Cervical insufficiency and transvaginal cerclage — indication pathways, McDonald vs Shirodkar technique, and cervical length surveillance
Cervical insufficiency is classically defined as the inability of the cervix to retain a pregnancy in the second trimester in the absence of contractions or labor — painless cervical dilation leading to membrane prolapse, rupture, and delivery of an immature fetus. Because no single diagnostic test confirms cervical insufficiency prospectively, candidacy for cerclage is instead determined by one of three overlapping but distinct clinical pathways.
History-indicated (prophylactic) cerclage: • Based on obstetric history alone, placed empirically at 12–14 weeks regardless of current cervical length • Classic candidates: ≥1 prior second-trimester pregnancy loss preceded by painless cervical dilation, or ≥3 prior preterm births/second-trimester losses • Evidence: modest benefit demonstrated in the largest RCT (MRC/RCOG, 1993) primarily in women with ≥3 prior losses; benefit less clear with only one prior loss
Ultrasound-indicated cerclage: • Requires BOTH a qualifying history (prior spontaneous preterm birth <34–37 weeks depending on the series) AND a short cervix (<25mm) found on serial transvaginal ultrasound before 24 weeks • Without the qualifying prior-birth history, a short cervix is managed with vaginal progesterone rather than cerclage — progesterone and cerclage have not shown added benefit when combined routinely • Evidence: meta-analyses (Owen et al.; individual patient data meta-analysis) show reduced preterm birth <35 weeks and improved perinatal outcomes in this specific population
Physical exam-indicated (rescue) cerclage: • No preceding short-cervix surveillance triggers this pathway — dilation is discovered incidentally on routine exam or with prolapsing membranes • Placed emergently, without the benefit of a "normal" starting cervical length, so outcomes are inherently less favorable than elective cerclage • Evidence: retrospective series show significant pregnancy prolongation (weeks) and improved neonatal survival compared with expectant management alone, despite lower absolute success than elective cerclage
History-indicated cerclage is placed prophylactically early in the second trimester based purely on obstetric history, before ultrasound evidence of cervical shortening is required or even expected. Timing at 12–14 weeks allows first-trimester screening and confirmation of fetal viability and normal anatomy to be completed first, while still placing the suture well before the historical gestational age of prior losses.
The strength of evidence for history-indicated cerclage scales with the number and character of prior losses. Women with a single prior second-trimester loss that may have had an alternative explanation (e.g., infection, abruption, or a structural anomaly rather than painless dilation) do not clearly benefit and are often better served by ultrasound surveillance with cerclage reserved for those who develop a short cervix (the ultrasound-indicated pathway) rather than automatic prophylactic placement.
A thorough history should specifically characterize whether prior losses were truly painless and followed a pattern of progressive, silent dilation — versus preterm labor with contractions, PPROM, or an infectious/inflammatory process — since only the former pattern is classically attributed to cervical insufficiency and predicts benefit from mechanical reinforcement.
For women with a prior spontaneous preterm birth, transvaginal ultrasound cervical length surveillance every 1–2 weeks from 16 to 24 weeks gestation identifies progressive shortening before it becomes clinically apparent, allowing cerclage to be offered at the point cervical length crosses the 25mm threshold — capturing benefit for a population broader than just those with the classic painless-dilation history.
Cervical length is measured transvaginally with an empty bladder, using the straight-line distance between the internal and external os on a mid-sagittal image, with gentle fundal pressure applied to unmask funneling that may not be apparent at rest.
Funneling describes membrane protrusion into a dilated internal os, typically described by the percentage of the endocervical canal involved (e.g., ">25% funneling") or by shape (T, Y, V, or U-shaped os, in order of increasing severity). Funneling in the setting of a shortening cervix reflects the internal os beginning to open first, with the total measurable closed length progressively consumed from the top down — a pattern distinct from cervical shortening due to prior surgery (e.g., LEEP) which can produce a uniformly short but non-funneling canal.
Rate of change matters as much as the absolute value: a cervix shortening rapidly across two consecutive scans (e.g., 32mm → 22mm in two weeks) in a high-risk patient warrants closer surveillance or intervention even before crossing the 25mm threshold outright.
Rescue cerclage addresses the most advanced presentation: cervical dilation with or without visibly bulging membranes, discovered on physical exam in the absence of labor, bleeding, ruptured membranes, or clinical infection — most often up to approximately 23 to 24 weeks gestation, beyond which the risks of the procedure generally outweigh any benefit given the approaching threshold of viability.
Before rescue cerclage, evaluation must exclude the contraindications that make cerclage unsafe or futile: active preterm labor, clinical chorioamnionitis (fever, uterine tenderness, fetal tachycardia, maternal leukocytosis), ruptured membranes, vaginal bleeding suggestive of abruption, and a lethal or major fetal anomaly. Amniocentesis to exclude subclinical intra-amniotic infection is used selectively at some centers before proceeding.
When membranes are visibly prolapsing beyond the external os, technical maneuvers to reduce them back into the uterine cavity before suture placement include steep Trendelenburg positioning, bladder filling, and gentle manual or moistened-sponge reduction; a Foley catheter balloon can also be used to gently push membranes cephalad while the purse-string suture is placed and cinched around it before balloon deflation and removal.
Counseling before rescue cerclage must be explicit that success is variable and unpredictable, that the procedure itself carries a small risk of precipitating rupture of membranes or labor, and that the alternative of expectant management (with or without amnioreduction) remains reasonable in patients who prefer to avoid the procedure.
Both McDonald and Shirodkar cerclage aim to mechanically reinforce the internal os with a circumferential purse-string suture, but differ in surgical complexity: McDonald is a simple, quick, and reversible transvaginal suture without tissue dissection, while Shirodkar requires bladder mobilization to place the suture submucosally at a higher level near the internal os itself.
McDonald cerclage: • Purse-string suture of nonabsorbable material placed circumferentially through the cervical stroma at the cervicovaginal junction, typically with 4–5 bites, without incising the vaginal mucosa or mobilizing the bladder • Advantages: technically simple, quick (typically 15–20 minutes), easily reversible/removable, can usually be performed under regional anesthesia in an outpatient or short-stay setting • Standard first-line choice for the vast majority of both history- and ultrasound-indicated cerclages
Shirodkar cerclage: • The bladder is mobilized anteriorly and the suture is tunneled submucosally, allowing placement closer to the true internal os than McDonald permits, with the vaginal mucosa closed over the knot • Advantages: theoretically superior mechanical support for a very short, effaced, or scarred cervix (e.g., after prior cone biopsy/LEEP) where there is minimal ectocervix available for a McDonald suture • Disadvantages: longer operative time, more technically demanding, and typically requires more extensive dissection to remove (sometimes requiring re-opening the vaginal mucosa)
Abdominal (transabdominal) cerclage: • Placed at the true internal os via laparotomy or laparoscopy, either pre-conception or in early pregnancy • Reserved for patients with a prior failed transvaginal cerclage or an anatomically inadequate cervix (e.g., after radical trachelectomy) • Requires cesarean delivery, since the suture is not accessible/removable transvaginally
The cerclage suture functions primarily as a mechanical band that redistributes the hydrostatic and gravitational forces of the growing uterine contents circumferentially around the cervix, rather than "sealing" the cervix from ascending infection — an important distinction that shapes counseling about residual infection and preterm labor risk even after successful placement.
Standard McDonald procedure sequence: 1. Preoperative confirmation of fetal viability and normal anatomy survey (if not already done), and exclusion of contraindications (labor, infection, ruptured membranes, bleeding, lethal anomaly) 2. Regional anesthesia (spinal most common), dorsal lithotomy positioning 3. Cervix exposed with weighted speculum and retractors; anterior/posterior lip grasped with ring forceps 4. Nonabsorbable suture (5mm Mersilene tape is most common) placed as a purse-string at the cervicovaginal junction with 4 bites, avoiding the bladder anteriorly and rectum posteriorly, entering and exiting the stroma without penetrating the endocervical canal 5. Suture cinched to approximate but not fully close the external os (a fingertip or small probe should still pass, allowing for cervical mucus drainage) and tied, with knot typically placed anteriorly or posteriorly for ease of later identification and removal 6. Suture ends left long enough to be easily grasped at removal
Postoperative care: pelvic rest (no intercourse, nothing per vagina) is commonly recommended though evidence for its benefit is limited; patients are counseled on warning signs of labor, ruptured membranes, and infection requiring prompt evaluation.
Following cerclage placement, serial transvaginal ultrasound continues in many protocols to track cervical length above the suture and to identify funneling or shortening that might indicate the mechanical support is being overwhelmed — findings that inform counseling and, in select circumstances, the decision to add or adjust other interventions such as vaginal progesterone.
A well-placed cerclage typically produces visible stabilization of the measured cervical length compared to the pre-placement trajectory, because the suture itself becomes part of the measured "closed" segment and mechanically resists further internal os dilation. Continued shortening above the suture, new or worsening funneling below the suture level, or the suture becoming visibly loose or displaced on exam are all findings that warrant individualized re-evaluation — options include activity modification, closer surveillance, or in rare cases suture revision, though data guiding these decisions are limited.
Routine digital cervical exams are generally minimized after cerclage placement outside of specific clinical indications, to avoid disrupting the mucus plug or mechanically stressing the suture line.
Elective cerclage removal is planned at 36–37 weeks gestation in an uncomplicated pregnancy, timed to allow the cervix to prepare for labor and vaginal delivery while remaining close enough to term to avoid unnecessary risk. Certain circumstances mandate urgent removal regardless of gestational age, because leaving the suture in place would risk cervical laceration, uterine rupture, or obstructed labor.
Absolute contraindications to placing (or reasons to promptly remove an existing) cerclage include: • Active preterm labor • Clinical chorioamnionitis • Ruptured membranes (PPROM) — the suture no longer serves a protective purpose once the amniotic sac has ruptured, and retention increases infectious risk • Significant vaginal bleeding suggestive of placental abruption or placenta previa with active bleeding • Lethal fetal anomaly or fetal demise • Nonreassuring fetal status where continuing the pregnancy is no longer the goal
In each of these urgent scenarios the suture is removed promptly (sometimes emergently in the operating room if labor is advanced or bleeding is significant) to avoid cervical laceration extending into the lower uterine segment, which can cause major hemorrhage and future obstetric morbidity.
Elective removal at 36–37 weeks is typically a brief office or labor-and-delivery procedure without anesthesia for a McDonald suture; Shirodkar and transabdominal cerclages may require an operative approach depending on how they were originally placed.
A cerclage left in place during active labor or with ruptured membranes is a genuine obstetric emergency risk — the suture must be identified and removed promptly whenever labor, PPROM, bleeding, or infection develops, regardless of gestational age or how close to the planned elective removal date the patient may be.