Cervical cerclage — structured SAQ (15 marks)

(a) The three indications and which applies here (4 marks)

One mark for the correct identification (ultrasound-indicated), and one mark each for the trigger, timing and evidence base of the three indications. ^[1][2][3]

This woman meets the ultrasound-indicated criteria squarely — a singleton, a prior spontaneous preterm birth before 34 weeks, and a transvaginal cervical length of 17 mm before 24 weeks. Note that on the Queensland Clinical Guidelines history criterion (one or more prior spontaneous preterm births and/or mid-trimester losses) she would also qualify on history alone; the two routes are not mutually exclusive, and the mark is for recognising that the short cervix in this specific population is where the randomised evidence is strongest. The Berghella meta-analysis in exactly this population reduced preterm birth before 35 weeks from 41.3% to 28.4% (RR 0.70, 95% CI 0.55 to 0.89), with composite perinatal mortality and morbidity down from 24.8% to 15.6% (RR 0.64, 95% CI 0.45 to 0.91).^[1][3]

(b) Three questions and four further contraindications (3 marks)

One mark for the three questions, and half a mark each for any four further contraindications (totalling two marks). ^[1][8]

The three questions before any cerclage: Is there infection? Are the membranes ruptured? Is she contracting? A yes to any one means no cerclage. Phrase all three so that "yes" is the answer that stops you — a candidate who asks "are the membranes intact?" and then says "yes means no cerclage" has just contraindicated every safe case.^[1]

Four further contraindications (any four): active uterine contractions or established preterm labour; clinical chorioamnionitis (fever, maternal or fetal tachycardia, uterine tenderness, offensive discharge); ruptured membranes; active vaginal bleeding or suspected abruption; a lethal fetal anomaly or confirmed fetal demise; gestation beyond the point at which the procedure can plausibly help (around 24 weeks for rescue).^[1][8]

(c) McDonald versus Shirodkar, suture material, and high versus low vaginal cerclage (5 marks)

One mark each for: McDonald technique; Shirodkar technique; the comparative practical point (knot, bladder, removal); C-STICH suture material result; and MAVRIC high versus low vaginal result. ^[1][5][6]

• McDonald (transvaginal, low): purse-string suture at the cervicovaginal junction, as high as practicable, taking four to five bites and avoiding the bladder anteriorly and the rectum posteriorly. Knot tied anteriorly with ends left long for easy removal; the workhorse, usually removed as an outpatient.^[1]
• Shirodkar (transvaginal, high): anterior and posterior colpotomies allow the bladder to be reflected upward so the suture sits at the level of the internal os, with the knot buried. Takes longer, bleeds more, and often needs anaesthesia for removal.^[1]
• Comparative: no proven superiority of Shirodkar over McDonald — the choice is largely operator preference.^[1]
• C-STICH (suture material): monofilament versus braided in 2049 women — pregnancy loss 8.0% versus 7.6% (adjusted RR 1.05, 95% CI 0.79 to 1.40), no significant difference. Use what you are competent with.^[5]
• MAVRIC (high versus low vaginal): in women with a previous failed cerclage, high vaginal cerclage was no better than low vaginal (birth before 32 weeks 38% versus 33%, RR 1.15, 95% CI 0.62 to 2.16) — only transabdominal cerclage was superior in that population.^[6]

(d) PPROM with a suture in situ (3 marks)

One mark for the default (removal), one mark for the latency finding, and one mark for the infection and neonatal finding. ^[7]

• Default: remove the suture, give broad-spectrum antibiotics and steroids as indicated by gestation, and deliver if infection develops. Short-term retention might be considered only at an extremely early gestation with no clinical or biochemical evidence of infection, after a consultant discussion with neonatology and the woman, with an explicit trigger to remove.^[1][7]
• Latency: the Zullo meta-analysis (six studies, 377 women) found retention did prolong pregnancy — the removal group had significantly lower prolongation beyond 48 hours (OR 0.15) and beyond 7 days (OR 0.30).^[7]
• Infection and neonatal cost: retention significantly increased chorioamnionitis (OR 0.57 favouring removal) and a 5-minute Apgar under 7 (OR 0.22 favouring removal). Retention buys latency but costs infection and a worse baby at birth.^[7]