<p>This study investigated the relationship between protein and energy intake during the early (days 1–2) and late (days 3–7) acute phase and prognosis in septic patients with nutritional risk screening 2002 ≥ 3. Cox regression analysis was performed to identify factors associated with 28-day mortality. Among 293 shock patients, higher energy intake in days 3–7 was signifcantly associated with lower 28-day mortality (hazard ratio, 0.969; 95% confidence interval, 0.942–0.997; <i>P</i> = 0.033). Kaplan-Meier analysis showed achieving 75% of energy target in days 3–7 correlated with decreased 28-day mortality (<i>P</i> = 0.040). In 309 non-shock patients, no significant relationships between nutritional therapy during acute phase and 28-day mortality were observed. Enhanced energy intake in days 3–7 showed no correlation with improvements in length of intensive care unit/hospital stay, ventilator-free days, non-ICU survival days or hospitalization costs for septic patients. In shock patients, energy intake in days 3–7 demonstrated no relationship with secondary infections; however, among non-shock patients, achieving 50% of energy intake in days 3–7 was linked to lower incidence of fungus infections (<i>P</i> = 0.006) and intra-abdominal infections(<i>P</i> = 0.003). Our findings demonstrate statistical associations between augmenting energy intake during the late acute phase and potential benefits in sepsis.</p>