Objectives: To quantify the increase in small for gestational age (SGA) neonates by using Tacna-Peru intrauterine growth curves and to study associated risk factors. Material and methods: From 1995 through 2000 15,739 neonates were studied at the Hipolito Unanue Tacna Hospital by applying both Lubchenco and Tacna-Peru intrauterine growth curves; the additional SGA neonates identified by the Peruvian curves were classified as "new SGAs", and their risk factors were studied. The Perinatal Computer System, odds ratio and chi square test were used in the statistical analysis. Results: When Lubchenco's and the Peruvian curves were applied, the percentage of large for gestational age (LGA) neonates decreased from 15.6 to 9.5%, that of adequate for gestational age (AGE) neonates was similar at 83.2 and 81.3% and that of SGA neonates increased from 1.2 to 9.2%; these changes were statistically significant. The Tacna-Peru curves identified 1,448 SGA neonates, while Lubchenco's curves identified 189. The difference of 1,259 neonates (8% of the total population) corresponded to the group of "new SGA neonates". Risk factors were multiple pregnancy, heart disease, hypertensive disease of pregnancy, primiparity, and urinary infection; risks included increased morbidity, nutritional metabolic abnormalities, infections, congenital defects, moderate depression, and neonatal mortality. Conclusions: The results of the present study confirm that Lubchenco curves overestimate the number of LGA neonates and underestimate that of SGA neonates and are not suitable for use in Peruvian neonates. The new SGA group showed high morbidity and mortality. We recommend that perinatologists design and use curves specific to their populations.