A higher incidence of new-onset diabetes mellitus (NOD) has been consistently observed in people living with HIV (PLHIV) after receiving highly active antiretroviral therapy (HAART). Most NOD in PLHIV occur after use of HAART and both insulin resistance and impaired β-cell insulin secretion have been found to underlie the disturbed glucose metabolism. The use of glycated hemoglobin (HbA1c) to diagnose or screen DM has been noted to significantly underestimate glycemia status and to have the least yield when compared to using standard oral glucose tolerance test (OGTT) or fasting glucose levels, and should be interpreted with caution. Low hemoglobin values, or hemolysis that causes shortened erythrocyte lifespan have been associated with lower HbA1c values. For better yield of making earlier diagnosis, the conjunction of HbA1c with either fasting plasma glucose levels or standard OGTT is thus recommended in HIV-infected patients. Treatment of DM according to the contemporary guidelines is practical but caution must be taken when there could exist HAART-associated mitochondrial toxicity which may result in buildup of lactic acid, in which case metformin use is contraindicated. HAART-induced hepatotoxicity should be evaluated before prescription of thiazolidinedione. The DPP4 inhibitors have a safe profile regarding immune or virological status, but the dose needs to be reduced when co-administered with a CYP3A4/5 inhibitor (such as protease inhibitors) due to altered plasma concentration profile of the gliptins. With increasing incidence of DM among PLHIV receiving HAART, clinicians are encouraged to keep high profile of clinical vigilance to ensure an appropriate management for patients who may suffer from a new disease while having been treated for an existing one. An understanding of the underlying pathophysiological processes of glucose dysmetabolism linked to antiretroviral therapy will help clinicians treat their patients in a more competent manner.