Objective

To study the relationship between neonatal morbidity (NNM) and two‐year neurodevelopmental impairment (NDI) in surviving children after early fetal growth restriction (FGR).
Design

Secondary analysis of a European randomised trial (TRUFFLE) of delivery for very preterm fetuses dependent on venous Doppler or cardiotocographic criteria.
Setting

Tertiary perinatal centres, participants in TRUFFLE.
Population

402 surviving children after early FGR.
Methods

Prospective data were collection from the recognition of FGR until the corrected age of two years. We studied the association between NNM and NDI, retaining trial allocation in all statistical models. NNM included any of bronchopulmonary dysplasia, brain injury, sepsis or necrotising enterocolitis. NDI was a composite of Bayley cognitive score < 85, cerebral palsy or severe sensory impairment.
Main outcome measure

NDI in relation to NNM.
Results

NNM occurred in 104 cases (26%) and was more frequent in 17 of 39 infants with NDI (44%) than in the 87 of 363 infants with normal outcome (24%) [odds ratio 2.5 (95% CI, 1.3–4.8); P = 0.01]. In 22 of 39 NDI cases (56%) there was no preceding NNM. NNM was inversely related to gestational age, but NDI did not vary by gestational age. In multivariable analyses, cerebral ultrasound abnormalities were most strongly associated with NDI, together with trial group allocation, birthweight ratio, infant sex and Apgar score.
Conclusions

With the exception of cerebral ultrasound abnormalities, commonly used NNMs are poor markers of later NDI and should not be used as surrogate outcomes for NDI.