Twin Growth Discordance Calculator

Percentage discordance with the LARGER twin as the denominator, which is what the published sources print – and the mean-denominator version alongside it, because the same pair reads 25.00% one way and 28.57% the other.

Last updated 8 Oct 2026

Percentage weight discordance

Denominator: the larger twin

First twin, weight or estimated weight (g)Either birth weight or estimated fetal weight; the arithmetic is the same and the thresholds have been derived on both. Order does not matter: the page sorts the two, so entering the smaller first gives the same answer rather than a negative percentage.

Second twin, weight or estimated weight (g)Where these are estimated rather than birth weights, each estimate carries its own error and a discordance inherits both: the 95% interval on a single ultrasound weight estimate exceeds 14% of birth weight for every method examined in one systematic review.

ChorionicityThis changes what the question is, not the arithmetic. Discordance in dichorionic twins raises a growth question; in monochorionic twins it raises twin-twin transfusion syndrome and selective growth restriction from unequal placental sharing, which are staged by umbilical artery Doppler rather than defined by a percentage.

22.00per centExample

Twins of 2,000 g and 1,560 g, dichorionic

The denominator, and what changes when you change it

Discordance = (larger − smaller) ÷ larger × 100
Over the mean: (larger − smaller) ÷ ((larger + smaller) ÷ 2) × 100 · over the smaller: (larger − smaller) ÷ smaller × 100

the larger twin is the denominator: three independent sources print it that way: Patterson and Wood 1990, D’Antonio and colleagues’ review of 22 studies and 10,877 pregnancies, and Domingues and colleagues’ series of 1,132 twins. Blickstein and Kalish’s review of the definitions describes no mean-based denominator for a twin pair at all
but the denominator changes the answer: at 2,000 g and 1,500 g: 25.00% over the larger, 28.57% over the mean, 33.33% over the smaller. The mean convention reads 3.57 percentage points higher and the smaller convention 8.33 points higher. The three agree ONLY where the twins are equal, and then all three are zero. A 2,000 g and 1,540 g pair is 23.00% over the larger and 25.99% over the mean, so it fails the 25% consensus threshold on one convention and meets it on the other
the page is symmetric in the two weights: it sorts them internally, so entering the smaller twin first gives the same answer. Written as (first − second) ÷ first it would return a NEGATIVE percentage half the time, and that is the natural way to build this page wrongly
a percentage ignores absolute size: Blickstein and Kalish’s point: 20% describes a 1,500 g and 1,200 g pair and a 3,000 g and 2,400 g pair equally. They also note discordance is not Gaussian, so a mean and standard deviation is inaccurate and only percentiles should be used

Worked example

Twins of 2,000 g and 1,560 g, dichorionic
Larger 2,000 g, smaller 1,560 g, difference 440 g
Over the larger: 440 ÷ 2000 × 100 = 22.00%
Over the mean: 440 ÷ 1780 × 100 = 24.72%, which is 2.72 percentage points higher for the same pair
Over the smaller: 440 ÷ 1560 × 100 = 28.21%
Enter the twins the other way round and the answer is still 22.00% — the page sorts them, so it cannot return a negative percentage
At 2,000 g and 1,500 g the three conventions read 25.00%, 28.57% and 33.33%; at 2,000 g and 1,540 g they read 23.00% and 25.99%, so that pair fails the 25% consensus threshold one way and meets it the other. The three agree only when the twins weigh the same, and then all three are zero

The same pair under three denominators

The implemented convention is in bold. The third row is the case that matters: a pair below the 25% consensus figure on the larger-twin convention and above it on the mean. A percentage quoted without its denominator cannot be compared with a threshold.
Pair	Over the larger	Over the mean
2,000 g and 1,500 g	25.00%	28.57%
2,000 g and 1,560 g	22.00%	24.72%
2,000 g and 1,540 g	23.00%	25.99% — crosses 25% where the other does not
3,000 g and 2,400 g	20.00%	22.22%
2,500 g and 2,500 g	0.00%	0.00% — the only point all three agree

Published thresholds and what each predicted

Three thresholds, no consensus, and a dichorionic odds ratio series that is lower at 20% than at 15%. That non-monotonicity is in the published review and is printed rather than smoothed; it is a reminder that these are pooled estimates over heterogeneous retrospective series.
Threshold	Finding	Source and cohort
Above 18%	Morbidity and mortality rose; hazard ratio 2.2 dichorionic, 2.6 monochorionic without transfusion syndrome, with risk roughly doubling even where both twins were appropriately grown	Breathnach and colleagues, 1,028 unselected pairs, 977 analysed
15 to 20%	Described as significant; 20% used in the series	Domingues and colleagues, 1,132 twins
25% or more	One of three contributory parameters in the consensus definition of selective fetal growth restriction in dichorionic twins	Delphi consensus, as quoted in a 2023 systematic review
Intrauterine death, dichorionic	Odds ratios 9.8 at 15%, 7.0 at 20%, 17.4 at 25%, 22.9 at 30% — not a monotone series	D’Antonio and colleagues, 22 studies, 10,877 pregnancies
Intrauterine death, monochorionic	Odds ratios 2.8 at 20% and 3.2 at 25%, transfusion syndrome excluded	Same review

The denominator, and why the number alone is not enough

Percentage discordance is the difference between the twins divided by the larger of them. Three independent sources print that convention: Patterson and Wood’s 1990 paper, whose title is the question, D’Antonio and colleagues’ review of 22 studies and 10,877 twin pregnancies, and Domingues and colleagues’ series of 1,132 twins. Blickstein and Kalish’s review of the definitions says the percent definition is calculated as a percentage of the larger infant and describes no mean-based denominator for a twin pair; the only average it proposes is for the middle triplet.

But the denominator is not cosmetic. At 2,000 g and 1,500 g the same pair reads 25.00% over the larger twin, 28.57% over the mean and 33.33% over the smaller — 3.57 and 8.33 percentage points higher. The three agree only when the twins weigh the same, and then all three are zero. A 2,000 g and 1,540 g pair is 23.00% on the implemented convention and 25.99% over the mean, so it fails the 25% consensus threshold one way and meets it the other. That is why all three are computed above: a percentage quoted without its denominator cannot be compared with a threshold. The page is also symmetric in the two weights, because the obvious implementation, first minus second over first, returns a negative percentage half the time.

Three published thresholds and no consensus between them. Breathnach and colleagues followed 1,028 unselected twin pairs and found morbidity and mortality rising above 18%, hazard ratio 2.2 in dichorionic pairs and 2.6 in monochorionic pairs without twin-twin transfusion syndrome, with the risk roughly doubling even where both twins were appropriately grown. Domingues and colleagues describe 15 to 20% as significant and say plainly that there is no consensus. And 25% or more is one of three contributory parameters in the Delphi consensus definition of selective growth restriction in dichorionic twins, where two of the three are needed, so discordance alone does not meet it. Chorionicity changes the question rather than the arithmetic: in monochorionic twins discordance raises transfusion syndrome and selective growth restriction from unequal placental sharing, staged by umbilical artery Doppler in the smaller twin rather than defined by a percentage. Where these are estimated weights each carries its own error, and the weight estimation and centile pages set out how large that is. This page computes a number, or applies published criteria, and names the body that publishes them. It renders no clinical decision: nothing here says whether this fetus is compromised, whether a tracing should be acted on, whether a scan should be repeated, whether labour should be induced or when a pregnancy should be delivered. An obstetric measurement is read alongside the whole clinical picture – history, gestational age, serial trend, the other biometry – and never instead of it. It supports a clinician’s judgement rather than replacing it. Thresholds here are the ones their named sources print. Reference standards differ between guidelines, a finding can cross a threshold purely by changing standard, and the cut-off your own unit works to takes precedence over anything on this page.

Frequently asked questions

Is the denominator the larger twin or the mean?

The larger twin, in every published source read for this page: Patterson and Wood 1990, D’Antonio and colleagues’ review of 22 studies, and Domingues and colleagues’ series of 1,132 twins. Blickstein and Kalish’s review of the definitions describes no mean-based denominator for a twin pair. The other two are computed above anyway, because a percentage quoted elsewhere may have been calculated either way.

How much difference does the convention make?

At 2,000 g and 1,500 g: 25.00% over the larger twin, 28.57% over the mean, 33.33% over the smaller. The mean convention reads 3.57 percentage points higher and the smaller convention 8.33 points higher. They coincide only when the twins weigh the same. A 2,000 g and 1,540 g pair reads 23.00% and 25.99%, which straddles the 25% consensus figure.

Which threshold should be used?

There is no consensus, and the page prints all three with their sources: 18% from Breathnach and colleagues’ 1,028-pair cohort, 15 to 20% from Domingues and colleagues’ series, which used 20%, and 25% or more as one of three contributory parameters in the Delphi consensus definition. Use the threshold your own unit works to.

Does chorionicity change the calculation?

Not the arithmetic, but it changes what the number means. In dichorionic twins discordance is a growth question and the published thresholds were mostly derived there. In monochorionic twins it raises transfusion syndrome and selective growth restriction from unequal placental sharing, staged by umbilical artery Doppler in the smaller twin. D’Antonio and colleagues found intrauterine death odds ratios far larger in dichorionic twins at the same threshold.

Does 25% discordance mean selective growth restriction?

No. In the Delphi consensus definition for dichorionic twins, discordance of 25% or more is one of three contributory parameters, and at least two of the three are needed; the other two are one twin’s estimated weight below the 10th centile and an umbilical artery pulsatility index in the smaller twin above the 95th centile. Alternatively the solitary parameter, one twin’s estimated weight below the 3rd centile, meets the definition on its own. Discordance alone does not.

Related calculators

References

Patterson RM, Wood RC. What is twin birthweight discordance? Am J Perinatol. 1990;7(3):217-19.
D’Antonio F, Khalil A, Dias T, Thilaganathan B. Weight discordance and perinatal mortality in twin pregnancy: systematic review and meta-analysis. Ultrasound Obstet Gynecol. 2017.
Breathnach FM, McAuliffe FM, Geary M, et al. Definition of intertwin birth weight discordance. Obstet Gynecol. 2011;118(1):94-103.
Blickstein I, Kalish RB. Birthweight discordance in multiple pregnancy. Twin Res. 2003;6(6):526-31.
Domingues AP, Fonseca E, Vasco E, et al. Birth weight discordant twins have increased prenatal mortality and neonatal morbidity: an analysis of 1,132 twins. J Pediatr Neonat Individual Med. 2015;4(1):e040113.
D’Antonio F, Prasad S, Masciullo L, Eltaweel N, Khalil A. Selective fetal growth restriction in dichorionic diamniotic twin pregnancies: systematic review and meta-analysis. Ultrasound Obstet Gynecol. 2023.

Not medical advice. For healthcare professionals and education. Reference intervals vary by laboratory and assay — always use your own laboratory's. Never base a dose or a treatment decision on this page alone. Full disclaimer at calcengines.com/disclaimer/