Abstract
Cervical varix during pregnancy is a rare condition, and standard management for bleeding from a varix has not been established. We performed cross double cervical cerclage and effectively stopped bleeding. A 41-year-old female had a twin pregnancy. The development of a cervical varix was observed during pregnancy and bleeding from ruptured varix started at 20 weeks of gestation. We performed surgical hemostasis by cervical cerclage. In the first cerclage, we could not stop the bleeding from the varix. For further restriction of blood supply to the cervical varix, we performed a second cerclage in a crossed position on a deeper side of the vagina than the first cerclage. Then the bleeding completely stopped and there was no bleeding until delivery. The “cross double McDonald cerclage” performed in our patient may be a useful modified cerclage method for stopping intractable bleeding from the cervix during pregnancy.
Keywords: Haemostasis, uterine cervix cerclage
INTRODUCTION
Cervical varix during pregnancy may cause moderate to severe hemorrhage if varix ruptures.[1] Some reported cases needed termination of pregnancy or hysterectomy for uncontrollable bleeding.[2,3]
In our case, massive hemorrhage caused by the rupture of cervical varix occurred at 20 weeks of gestation. She needed a blood transfusion and the bleeding could not be controlled by conservative treatment using gauze packing. Therefore, after a detailed explanation about the surgery, we tried cervical cerclage for hemostasis. We were able to stop the bleeding using a modified double McDonald cerclage technique. The clinical course and the method are described in detail in the following.
CASE REPORT
A 41-year-old woman, G1P0, had a DD (dichorionic-diamniotic) twin pregnancy by in vitro fertilization after 2 years of treatment for primary infertility. She had a history of cervical conization for adenocarcinoma in situ. At 20 weeks of gestation, she visited our hospital with a complaint of abnormal bleeding from her vagina. We found abnormal development of cervical varix by vaginal ultrasound and the bleeding seemed to be caused by varix. We performed vaginal gauze packing for hemostasis and the bleeding was decreased for a week. We performed a cervical pap test at that time and the result was negative for intraepithelial lesion or malignancy (NILM). However, at 21 weeks of gestation, massive bleeding occurred and the amount of bleeding reached almost 800 g. We recommended surgical hemostasis or termination of the pregnancy, but she refused both opinions because of the risks of the surgery, including infection and premature rupture of the membrane, and she wished to have further preservative treatment. Therefore, we added vaginal gauze packing once more and observed her with the administration of blood transfusion. However, the bleeding persisted and the amount of bleeding gradually increased. The total amount of bleeding reached about 2000 g at 22 weeks of gestation. We informed her that the bleeding was becoming life-threatening and that we must perform surgical treatment or termination of the pregnancy. She and her husband finally accepted to try surgical hemostasis by cervical cerclage.
Under spinal anesthesia, we detected two bleeding points on the surface of the cervix caused by rupture of cervical varix, the findings of a developed varix by vaginal ultrasound seemed to be match the bleeding points [Figure 1a]. We gently held the cervix with the varix using segond uterine forceps and performed the first cerclage by the usual McDonald cerclage method using Teflon tape, enclosing the ruptured points of the varix. However, the bleeding only decreased slightly after the first cerclage and the bleeding persisted from each ruptured point. Therefore, we performed a second cerclage at a 45° rotated position on a deeper side of the vagina than the first cerclage. After the second cerclage, bleeding completely stopped. Vaginal ultrasound in color Doppler examination showed that the blood supply to the cervix appeared to be greatly decreased [Figure 1b]. Total operative time was 29 min, and blood loss was 1650 g.
Figure 1.
Method for cross double McDonald cerclage and findings in intraoperative transvaginal ultrasound. (a) Routes of the cross double cerclages and the bleeding point (gray circle), and the bleeding points are marked with a white triangle in transvaginal ultrasound. (b) Sagittal view from the right side of the cervix after cerclage, and sutured strings (marked with a white triangle)
After hemostasis by the cross double cerclage, there was no vaginal bleeding until delivery. From 30 weeks of gestation, the fetuses showed a discrepancy in estimated fetal weight. The second twin showed selective intrauterine growth restriction (sIUGR) even though her pregnancy was dichorionic, and we suspected an effect of the surgery for cervical varix, but the cause was undetectable antenatally. Finally, we performed a cesarean section due to the significant growth arrest of the second twin at 34 weeks gestation. The first twin was male with a weight of 2010 g (appropriate for dates) and the second twin was female with a weight of 1248 g (small for dates). The placenta of the second twin showed velamentous insertion of the umbilical cord. We diagnosed velamentous insertion as the cause of the growth restriction of the second twin. Cervical varix rapidly disappeared after delivery of the fetuses, and the bleeding did not persist and needed no further treatment. We could remove the two Teflon tapes immediately after the cesarean section without any trouble. The patient was discharged from the hospital without any problems after surgery, and the babies were also discharged without any complications after temporary treatment in the neonatal care unit.
DISCUSSION
During pregnancy, general varices in the legs, vulva, and hemorrhoidal plexus are commonly observed in approximately 15% of all pregnancies. However, a uterine cervical varix is rare.[3] It was previously reported that the risk of uterine cervical varices was increased by exposure to diethylstilbestrol (DES) during pregnancy.[3] However, there have been few such reports recently DES has been defined as a contraindication to pregnancy. Some studies have suggested a relationship of cervical varices with surgeries for uterine cervices such as trachelectomy and conization.[1,4] The invasiveness of surgeries may induce the development of the vessels in uterine cervices.
As far as we know, there have been only 14 published cases of cervical varices that included eight cases of vaginal bleeding with cervical varices. The previously reported cases and our case are summarized in Table 1.[2,5,6,7,8,9]
Table 1.
Reported cases of abnormal bleeding from cervical varix in pregnancy
| Report | Age | Gravida/Parity | GA at diagnosis | GA at termination | Treatment | Position of placenta | Reason for termination | Delivery mode | Blood loss |
|---|---|---|---|---|---|---|---|---|---|
| Follen et al. (1985) | 32 | 4/3 | 22 | 35 | Gauze packing, bed rest | NA | NA | (g) | NA |
| 31 | 5/1 | 13 | 31 | NA | PROM | NA | |||
| 32 | 3/0 | 28 | Term | Cervical cerclage, bed rest Gauze packing, bed rest | NA | NA | |||
| Fleming et al. (1993) | 26 | 2/1 | 13 | 14 | Cervical cerclage | NA | PROM | CS | NA |
| Hurton et al. (1998) | 33 | 3/0 | 33 | 34 | Tocolyosis | Not low | Abnormal bleeding | CS with hysterectomy | 5000 |
| Yohimura et al. (2004) | 34 | 1/0 | 18 | 27 | Tocolyosis, bed rest | Previa | Bleeding from previa | CS | 920 |
| Kusanovic et al. (2006) | 26 | 3/0 | 21 | 32 | Tocolysis, gauze packing | Low lying | Tocolysis failure, abnormal bleeding | CS | NA |
| Kumazawa et al. (2007) | 30 | 1/0 | 27 | 32 | Tocolysis, gauze packing | Previa | Tocolysis failure | CS | 1814 |
| Present case | 41 | 1/0 DD twin | 20 | 34 | Double cerclage tocolysis, bed rest | Low lying | Fetal growth restriction | CS | 3643 |
GA: Gestational age (week), CS: Cesarean section, VD: Vaginal delivery, NA: Not assessed; PROM: preterm rupture of membranes, DD twin: dichorionic diamniotic twin
In eight of the nine cases, live infants were born between 27 weeks and 35 weeks of gestation by cesarean section, only one case being a term vaginal delivery. The main reason for the termination of pregnancy was vaginal bleeding of a cervical varix (2 of the 9 cases). Four of the nine cases were complicated with placenta previa or low-lying placenta. Placenta previa or low-lying placenta may contribute to the development of cervical varices through increased volume of the uterine cervical vasculature. In our case, the placenta was in a low position and the delivery mode was cesarean section due to significant growth restriction of the second twin not to vaginal bleeding. The pathology of the placenta revealed that the cause of selective IUGR was the velamentous insertion of the cord. In all cases, treatment was needed to stop the bleeding of the cervical varix. Cervical cerclage was performed in three of the nine cases. In one case, 1 week after cerclage placement, spontaneous abortion occurred due to premature rupture of the membrane.[6] In the two other cases (including our case), cerclage was effective for hemostasis and the duration of pregnancy was extended for 14 weeks and 18 weeks. On the other hand, preservative treatments such as bed rest and packing extended the pregnancy by only 8.5 weeks on average.
Fleming and Anderson. first reported the use of cervical cerclage for preventing bleeding of cervical varix in 1993.[6] Tsai et al. first reported double cerclage, but their purpose was preventing preterm labor, not bleeding of cervical varix. However, the reported double cerclages were not crossed position.[10] In the present case, we performed the second cerclage in a rotated cross position for more efficient restriction of blood flow, as shown in Figure 1b. We explained to the patient about the risk of necrosis of the cervix by the loss of the blood supply after surgery. However, the condition we feared did not occur. We think that the gap between the two cerclages may have supplied the minimum necessary blood flow to the cervix. As for the difficulty of the method, we could perform the first and second cerclages without any failure even in poor condition of the operating field caused by massive bleeding from cervical varix and severe defect of the cervix. McDonald cerclage is an easier method than Shirodkar cerclage for cervical cerclage, and in the second cerclage, we can tug and control the cervix by holding the string of the first cerclage. Therefore, we could set the second cerclage on a deeper side of the cervix effectively without any problem.
Finally, we conclude that cross double McDonald cerclage is an effective and reliable method for hemostasis of a ruptured cervical varix. This method should be tried before consideration of termination of pregnancy in an early phase of pregnancy.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent forms. In the form, the patient has given her consent for her images and other clinical information to be reported in the journal. The patient understands that name and initials will not be published and due efforts will be made to conceal identity, but anonymity cannot be guaranteed.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
REFERENCES
- 1.Brown JV, 3rd, Mills MD, Wong H, Goldstein BH. Large volume cervical varix bleeding in a gravid patient. Gynecol Oncol Case Rep. 2012;4:20–2. doi: 10.1016/j.gynor.2012.12.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Hurton T, Morrill H, Mascola M, York C, Bromley B. Cervical varices: An unusual etiology for third-trimester bleeding. J Clin Ultrasound. 1998;26:317–9. doi: 10.1002/(sici)1097-0096(199807/08)26:6<317::aid-jcu7>3.0.co;2-l. [DOI] [PubMed] [Google Scholar]
- 3.Tanaka M, Matsuzaki S, Kumasawa K, Suzuki Y, Endo M, Kimura T. Cervical varix complicated by placenta previa: A case report and literature review. J Obstet Gynaecol Res. 2016;42:883–9. doi: 10.1111/jog.12969. [DOI] [PubMed] [Google Scholar]
- 4.Okugawa K, Kobayashi H, Sonoda K, Kaneki E, Kawano Y, Hidaka N, et al. Oncologic and obstetric outcomes and complications during pregnancy after fertility-sparing abdominal trachelectomy for cervical cancer: A retrospective review. Int J Clin Oncol. 2017;22:340–6. doi: 10.1007/s10147-016-1059-9. [DOI] [PubMed] [Google Scholar]
- 5.Follen MM, Fox HE, Levine RU. Cervical vascular malformation as a cause of antepartum and intrapartum bleeding in three diethylstilbestrol-exposed progeny. Am J Obstet Gynecol. 1985;153:890–1. doi: 10.1016/0002-9378(85)90697-0. [DOI] [PubMed] [Google Scholar]
- 6.Fleming AD, Anderson TL. Cervical AV malformation with in utero DES exposure. Nebr Med J. 1993;78:138–9. [PubMed] [Google Scholar]
- 7.Yoshimura K, Hirsch E, Kitano R, Kashimura M. Cervical varix accompanied by placenta previa in twin pregnancy. J Obstet Gynaecol Res. 2004;30:323–5. doi: 10.1111/j.1447-0756.2004.00202.x. [DOI] [PubMed] [Google Scholar]
- 8.Kusanovic JP, Soto E, Espinoza J, Stites S, Gonçalves LF, Santolaya J, et al. Cervical varix as a cause of vaginal bleeding during pregnancy: Prenatal diagnosis by color Doppler ultrasonography. J Ultrasound Med. 2006;25:545–9. doi: 10.7863/jum.2006.25.4.545. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Kumazawa Y, Shimizu D, Hosoya N, Hirano H, Ishiyama K, Tanaka T. Cervical varix with placenta previa totalis. J Obstet Gynaecol Res. 2007;33:536–8. doi: 10.1111/j.1447-0756.2007.00553.x. [DOI] [PubMed] [Google Scholar]
- 10.Tsai YL, Lin YH, Chong KM, Huang LW, Hwang JL, Seow KM. Effectiveness of double cervical cerclage in women with at least one previous pregnancy loss in the second trimester: A randomized controlled trial. J Obstet Gynaecol Res. 2009;35:666–71. doi: 10.1111/j.1447-0756.2008.01006.x. [DOI] [PubMed] [Google Scholar]