We have reviewed the literature of immune therapy in autoimmune encephalitis associated with antibodies to cell surface antigens including N-methyl-D-aspartate receptor (NMDAR), leucine-rich, glioma-inactivated protein-1 (LGI1), contactin-associated protein-2 (Caspr2), the α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor (AMPAR), γ-aminobutyric acid-A receptor (GABAAR), γ-aminobutyric acid-B receptor (GABABR), Glycine R and other rarer antigens. Most studies are retrospective cohorts, and there are no randomised controlled trials. Most clinicians use first-line therapy (steroids, intravenous immunoglobulin, plasma exchange), and if severe or refractory, second-line therapy (rituximab, cyclophosphamide). When present, tumours should be removed. There are common therapeutic themes emerging. Firstly, patients given immune therapy do better and relapse less than patients given no treatment. Secondly, patients given early treatment do better. And thirdly, when patients fail first-line therapy, second-line therapy improves outcomes and reduces relapses. Given the retrospective uncontrolled data, the literature has inherent bias, including severity and reporting bias.