Cervical pregnancy, where the ectopic gestation implants in the endocervical canal, is a life-threatening and an uncommon variety of ectopic pregnancy. Its incidence ranges from 1 in 1,000 to 1 in 18,000 pregnancies.1
In the past, cervical pregnancy was treated with hysterectomy, hence precluding future pregnancies. However, recent advances in ultrasound equipment and easy access to quantitative β-hCG have made the diagnosis of early ectopic pregnancies possible. This enables successful application of an array of conservative treatments such as local and systemic methotrexate (MTX) injection or in combination with mifepristone, local potassium chloride (KCl), or vasopressin injection. Minimally invasive procedures include hysteroscopic endocervical resection, electrodessication with Kleppinger device, or cervical amputation. Additional procedures include cervical cerclage or intracervical balloon tamponade after curettage or aspiration, hypogastric artery, bilateral uterine artery, or unilateral internal iliac artery ligation or embolization for controlling excessive bleeding when MTX injection proves unsuccessful.1–26 These procedures lead to a significant decline in complications and preservation of the patients’ reproductive potential.
With the aid of a high-resolution transvaginal ultrasound examination, a chemotherapeutic agent such as MTX or KCl could be locally injected into the ectopic cervical implantation. This treatment procedure is known to be advantageous for patients with cervical pregnancies for whom surgical management can limit reproductive capacity.
There have been several case studies of live cervical ectopic pregnancies that used local injection of KCl or MTX or both.2,3 We have sought to examine a very rare condition and described our experience with early cervical pregnancies managed conservatively with sonographically guided injection of MTX and KCl.
MATERIALS AND METHODS
Between 1993 and 2004, patients who were referred by their primary physicians to either of two tertiary medical centers (Cathay General Hospital and Mackay Memorial Hospital, Taipei, Taiwan) for reevaluation and conservative management of a suspected cervical pregnancy were included in this prospective study, which was approved by the institutional review boards of the hospitals. The patients had no previous therapeutic evaluations or procedures other than ultrasonography. Before treatment, the potential risks and benefits were explained to the patients. Informed consent for local MTX and KCl was also obtained from each of them.
In all cases, age, parity, body weight, hemoglobin level before and after the procedure, and the operation time were recorded. The gestational age at diagnosis, the presence or absence of cardiac activity, β-human chorionic gonadotropin (β-hCG) levels before and after treatment, the time it took for the β-hCG to become negative, the time to sonographic resolution, the time of resumption of menstruation, and the occurrence and outcomes of subsequent pregnancies were also considered. Each patient was judged to be suitable for the chemotherapy after a complete blood count and liver and renal function tests. Rh blood typing was also determined.
A review of the patients’ histories revealed that most of the patients (23 of 38) had had a previous dilation and curettage due to an unwanted prior pregnancy or incomplete abortion. Two patients had intrauterine devices at the time of pregnancy, and one patient had an intramural myoma. There were no patients who conceived with in vitro fertilization.
The enrolled patients were reevaluated in our institution by undergoing a transvaginal ultrasonography. Cervical ectopic pregnancy was diagnosed when the following conditions were observed: 1) the entire gestational sac identified in the dilated cervix had a well-formed spherical ovoid shape2; 2) the gestational sac may contain the yolk sac or an embryo with cardiac activity and was usually located below the level of the internal os along with an empty uterine cavity (Fig. 1)3; 3) the endometrial stripe could be visualized; and 4) the hourglass configuration of the uterus due to a barrel-shaped cervix would also suggest cervical pregnancy.4 The possibility of an aborted gestational sac transit in the cervical canal was excluded by performing a Doppler study. According to Jurkovic et al,5 a nonviable gestational sac passing through the cervix would not show peritrophoblastic flow.
After consent was obtained and the patients were found to be hemodynamically stable, they were treated with transvaginal sonographically guided injection of MTX or KCl into the gestational sac or embryo. In our institution, local application of KCl or MTX is performed with a percutaneous transhepatic cholangiodrainage gauge 22 needle (Terumo Company, Tokyo, Japan), originally developed for injection of contrast agent during percutaneous transhepatic cholangiography and drainage. Under light general anesthesia (intravenous propofol), a needle guide was attached to the ultrasound transducer (Acuson XP/10 Computed Sonography System, Mountain View, CA), and the needle was directed into the gestational sac. Fifty milligrams of MTX was introduced into the gestational sac. Mechanical disruption of the pregnancy (by aspiration of the sac) was also attempted before MTX injection. When a fetal heart beat was present, an intracardiac or intrathoracic injection of 2 mL KCl (Fig. 2A), at a concentration of 2 mEq/mL, was given, followed by MTX injection into the amniotic cavity and placenta, if present (Figs. 2B and 2C). The endpoint of the KCl injection was the cessation of cardiac activity. The puncture site was then observed for another 3 minutes for any postprocedure bleeding. The procedures were performed either on an outpatient basis or with brief hospitalization. If increased vaginal bleeding was noted during the 2–3 hours after the procedure, then the patient was hospitalized. The patients were informed in advance of the possible need to perform additional procedures in the event of uncontrollable bleeding. They were informed that such additional procedures might include intracervical placement of a tamponade Foley catheter balloon for 3 days and an additional intramuscular injection of 50 mg MTX.
Complete blood count, liver function tests, blood urea nitrogen, and creatinine levels were again obtained twice (at 2-week intervals) during the patients’ follow-up to identify any MTX-induced liver or renal toxicity, especially for those patients who received multiple doses of MTX. After discharge, the patients were followed with twice weekly quantitative β-hCG titers for 2 weeks, then weekly until levels were undetectable or below 5 milli-International Units/mL. Transvaginal sonography was performed during each of these follow-ups, looking for collapse of the gestational sac (Fig. 3) or complete disappearance of the cervical gestation with the resumption of normal menses. Patients who had a persistent trophoblast (as observed after a series of ultrasound examinations) or plateau or an arrest in the decline of β-hCG titer (failure to decline by more than 15% in 48 hours) received a variable-dose regimen involving alternating MTX 50 mg intramuscularly.
Patient compliance with regard to 1-year follow-up was good since most of the patients were young and were concerned about recurrence of cervical ectopic pregnancy. Through patient interviews, subsequent fertility outcomes were tabulated at least a year after the procedure (average 4.5 years). Patients were evaluated for any recurrence of cervical pregnancy or cervical incompetence during pregnancy. Patients were advised to wait for 3 months before attempting another pregnancy.
From the 1993 to 2004, 38 women underwent transvaginal ultrasound-guided local MTX or KCl injection procedure for cervical pregnancy. The patient demographic characteristics, including age, body weight, initial hemoglobin, gestational age, initial serum β-hCG, time to resolution of β-hCG to less than 5 milli-International Units/mL, and time of involution of gestated sac are listed in Table 1. The presence of fetal heart beat in the embryos was documented in 22 cases, whereas the rest had an ovoid gestational sac with yolk sac. From among those with observable fetal heart beats, the mean quantitative β-hCG titer before operation was 38,948 (range 5,608–103,256) milli-International Units/mL, whereas among those without fetal heart beats, the mean β-hCG was 8,972 (range 2,765–18,648) milli-International Units/mL (Table 1).
There were two cases (12 weeks with serum β-hCG titer of 89,506 milli-International Units/mL and 11 weeks with β-hCG titer of 57,729 milli-International Units/mL) where an additional MTX intramuscular injection was given due to a nondeclining β-hCG titer; in both cases, such treatment was successful. No other complications were identified.
Operation time was recorded as 3–8 minutes (mean 4.5 minutes). There were three patients who had significant bleeding after rupture of the amniotic membrane. Patients were 6.5, 7.3, and 7.6 weeks pregnant (viable pregnancy) and had an estimated blood loss of 350, 500, and 400 mL, respectively. These patients were managed conservatively by intracervical tamponade with a Foley catheter balloon for 3 days. They were also given additional 50-mg intramuscular doses of MTX. Although the potential for serious toxic effects exists, the low doses of methotrexate used in these patients did not cause any adverse effects. Because all of the patients in this study were Rh positive, anti-D administration was not necessary.
All of the subjects retained their uteri. No patients were lost to follow-up. Menstruation resumed after β-hCG was determined to be close to negative. All patients tolerated the procedure and were satisfied with the results. Each patient underwent a mean of nine follow-up ultrasound examinations (range 6–15).
After the procedure, 21 patients expressed a desire for future pregnancy. From among these 21 subjects, 18 had subsequent pregnancies. Fifteen women carried their pregnancies to term, which resulted in liveborn infants. One had preterm delivery (at 34 weeks) due to preterm labor. Two resulted in first-trimester pregnancy losses that were then subsequently followed by term pregnancies. There were no reports of cervical incompetence during the said pregnancies. There were three patients who were diagnosed as infertile due to an unknown cause. The remaining 17 patients did not desire pregnancy during a mean 4.5 years of follow-up (range 1–6 years).
The etiology of cervical ectopic pregnancy is still unclear. Several authors have observed that damage to the endometrial lining and the cervix during operative procedures to the uterus, dilatation and curettage resulting in decidual proliferation of endocervical epithelium, chromosomal abnormalities in the embryo, tumors and anomalies of the reproductive organs, Asherman’s syndrome, the use of intrauterine devices, and in vitro fertilization all predispose to cervical pregnancy.1,26
The diagnosis of early cervical pregnancy is now possible by transvaginal ultrasonography and serum β-hCG determination. Several decades ago, cervical ectopic pregnancy was managed radically by performing hysterectomy because the condition was usually not recognized early enough, leading to catastrophic and severe hemorrhage. Because the cervix is devoid of the contractile myometrium, hemorrhage is inevitable due to tissue disruption brought about by the ectopic implantation.28
There are several conservative methods of managing ectopic pregnancies. Variable-dose, single-dose, and low-dose MTX regimens have been studied for tubal pregancies.33–36 Direct injection into the conceptus has also been investigated. Because the use of methotrexate in ectopic pregnancy is relatively new, a standardized protocol is not yet available, even for tubal pregnancy.37
Methotrexate treatment of cervical pregnancy was first performed by Farabow et al in 1983.7 Since then, several cases have been reported in literature. Treatment of cervical pregnancy with local administration of MTX alone was described in five cases by Timor-Tritsch et al,3 three cases by Hung et al,27 and two cases by Monteagudo et al.9 Most other case reports or series describe a combination of strategies with feticide by direct puncture, feticide by KCl, systemic MTX before or after the local MTX injection, subsequent cervical curettage, and subsequent prophylactic embolization.
The intramuscular administration of MTX has been used to successfully treat common ectopic pregnancies, especially those located in the isthmic, ampullary, or fimbrial portions of the fallopian tube. According to Pansky et al, it seems that the systemic route for MTX in treating ectopic pregnancy has advantages over local injection in terms of simplicity, applicability to more patients, success rate, and fertility outcome (Pansky M. Methotrexate treatment for ectopic pregnancy: systemic versus local injection [abstract]. Scientific abstracts: proceedings of the First Congress on Controversies in Obstetrics and Gynecology, Prague, Czech Republic, 1999). Although not absolutely contraindicated in solitary cervical pregnancy, systemic MTX often fails in such pregnancies, sometimes requiring subsequent hysterectomy.2,38,39
Methotrexate is a chemotherapeutic agent which has the ability to inhibit growth of the trophoblast by inhibiting DNA synthesis and cell division.29 It is usually given at a dose of 50 mg locally into the ectopic site or systemically. In this study, local intra-amniotic instillation of 50 mg MTX was performed (Fig. 2B). There are several known advantages of the direct MTX injection into the ectopic site: 1) greater effectiveness, 2) shorter treatment time, and 3) reduced dosage with decreased adverse effects or toxicity compared with parenteral administration (gastrointestinal irritation, myelosuppression, and hepatotoxicity).
According to Hung et al,30 failure of single MTX chemotherapy is likely if the gestational age is less than 9 weeks, fetal viability is documented, serum β-hCG titer is less than 10,000 milli-International Units/mL, or fetal crown to rump length is greater than 10 mm. Hence, in this study, KCl was administered into the fetal heart, producing immediate cessation of cardiac activity. More than 50% of our patients had a gestational age of more than 9 weeks. Furthermore, the procedure was noted to be successful even in one of the patients who was 14 weeks pregnant. This is, by far, the most advanced gestational age reported in literature31 where MTX was administered locally. Bleeding was encountered in three patients whose gestational ages were less than 9 weeks and was due to rupture of the intra-amniotic membrane. Doubilet et al2 in 2004 stated that ectopic pregnancies containing embryos with fetal cardiac activity are ideal candidates for treatment with sonographically guided KCl injection. Moreover, Kung and Chang6 in 1999 reported that chemotherapy with MTX combined with adjuvant conservative procedure in treating cervical pregnancy has a reported success rate of 91%.
When vascular wall invasion by the trophoblast, neovascularization, and arteriovenous malformations are present, feticide and MTX, both systemically and locally, may be unsuccessful. Multiple blood transfusions or hysterectomy or both may be required, as in the case reported by Hajenius et al32 However, local medical treatment was successful in all our cases. We believe that direct fetal heart puncture and intrachorionic MTX injection were key factors. After the procedure, the patients noted vaginal bleeding and pain. Sometimes, intractable bleeding can occur with MTX. This may be attributed to tissue sloughing and necrosis or the shedding of trophoblast from an atonic cervix.27 When MTX therapy becomes complicated by significant hemorrhage, this may be amenable to simple local measures such as the inflation of a Foley catheter balloon in the endocervical canal. Hwang et al19 and Hsieh et al20 recommended vasopressin injection for cervical hemorrhage after local MTX.
The administration of MTX locally was chosen to avoid the adverse effect of systemic administration of MTX. Hung et al27 in 1998 observed that MTX alone or combined with procedures such as curettage or cervical tamponade is effective in ectopic pregnancies of 12 weeks of gestation or less. However, in this study, our results demonstrate that local MTX alone or in combination with KCL injection for viable cervical pregnancies and additional hemostatic procedures such as cervical tamponade can be effective for the treatment of cervical pregnancies even in advanced gestational age (up to 14 weeks). However, patients should be thoroughly counseled and advised about the procedure risks and the difficulty of predicting who may develop complications after treatment. If bleeding cannot be controlled by conservative measures, more invasive procedures, such as emergency embolization, artery ligation, or in some cases, hysterectomy, should be expected.25 Experience in invasive ultrasound-guided procedures is indeed crucial for the success of this treatment option, but careful selection of patients and their compliance is also important because patient follow-up is long and tedious, with β-hCG and pregnancy resolution of considerably less than 38 days (average of 38 and 49 days, respectively). In a study of nine patients managed with intra-amniotic MTX injection and subsequent cervical curettage, the time to return to a nonpregnant range varied from 22 to 72 days.15
There is still no consensus about the most appropriate treatment for cervical pregnancy. Therefore, large prospective randomized trials are still needed to establish common selection criteria, dosage of MTX, and length of follow up. But in women who desire future pregnancy, this type of conservative treatment may be an acceptable and promising alternative to hysterectomy in the management of cervical pregnancy.
1. Yankowitz J, Leak J, Huggins G, Gazaway P, Gates E. Cervical pregnancy case reports and current literature review. Obstet Gynecol Surv 1994;49:49–54.
2. Doubilet PM, Benson CB, Frates MC, Ginsburg E. Sonographically guided minimally invasive treatment of unusual ectopic pregnancies. J Ultrasound Med 2004;23:359–70.
3. Timor-Tritsch IE, Monteagudo A, Mandeville EO, Peisner DB, Anaya GP, Pirrone EC. Successful management of viable cervical pregnancy by local injection of methotrexate guided by transvaginal ultrasonography. Am J Obstet Gynecol 1994;170:737–9.
4. Hofmann HM, Urdl W, Hofler H, Honigl W, Tamussino K. Cervical pregnancy: case report and current concepts in diagnosis and treatment. Arch Gynecol Obstet 1987;241:63–9.
5. Jurkovic D, Hacket E, Campbell S. Diagnosis and treatment of early cervical pregnancy: a review and report of two cases treated conservatively. Ultrasound Obstet Gynecol 1996;8:373–80.
6. Kung FT, Chang SY. Efficacy of methotrexate treatment in viable and nonviable cervical pregnancies. Am J Obstet Gynecol 1999;181:1438–44.
7. Farabow WS, Fulton JW, Fletcher V, Jr, Velat CA, White JT. Cervical pregnancy treated with methotrexate. N C Med J 1983;44:91–3.
8. Pascual MA, Ruiz J, Tresserra F, Sanuy C, Grases PJ, Tur R, Barri PN. Cervical ectopic twin pregnancy: diagnosis and conservative treatment. Hum Reprod 2001;16:584–6.
9. Monteagudo A, Minior VK, Stephenson C, Monda S, Timor-Tritsch IE. Non-surgical management of live ectopic pregnancy with ultrasound-guided local injection: a case series. Ultrasound Obstet Gynecol 2005;25:282–8.
10. Cepni I, Ocal P, Erkan S, Erzik B. Conservative treatment of cervical ectopic pregnancy with transvaginal ultrasound-guided aspiration and single-dose methotrexate. Fertil Steril 2004;81:1130–2.
11. Heikinheimo O, Leminen A, Cacciatore B, Rutanen EM, Kajanoja P. Advanced cervical pregnancy: uterus-sparing therapy initiated with a combination of methotrexate and mifepristone followed by evacuation and local hemostatic measures. Acta Obstet Gynecol Scand 2004;83:211–6.
12. Feinberg E, Confino E. Electrodessication of a cervical heterotopic pregnancy. Fertil Steril 2004;82:448–9.
13. Radpour CJ, Keenan JA. Consecutive cervical pregnancies. Fertil Steril 2004;81:210–3.
14. Gyamfi C, Cohen S, Stone JL. Maternal complication of cervical heterotopic pregnancy after successful potassium chloride fetal reduction. Fertil Steril 2004;82:940–3.
15. Mesogitis S, Pilalis A, Daskalakis G, Papantoniou N, Antsaklis A. Management of early viable pregnancy. BJOG 2005;112:409–11.
16. Sherer DM, Dalloul M, Santoso P, Stimphil R, Sokolovski M, Abulafia O. Complete abortion of a nonviable cervical pregnancy following methotrexate treatment. Am J Perinatol 2004;21:223–6.
17. Celik C, Bala A, Acar A, Gezginc K, Akyurek C. Methotrexate for cervical pregnancy. A case report. J Reprod Med 2003;48:130–2.
18. Mendilcioglu I, Zorlu CG, Simsek M. Successful termination of a cervical pregnancy with misoprostol. Eur J Obstet Gynecol Reprod Biol 2003;106:96.
19. Hwang JL, Hsieh BC, Huang LW, Seow KM, Pan HS, Chen HJ. Successful treatment of a cervical pregnancy by intracervical vasopressin. BJOG 2004;111:387–8.
20. Hsieh BC, Lin YH, Huang LW, Chang JZ, Seow KM, Pan HS, et al. Cervical pregnancy after in vitro fertilization and embryo transfer successfully treated with methotrexate and intracervical injection of vasopressin. Acta Obstet Gynecol Scand 2004;83:112–4.
21. Takano M, Hasegawa Y, Matsuda H, Kikuchi Y. Successful management of cervical pregnancy by selective uterine artery embolization: a case report. J Reprod Med 2004;49:986–8.
22. Kung FT, Lin H, Hsu TY, Chang CY, Huang HW, Huang LY, et al. Differential diagnosis of suspected cervical pregnancy and conservative treatment with the combination of laparoscopy-assisted uterine artery ligation and hysteroscopic endocervical resection. Fertil Steril 2004;81:1642–9.
23. Saygili Yilmaz ES, Aydin D, Yilmaz Z. Conservative treatment of cervical pregnancy by evacuation after transvaginal suture ligation of the cervicovaginal branches of uterine arteries. Acta Obstet Gynecol Scand 2002;81:988–90.
24. Sherer DM, Lysikiewicz A, Abulafia O. Viable cervical pregnancy managed with systemic Methotrexate, uterine artery embolization, and local tamponade with inflated Foley catheter balloon. Am J Perinatol 2003;20:263–7.
25. Suzumori N, Katano K, Sato T, Okada J, Nakanishi T, Muto D, et al. Conservative treatment by angiographic artery embolization of an 11-week cervical pregnancy after a period of heavy bleeding. Fertil Steril 2003;80:617–9.
26. Dicker D, Feldberg D, Samuel N, Goldman JA. Etiology of cervical pregnancy: association with abortion, pelvic pathology, IUDs and Asherman’s syndrome. J Reprod Med 1985;30:25–7.
27. Hung TH, Chiu TH, Hsu JJ, Chen KC, Hsieh CC, Hsieh TT. Sonographic evolution of a living cervical pregnancy treated with intraamniotic instillation of methotrexate. J Ultrasound Med 1997;16:843–7.
28. Fylstra DL, Coffey MD. Treatment of cervical pregnancy with cerclage, curettage and balloon tamponade: a report of three cases. J Reprod Med 2002;46:71–4.
29. Rock JA, Thompson JD. Te Linde’s operative gynecology. 8th ed. Philadelphia (PA): Lippincott-Raven; 1997.
30. Hung TH, Jeng CJ, Yang YC, Wang KG, Lan CC. Treatment of cervical pregnancy with methotrexate. Int J Gynaecol Obstet 1996;53:243–7.
31. Jeng CJ, Lou CN, Tzeng CR, Yang YC. Successful conservative treatment of a 14-week gestational age cervical pregnancy by primary local injection of single dose intra-amniotic methotrexate and intracardiac potassium chloride feticide. Acta Obstet Gynecol Scand 2006;85:368–70.
32. Hajenius PJ, Roos D, Ankum WM, Van der Veen F. Are serum human chorionic gonadotropin clearance curves of use in monitoring methotrexate treatment in cervical pregnancy? Fertil Steril 1998;70:362–5.
33. Hajenius PJ, Engelsbel S, Mol BW, Van der Veen F, Ankum WM, Bossuyt PM, et al. Randomised trial of systemic methotrexate versus laparoscopic salpingostomy in tubal pregnancy. Lancet 1997;350:774–9.
34. Stovall TG, Ling FW. Single-dose methotrexate: an expanded clinical trial. Am J Obstet Gynecol 1993;168:1759–62.
35. Parker J, Bisits A, Proietto AM. A systematic review of single-dose intramuscular methotrexate for the treatment of ectopic pregnancy. Aust N Z J Obstet Gynaecol 1998;38:145–50.
36. Korhonen J, Stenman UH, Ylostalo P. Low-dose oral methotrexate with expectant management of ectopic pregnancy. Obstet Gynecol 1996;88:775–8.
37. American College of Obstetricians and Gynecologists. Medical management of tubal pregnancy. ACOG Practice Bulletin 3. Int J Gynaecol Obstet 1999;65:97–103.
38. Weston G, Kashyap R. Failed conservative management of cervical pregnancy despite falling beta-HCG. Aust N Z J Obstet Gynecol 2001;41:346–7.
39. Bai SW, Lee JS, Park JH, Kim JY, Jung KA, Kim SK, et al. Failed methotrexate treatment of cervical pregnancy, predictive factors. J Reprod Med 2002;47:483–8.
Figure. No caption available.