Data from India regarding the disease spectrum and surgical results of neuroendocrine tumours (GEPNETs) are sparse. Tempered surgical radicality in a high-volume oncology centre, conforming to existing guidelines, may further our understanding of tumour characteristics and behavioural patterns of nonfunctional GEPNETs. Surgical outcomes of patients with histopathologically confirmed GEPNETs from January 2003 to December 2013 were analyzed from a prospectively maintained database. Tumour grade, organ of primary tumour, perioperative factors, quality/radicality of resection and presence of metastatic disease were correlated with perioperative outcomes, overall survival and disease-free survival. Ninety of the 101 operated patients had nonfunctional tumours. These comprised radical resections (n = 69), organ-preserving procedures (n = 16) and inoperable tumours (n = 5). The primary tumour sites were pancreatic in 48 patients and gastroenteric in 42 patients. The overall perioperative morbidity and mortality rates were 30 and 3 %, respectively. Fifteen patients harboured metastatic disease at presentation. At a median follow-up of 22 months, 18 patients had residual disease, 7 developed recurrences and 10 patients died. The estimated actuarial 5-year overall survival was 81.6 %, and disease-free survival was 67.2 %. Tumour grade and organ of origin (pancreatic vs. gastroenteric) did not influence long-term survival (p = 0.315 and p = 0.624, respectively), but presence of metastatic disease at presentation significantly affected long-term survival (p = 0.009). Nonfunctional pancreatic/duodenal neuroendocrine tumours (NETs) accounted for 76 % of surgical resections at our centre with the minority being other resections. In selected patients with nonfunctional NETs, organ-preserving surgery may provide equivalent long-term survival with decreased operative morbidity. Although tumour grade is considered to be an important prognostic factor, the presence of metastatic disease at presentation also determines long-term survival. The referral bias suggests the need for greater awareness given the favourable long-term outcomes of these tumours. There is a need to correct this referral bias by increasing the awareness of GEPNETs in India.