Background: The association of the interval between neoadjuvant chemo-radiation and surgery (CRT-S), and cancer outcomes in patients with esophageal cancer is not clear. We aimed to determine the relationship between CRT-S interval and pathological complete response rate (pCR), short and long overall survival (OS). Methods: Patients listed on the National Cancer Data Base from 2004 to 2013 were studied. We included patients with CRT followed by surgery in 15-90 days. All patients had reported pT, pN cancer stages and survival status. CRT-S interval was studied as continuous (weeks) and categorical variables (quintiles). Results: A total of 5181 patients were included; 81% were adenocarcinomas, 84% were males and mean age was 62 years. They were divided into CRT-S interval quintiles (15 to 37, 38 to 45, 46 to 53, 54 to 64 and 65 to 90 days) (n = 1016, 1063, 1081, 1083 and 938 patients), respectively. There was a significant increase of pCR rate across the CRT-S quintiles (18%, 21%, 24%, 25% and 29%, p < 0.001). This advantage persisted when CRT-S was measured as continuous variable in weeks (OR: 1.11, 95% CI = 1.078-1.143, p < 0.001). However, 90-day mortality significantly increased as CRT-S increased across quintiles (5.7%, 6.2%, 6.8%, 8.5% and 8.2%, p = 0.02) and through weeks (OR = 1.05, 95%CI = 1.005-1.106, p = 0.03). Mean OS across CRT-S quintiles was 59.2, 58.8, 55.4, 56.6 and 51.5 months, respectively. Multivariate Cox regression showed significantly worse OS per week increase in CRT-S interval (HR 1.02, 95% 1.003-1.037, p = 0.02), especially among the last quintile (CRT-S = 65-90 days: HR 1.2, 95% CI 1.04-1.32, p = 0.009). Those with no-pCR had worse OS as time to surgery increased (p < 0.001), while pCR group had similar OS across CTR-S intervals. Conclusions: Despite higher pCR rate as CRT-S interval increasing, surgery is preferred to be done in less than 65 days after CRT to avoid worse 90-day mortality and achieve better OS. Further randomized studies are needed to consolidate our findings.