An injection of hydrogel-encapsulated islets that controls blood glucose levels over long term would provide a much needed alternative treatment for type 1 diabetes mellitus (T1DM). To this end, we tested the feasibility of using an injectable polyethylene glycol (PEG) hydrogel as a scaffold for islet encapsulation. Encapsulated islets cultured in vitro for 6 days showed excellent cell viability and released insulin with higher basal and stimulated insulin secretion than control islets. Host responses to PEG hydrogels were studied by injecting PEG hydrogels (no treatment and vehicle controls used) into the peritoneal cavities of B6D2F1 mice and monitoring alterations in body weight, food and water intake, and blood glucose levels. After 2 weeks, peritoneal cavity cells were harvested, followed by hydrogel retrieval, and extraction of spleens. Body weights, food and water intake, and blood glucose levels were unaltered in mice injected with hydrogels compared to no treatment and vehicle-injected control mice. Frozen sections of a hydrogel showed the presence of tissues and small number of immune cells surrounding the hydrogel but no cell infiltration into the hydrogel bulk. Spleen sizes were not significantly different under the experimental conditions. Peritoneal cavity cells were slightly higher in mice injected with hydrogels compared to control mice but no statistical difference between vehicle- and hydrogel-injected mice was noted. As an in vivo feasibility study, streptozotocin-induced diabetic mice were injected with vehicle or hydrogels containing 50 islets each into two sites, the peritoneal cavity and a subcutaneous site on the back. Transient control of blood glucose levels were observed in mice injected with hydrogels containing islets. In summary, we developed an injectable PEG hydrogel that supported islet function and survival in vitro and in vivo and elicited only a mild host response. Our work illustrates the feasibility of using injectable PEG hydrogels for islet encapsulation.