Parenteral nutrition duration (PND) of 14 days or more was found to increase the risk of retinopathy of prematurity (ROP) in infants born in Sweden, according to a new study published in JAMA Ophthalmology. The researchers suggest considering updated Digital ROP (DIGIROP) 2.0 models to improve management of ROP.

ROP is a major cause of visual impairment in children. Low gestational age (GA), low birth weight, low early serum insulinlike growth factor-1 (IGF-1), poor early weight gain, and fluctuating oxygen concentration are all risk factors for ROP. This study aimed to assess the prognostic value of PND, or the length of time an infant is fed intravenously instead of by mouth, on ROP.

Participants came from the Swedish National Registry for ROP, which collects information for infants who were born prematurely and examined for ROP, and data were extracted until December 31, 2015. All participants were infants born from 2007 to 2020. Infants were excluded if they had missing data on birth weight. Researchers obtained medical records to collect data on weekly weights. Postnatal age, postmenstrual age, and GA were defined by the American Academy of Pediatrics.

Outcomes that were related to PND were any ROP and ROP treatment. PND, which is administered as early as possible after birth, was defined as number of days that the patient had parenteral protein and lipid supplementation. Predictors used in DIGIROP included GA, sex, birth weight, PND, and other important interactions.

There were 11,139 infants who were included in the study; 45.5% were girls, the mean (SD) GA was 28.5 (2.4) weeks, and the mean (SD) birth weight was 1172 (384) g. Any ROP was found in 3179 infants (28.5%) and the median (range) amount of time to first diagnosis of ROP was 8.4 (0.9-24.7) weeks. There were also 599 infants (5.4%) who received treatment for ROP, with a median (range) time to first ROP treatment of 12.6 (6.3-28.3) weeks.

Infants had a mean (SD) of 10.8 (16.7) days of PND in the overall cohort, with 7228 infants (64.9%) receiving PND for less than 14 days, 2308 (20.7%) receiving PND for 14 days or more, and 1603 (14.4%) having an unknown PND.

Infants who had PND for 14 days or more had faster ROP treatment compared with infants who had less than 14 days (median [range], 3.0 [0-19.1] vs 4.3 [0.1-15.9] weeks; GA-adjusted mean difference, –0.9; 95% CI, –1.5 to –0.3). PND for 14 days or more was also associated with higher odds ROP at any stage (adjusted odds ratio [aOR] per 1-week increase, 1.16; 95% CI, 1.13-1.20) after adjusting for GA, birth weight, and sex.

PND for 14 or more days had an aOR of 1.84 (95% CI, 1.62-2.10) for any ROP compared with PND of less than 14 days. PND was also found to be associated with treatment for ROP (aOR per 1-week increase, 1.12; 95% CI, 1.09-1.15). Infants who had PND for 14 days or more also had significantly more treatment for ROP compared with infants who had less than 14 days (aOR, 2.20; 95% CI, 1.73-2.80).

The study authors noted that editing errors could exist in the register data. The Swedish population is also homogenous in ethnicity, neonatal care, and socioeconomic status, which means the models for DIGROP need to be validated in other populations.

The researchers concluded that PND holds prognostic value for any ROP and for ROP that needed treatment. Infants who needed 14 or more days of PND had a higher risk of needing treatment for ROP.


Pivodic A, Holmström G, Smith LEH, et al. Prognostic value of parenteral nutrition duration on risk of retinopathy of prematurity: development and validation of the revised DIGIROP clinical decision support tool. JAMA Ophthalmol. Published online June 29, 2023. doi:10.1001/jamaophthalmol.2023.2336