Ruxolitinib has proved to be effective for the treatment of patients with myelofibrosis (either primary or secondary) and polycythaemia vera, and its approval led to a significant change in the current treatment algorithm. Despite its efficacy and beyond its well described haematological toxicity, a peculiar immunosuppressive effect emerged as our clinical experience grew, both within and outside of a clinical trial setting. Definite and negative interactions with multiple pathways of the immune system of patients have been reported so far, involving both adaptive and innate immune responses. These pathophysiological mechanisms may contribute to the increased risk of reactivation of silent infections (e.g., tuberculosis, hepatitis B virus and varicella zoster virus) that have been associated with the drug. Even though such infectious events may be fatal or may lead to significant impairment of organ function, compromising the eligibility of patients for an allotransplant procedure, there are no dedicated guidelines that may help us in assessing and managing the risk of developing serious infections. On this basis, our aim for the present work was to review the current knowledge on the pathophysiological mechanisms through which ruxolitinib may exert its immunosuppressive effect, and to illustrate our personal approach to the management of three peculiar clinical scenarios, for which a risk-based algorithm is suggested.