Objective: We sought to evaluate the growth status of twins at birth using individualized growth assessment methods and to compare this assessment with that obtained with conventional methods.
Study design: Twenty twin pregnancies were studied longitudinally with ultrasound. Measurements of the head and abdominal cubes (A,B), head circumference (HC), abdominal circumference (AC), thigh circumference (ThC), and femur diaphysis length (FDL) made in the 2nd trimester were used to specify Rossavik growth models for each parameter in each fetus. These models were used to predict weight (WT), HC, AC, ThC, and crown-head length (CHL) at birth. Actual birth measurements made within 24 hours of delivery were compared to predicted values, the latter corrected using singleton [corrects for both technical problems (TP)] or twin [corrects for both technical problems (TP) and decreased soft-tissue deposition (DSTD)] correction factors where appropriate. Two sets of growth potential realization index (GPRI) values and their corresponding neonatal growth assessment scores (NGAS) were calculated and compared to previously established normal values. Birth measurements were compared with appropriate population age-specific size curves. These data were used to characterize and classify the growth status of each twin neonate.
Results: Individualized growth assessment identified five primary types of growth outcomes: normal (Group I, 45%); primarily DSTD (Group 11,22.5%); IUGR (Group 111, 15%); above average soft-tissue deposition (Group IV, 5%); and growth acceleration (Group V, 7.5%). Within Group I was a subgroup with evidence of DSTD (Group Ib, 33.3% of Group I). Group I1 could be divided into two subgroups, one with only DSTD (Group IIa, 44.4% of Group 11) and one with both DSTD and other growth abnormalities (Group IIb, 55.6% of Group 11). Group I11 had multiple growth abnormalities which were more severe than those seen in Group 11. All normal neonates were AGA and had virtually all anatomic parameters within their respective normal ranges. Of the neonates with definite evidence of IUGR (Groups IIb and III), only 4 of 11 (36.4%) were SGA and only 6 of 11 (54.5%) had any of the five anatomic parameters below their respective normal ranges. Only 1 of 3 (33.3%) of neonates with growth acceleration was LGA and none (0%) of the five anatomic parameters were above their respective normal ranges.
Conclusions: Individualized growth assessment methods provide a more comprehensive assessment of growth outcome in twins and detect a decrease in soft-tissue deposition not identifiable with conventional growth assessment procedures. The latter procedures are also less sensitive in the detection of both IUGR and growth acceleration.