Reoperative insulinomas, 1927 to 1992: an institutional experience

Between 1927 and 1992, 313 patients have undergone surgery for insulinoma(s) at our institution, 12% of which were reoperations. We retrospectively reviewed all cases (n = 39) of reoperative surgery for persistent hyperinsulinism to identify changing patterns in surgical approach, morbidity, and outcome and to evaluate the influence of preoperative and intraoperative localization studies.

The diagnosis of endogenous hyperinsulinism has evolved from satisfying Whipple's triad to documenting concomitant hypoglycemia and endogenous hyperinsulinemia. Thirty-nine patients were divided into two groups for comparison: those treated before localization studies were available (1927 to 1967) (n = 17) and those treated since that time (n = 22). Initial operations were also compared with reoperations among these 39 patients.

There were 26 women and 13 men (mean age, 42 years). There was at least one positive preoperative localization study in 16 of 22 patients (73%). Intraoperative ultrasonogram and careful palpation successfully identified 10 of 11 tumors in the reoperative setting. Blind or completion pancreatectomies were common before 1967 (10 of 17 patients). Since 1967, 14 of 22 patients have undergone enucleation of their primary tumor. Operative morbidity increased from 21% to 58% with reoperation but decreased from 65% to 29% (p = 0.026) when comparing the preangiography to the postangiography eras. The development of iatrogenic diabetes mellitus occurred in 13 patients (33%) after reoperation. Forty-one percent had diabetes before 1967 and 27% since that time (p = not significant). Thirty-four of 36 patients (94%) without malignant tumors were cured by reoperation. There was one operative death. Survival for patients who underwent completion or total pancreatectomy was significantly reduced (p = 0.003).

Reoperations for persistent hyperinsulinism can be highly successful in experienced hands. These reoperations, however, are associated with increased morbidity and iatrogenic diabetes. With experience and the use of selected localization studies, first-time failures can be avoided in most cases thus reducing the need for reoperations and its inherent sequelae.