Ratio = forward ÷ reverse pulmonary vein VTI. Lower means more left atrial hypertension. Studies define “restrictive” differently: each row says whose cut-off it is.
| Ratio | What it predicted | Source |
|---|---|---|
| ≤2.7 | Severe illness within 2 h of birth: sensitivity 89%, specificity 97%, AUC 0.93. Also predicted emergent septal intervention and death or transplant 68 fetuses with critical left heart obstruction (52 HLHS, 14 critical aortic stenosis) already called restrictive or intact; last fetal scan, median 35.9 weeks. Emergent catheter intervention: sensitivity 81%, specificity 97%. Severe illness: arterial pH <7.15, venous pH <7.10, bicarbonate <16 mmol/L, lactate >5 mmol/L or median saturation <60% in the first 2 h. Ratio >2.7 with a pulmonary vein diameter ≤3.8 mm: 3.8% had it. Mid-gestation scans: best cut-off <3 (94%, 81%). | Gellis 2018 |
| ≤3 | Emergency atrial septostomy: no false positives (≤5 had 2), sensitivity unchanged. Late second-trimester values stayed on the same side of the cut-off in 30 of 32 39 fetuses with critical left heart obstruction; 8 had a ratio ≤5 and 6 of them had a restrictive septum confirmed after birth, so the ≤3 result rests on few events. | Divanović 2011 |
| <3 | Guideline cut-off for needing the septum opened urgently after birth. Missed some in one cohort: sensitivity 60%, specificity 100% (Mardy 2021) | AHA 2014; ASE 2023, Table 13 |
| <5 | Emergency atrial septoplasty: sensitivity 88%, specificity 97%, AUC 0.98 41 fetuses, 8 septoplasties, scanned at a mean of 33 weeks. Median ratio 3.4 with septoplasty vs 13.6 without. The text gives the specificity as 98%; the abstract and Table 3 give 97%. | Michelfelder 2005, Table 3 |
| <5.1 | Any septal procedure before stage 1: sensitivity 100%, specificity 91%, AUC 0.98. Did not reliably predict 2-year survival | Lowenthal 2012 |
AHA 2014, Class IIa (LOE B/C): with a restrictive or intact septum, planning specialized delivery room care is reasonable when the ratio is <3 or the third-trimester hyperoxia test is abnormal.
| Pattern | Doppler, and the newborns | Source |
|---|---|---|
| A | Continuous forward flow, small atrial reversal (reverse ÷ forward <0.18). Postnatal foramen 4.5 ± 2.1 mm | Taketazu 2004 |
| B | Continuous forward flow, larger atrial reversal (≥0.18). Postnatal foramen 1.6 ± 1.6 mm | Taketazu 2004 |
| C | Brief to-and-fro flow, little or no diastolic flow. All 3 live-born had an intact septum and needed emergent septoplasty 40 fetuses with spectra; 6 had pattern C, 3 of them live-born. Operative survival 91% (A), 100% (B), 33% (C). Pattern C veins were dilated (vein:PA diameter 1.53 vs 0.73–0.76). One pattern B fetus also had an intact septum, decompressed by a vein found after birth. Measuring the foramen itself in utero did not predict postnatal size, blood gases or the need for septoplasty. | Taketazu 2004 |
| Finding | What is reported | Source |
|---|---|---|
| How often | 12 of 195 HLHS fetuses; 22% of those with a ratio <5 | Campbell 2026 |
| Ratio still counts | With a vein: ratio <3, 3 of 4 died; ≥3, 6 of 8 alive without transplant at 1 year. Once the ratio was known, the vein itself was not independently associated with outcome (12 with a vein; wide confidence interval). Journal pre-proof | Campbell 2026 |
| Measure | Abnormal result and what it predicted | Source |
|---|---|---|
| Branch PA pulsatility index | Not vasoreactive: falls <10% at every site (last scan after 31 weeks); predicted immediate septal intervention, sensitivity 100%, specificity 94%, PPV 71%, NPV 100% 43 HLHS, 9 restrictive, 5 needed intervention, 2 false positives. | Szwast 2010 |
| Pulmonary vein ratio | Ratio <6.5 on oxygen: emergent septoplasty, sensitivity and specificity 100%. Baseline 3–7: a ratio that falls on oxygen separated the groups 27 fetuses, 5 septoplasties, tested at about 34 weeks. In this cohort the PA pulsatility index performed poorly (AUC 0.32). | Mardy 2021 |
ASE 2023: may be considered in select cases in the late third trimester.
| Finding | What is reported | Source |
|---|---|---|
| Left atrial size | Ratio ≤5 (54 fetuses): hypoplastic LA in 61%, complex septum or a ridge across the LA 19%, near-normal LA 17% (never intact), dilated LA with severe MR 4% | Dannesbo 2026 |
| Lymphangiectasia on fetal MRI | Seen with every septal type; went with a lower ratio | Dannesbo 2026 |
| Nutmeg lung, outcome | 4 of 44 HLHS fetal MRIs; all 4 died by 5 months (35% death or transplant without it) | Saul 2016 |
| Intact septum after birth | Small LA with a thick muscular septum: 4 of 18, all with obstructed decompression and the worst lung histology | Rychik 1999 |
| Item | What is said or reported | Source |
|---|---|---|
| Level of care | Level 4: caesarean in a cardiac facility, usually at 38–39 weeks, cardiac team in the delivery room, plan for catheterization, surgery or ECMO. HLHS without restriction is level 2 | AHA 2014, Table 19 |
| Recommendation | Specialized delivery room planning with a ratio <3 or an abnormal third-trimester hyperoxia test: Class IIa LOE B/C | AHA 2014 |
| Fetal echo as the screen | High risk (ratio <3 near term, thick septum, foramen <3 mm) for intervention, ECMO or death within 2 h: sensitivity 100%, specificity 95%, PPV 71%, NPV 100%. Planned cardiac operating-room delivery: 4 of 5 survived, 0 of 2 born elsewhere 52 HLHS, 7 high risk. None of the 45 low-risk HLHS born at the maternity hospital needed intervention before stage 1. An older series: 2 of 7 survived despite planned caesarean and immediate intervention (Glatz 2007). | Moray 2024 |
| Emergent intervention within 24 h | Restriction before 30 weeks, 11 of 13; from 30 weeks, 5 of 7; none, 4 of 174 Restriction = intact septum, a decompressing vein or a ratio <3. | Kikano 2026 |
Restriction can progress: 8 of 18 with comparable fetal and postnatal images were worse after birth than on the first fetal scan (Glatz 2007).
| Cohort | With restriction vs without | Source |
|---|---|---|
| Prenatal, any restriction | 2-year survival 43% vs 83%, intention to treat | Lowenthal 2012 |
| Left heart obstruction, ratio ≤3 | Survival to discharge 42% vs 81%; excess deaths mostly before stage 2 | Schneider 2020 |
| Intact or <2 mm on colour | 30-day survival 38% (intact septum with mitral or aortic atresia) vs 94% (lesser obstruction) | Glatz 2007 |
| Septostomy in the first 2 days | Mortality 48% before and soon after surgery, despite a successful septostomy | Vlahos 2004 |
| Multicentre, stage 1 | Septal procedure before stage 1 vs no restriction: operative mortality adjusted OR 2.01, 1-year 2.2 (3,644 infants) | Lin 2027 |
| Timing of onset | 30-day mortality after stage 1: 25% (before 30 weeks), 29% (from 30 weeks), 6% (none), p 0.007. Overall postoperative mortality 58%, 29%, 40%, not significant 196 HLHS. Postoperative mortality is from the paper's Table 2 (7 of 12 before 30 weeks); its abstract gives 62% for that group. | Kikano 2026 |
| Evidence | What is said or reported | Source |
|---|---|---|
| Recommendation | May be considered for a severely restrictive or intact septum: Class IIb LOE B/C | AHA 2014 |
| Pooled studies | 746 fetuses: survival to discharge 37% with vs 52% without (no difference); restrictive septum at birth 38% vs 88% | Mustafa 2024 |
| Registry | 47 procedures: 13% procedure-related fetal death; discharge survival 34% with vs 36% without | Jantzen 2017 |
| Single centre | Technically successful: survival 72% vs 21% for all others 42 fetuses with HLHS or a variant, after 18 judged extremely high risk were excluded. By intention to treat, survival was 52% with intervention vs 29% without (p 0.147); unsuccessful attempts included 5 peri-procedural deaths. Nutmeg lung improved in 8 of 9 re-imaged. Their criteria: 28 to <32 weeks, ratio ≤3 with to-and-fro or low diastolic flow, LA not severely hypoplastic. | Yilmaz Furtun 2025 |
| Defect size | Created defects ≥3 mm went with higher saturation and fewer procedures before surgery | Marshall 2008 |
Fetal HLHS and variants: predicting a restrictive or intact atrial septum, and what it means at birth