intrauterine pregnancy. Two pregnancies arose spontaneously, including one through the treated tube (the patient had previous salpingectomy for tubal pregnancy); another patient became pregnant after lUI, and 2 others after IVF (due to male factor infertility). Of these five pregnancies 1 patient later had an early miscarriage, 2 have given birth to three normal baVol. 58, No.6, December 1992

Table 1 Results of Transcervical Tubal Cannulation in 39 EPs No. of patients treated Age (y) Gestation (wk) RCG on day of cannulation (lUlL) Maximum hCG (lUlL) Median days to resolution

32.0 6.6 2,250 3,267 25

33 ± 4.6 ± 1.7 ± 2,361 ± 3,965 ± 10

bies (one twin pregnancy and one singleton), and two singleton pregnancies are still ongoing. The other patients remain symptom free. DISCUSSION

There is a general consensus that the clinical presentation of EP has changed over the past 20 years (1). Patients present earlier and a diagnosis can frequently be made without the recourse to surgery by transvaginal US and highly specific hCG assays. Patients who are undergoing assisted conception procedures are very closely monitored after conception and frequently an EP is diagnosed in asymptomatic patients. It has been argued that this changing presentation necessitates a reappraisal of therapeutic strategies for managing these patients with a move toward conservative surgery and medical management. At one end of the spectrum expectant management has been proposed in selected patients with quoted resolution rates of 57% to 92% (13-15). Moreover, Fernandez et al. (16) found that when initial hCG level is 3.5 cm in its greatest dimension (19). Methotrexate or other agents can be injected directly into the tubal gestation sac. This can be done under Risquez et al.

Tubal cannulation in ectopic pregnancy

1133

direct vision at laparoscopy but this requires the patient to be subject to a surgical intervention with associated morbidity and even mortality. Alternatively MTX can be administered using transvaginal US. This is only applicable if the gestation sac can be visualized (5). In our series this would only have been possible in 16% (5 of 31) of the cases. In a series of 191 patients reported by Pansky et al. (1), the EP was only definitely diagnosed by US in 7%. Other authors have reported visualizing a sac in up to 70% of patients with EP (20). The advantage of transcervical tubal cannulation is that it is possible to attempt the technique in all patients with an unruptured EP. As there is no needling of the gravid tube there is less risk of traumatic injury to vessels (21). The results of our study agree with previous observations that conservative therapy with MTX as the antitrophoblastic agent has a higher failure rate when the heG at the time of treatment exceeds 5,000 mIU/mL. There was no apparent relation between treatment failure and increased gestation age. Indeed gestation age was not correlated with heG levels. In 65% of patients (20/ 31) heG levels increased in the first 4 days after administration of MTX but in those cases associated with resolution of the EP the heG level had plateaued after 6 days. In vitro studies in choriocarcinoma cells show that there is a 10-fold increase in heG synthesis in these cells after exposure to MTX (22). A transient rise in heG should therefore not be taken as a sign of treatment failure. Two other important practical points emerged from this work. First, tubal rupture can occur even when heG levels are declining, and second, surgical intervention may be necessary in the presence of nearly baseline heG levels. This latter case is probably a rare occurrence associated with continued bleeding from a necrotic area around the implantation site (23). However, all conservative surgical or medical therapeutic strategies need careful follow up by monitoring heG levels and possibly by repeated transvaginal US and clinical examination until the EP has resolved both clinically and biochemically (24). Previous studies in which the MTX was injected into the gestational sac after aspiration of sac contents questioned whether the needle aspiration itself was implicated in the success of therapy (20). This study demonstrates that intraluminal MTX per se is capable of causing regression of the EP. The present study demonstrates that transcervical tubal catheterization in patients with tubal pregnancies is feasible and may be performed without anesthesia or analgesia in most cases. The ease of 1134

Risquez et aI.

Tubal cannulation in ectopic pregnancy

the procedure, the nonsurgical nature of this approach, the absence of drug toxicity, and the efficacy of the treatment is promising. It is surprising that failure to catheterize the fallopian tube did not occur, contrary to previous reports (25). It is possible that hormonal changes induced by the EP resulted in an enlargement of the tubal ostium compared with the nonpregnant state, allowing easy cannulation of the fallopian tubes. Other theoretical explanations of this phenomenon may include the presence of increased tubal pressure, secretion of materials that cause smooth muscle relaxation, or reflux of blood through the fallopian tube. Although the procedure is easy to perform, it requires experience with catheterization of fallopian tubes. In conclusion, the results presented in this multicenter study are preliminary in nature, and larger studies are required to establish the methodology, efficacy, and safety of this new approach. If these results are substantiated by other groups, it appears as if transcervical tubal cannulation may offer a new easy alternative for the management of selected patients with tubal EP. Ackrwwledgment. Special thanks are due to Miss Therese Marie Mignot for her statistical assistance.

REFERENCES 1. Pansky M, Golan A, Bukovsky I, Caspi E. Nonsurgical management of tubal pregnancy. Necessity in view of the changing

clinical appearance. Am J Obstet Gynecol 1991;164:888-5. 2. Forman RG. Acute pelvic sepsis and tubo-ovarian abscess. In: Oxford textbook of surgery. Oxford: Oxford University Press. In press. 3. Tanaka T, Hayashi H, Kutsuzawa S, Fujimoto S, Ichinoe K. Treatment of interstitial ectopic pregnancy with methotrexate: report of a successful case. Fertil Steril1982;37:851-2. 4. Pansky M, Golan A, Schneider D, Arieli S, Bukovsky I, Weinraub Z, et a1. Tubal patency after local methotrexate injection for tubal pregnancy. Lancet 1989;2:967-8. 5. Feichtinger W, Kemeter P. Conservative treatment of ectopic pregnancy by transvaginal aspiration under sonographic control and methotrexate injection. Lancet 1987;1:381-2. 6. Robertson DE, Moye MAH, Hansen IN, Serhal P, Smith W, Brinsden PR, et a1. Reduction of ectopic pregnancy by injection under ultrasound control. Lancet 1987;2:974-5. 7. Lindblom B, Hahlin M, Lundorff P, Thorburn J. Treatment of tubal pregnancy by laparoscope-guided injection of prostaglandin F 2 • Fertil Steril1990;54:404-8. 8. Egarter CH, Husslein P. Treatment of tubal pregnancy by prostaglandins. Lancet 1988;1:1104-5. 9. Lang P, Weiss PAM, Mayer HO. Local application of hyperosmolar glucose solution in tubal pregnancy. Lancet 1989;2:922-3. 10. Risquez F, Mathieson J, Pariente D, Foulot H, Dubuisson JB, Bonnin A, et a1. Diagnosis and treatment of ectopic pregFertility and Sterility

II.

12.

13. 14.

15.

16.

17.

18.

nancy by retrograde selective salpingography and intraluminal methotrexate injection: work in progress. Hum Reprod 1990;5:759-62. Stovall TG, Ling FW, Carson SA, Buster JE. Nonsurgical treatment of tubal pregnancy [letter]. Fertil Steril 1991;55: 1020. Risquez F, Mathieson J, Zorn JR. Twin pregnancy after diagnosis and treatment of ectopic implantation by retrograde selective salpingography and intraluminal methotrexate injection. Fertil Steril1990;54:1168-70. Lund JJ. Early ectopic pregnancy treated nonsurgically. J Obstet Gynaecol Br Emp 1955;62:70-6. Garcia AJ, Aubert JM, Sarna J, Josimovich JB. Expectant management of presumed ectopic pregnancies. Fertil Steril 1987;48:395-400. Carp HJA, Oelsner G, Serr DM, Mashiach S. Fertility after nonsurgical treatment of ectopic pregnancy. J Reprod Med 1986;31:119-22. Fernandez H, Rainhorn JD, Papiernik E, Bellet D, Frydman R. Spontaneous resolution of ectopic pregnancy. Obstet GynecoI1988;71:171-4. Tulandi T, Hemmings R, Khalifa F. Rupture of ectopic pregnancy in women with low and declining serum iJ-human chorionic gonadotropin concentrations. Fertil Steril 1991;56: 786-7. Sauer MV, Gorrill MJ, Rodi lA, Yeko TR, Greenberg LH, Bustillo M, et al. Nonsurgical management of unruptured

Vol. 58, No.6, December 1992

19.

20.

2I.

22.

23.

24.

25.

ectopic pregnancy: an extended clinical trial. Fertil Steril 1987;48:752-5. Stovall T, Ling F, Gray LA, Carson SA, Buster J. Methotrexate treatment of unruptured ectopic pregnancy: a report of 100 cases. Obstet GynecoI1991;77:749-53. Menard A, Crequat J, Maldelbrot L, Hauuy J-P, Madelenat P. Treatment of unrupted tubal pregnancy by local injection of methotrexate under transvaginal sonographic control. Fertil SteriI1990;54:47-50. Honigl W, Lang P, Drdl W, Weiss PAM. Rupture of tubal pregnancy after transvaginal puncture. Lancet 1991;336: 885-6. Speeg KV, Clifford Azizkhan J, Stromberg K. The stimulation by methotrexate of human chorionic gonadotropin and placental alkaline phosphatase in cultured choriocarcinoma cells. Cancer Res 1976;36:4570-6. Gretz E, Quagliarello J. Declining serum concentrations of the beta-subunit of human chorionic gonadotropin and ruptured ectopic pregnancy. Am J Obstet Gynecol 1987;164: 940-I. Dumesic DA, Hatez R. Delayed hemorrage of a persistent ectopic pregnancy following laparoscopy and methotrexate therapy. Obstet Gynecol 1991;78:960-2. Risquez F, Boyer P, Rolet F, Magnani M, Guichard A, Cedard L, Zorn JR. Retrograde tubal transfer of human embryos. Hum Reprod 1990;5:185-8.

Risquez et al.

Tubal cannulation in ectopic pregnancy

1135