BACKGROUND: Although people with haematological malignancies have to endure long phases of therapy and immobility, which is known to diminish their physical performance level, the advice to rest and avoid intensive exercises is still common practice. This recommendation is partly due to the severe anaemia and thrombocytopenia from which many patients suffer. The inability to perform activities of daily living restricts them, diminishes their quality of life and can influence medical therapy. OBJECTIVES: In this update of the original review (published in 2014) our main objective was to re‐evaluate the efficacy, safety and feasibility of aerobic physical exercise for adults suffering from haematological malignancies considering the current state of knowledge. SEARCH METHODS: We searched the Cochrane Central Register of Controlled Trials (CENTRAL) (the Cochrane Library, 2018, Issue 7) and MEDLINE (1950 to July 2018) trials registries (ISRCTN, EU clinical trials register and clinicaltrials.gov) and conference proceedings. We did not apply any language restrictions. Two review authors independently screened search results, disagreements were solved by discussion. SELECTION CRITERIA: We included randomised controlled trials (RCTs) comparing an aerobic physical exercise intervention, intending to improve the oxygen system, in addition to standard care with standard care only for adults suffering from haematological malignancies. We also included studies that evaluated aerobic exercise in addition to strength training. We excluded studies that investigated the effect of training programmes that were composed of yoga, tai chi chuan, qigong or similar types of exercise. We also excluded studies exploring the influence of strength training without additive aerobic exercise as well as studies assessing outcomes without any clinical impact. DATA COLLECTION AND ANALYSIS: Two review authors independently screened search results, extracted data and assessed the quality of trials. We used risk ratios (RRs) for adverse events, mortality and 100‐day survival, standardised mean differences (SMD) for quality of life (QoL), fatigue, and physical performance, and mean differences (MD) for anthropometric measurements. MAIN RESULTS: In this update, nine trials could be added to the nine trials of the first version of the review, thus we included eighteen RCTs involving 1892 participants. Two of these studies (65 participants) did not provide data for our key outcomes (they analysed laboratory values only) and one study (40 patients) could not be included in the meta‐analyses, as results were presented as changes scores only and not as endpoint scores. One trial (17 patients) did not report standard errors and could also not be included in meta‐analyses. The overall potential risk of bias in the included trials is unclear, due to poor reporting. The majority of participants suffered from acute lymphoblastic leukaemia (ALL), acute myeloid leukaemia (AML), malignant lymphoma and multiple myeloma, and eight trials randomised people receiving stem cell transplantation. Mostly, the exercise intervention consisted of various walking intervention programmes with different duration and intensity levels. Our primary endpoint overall survival (OS) was only reported in one of these studies. The study authors found no evidence for a difference between both arms (RR = 0.67; P = 0.112). Six trials (one trial with four arms, analysed as two sub‐studies) reported numbers of deceased participants during the course of the study or during the first 100 to 180 days. For the outcome mortality, there is no evidence for a difference between participants exercising and those in the control group (RR 1.10; 95% CI 0.79 to 1.52; P = 0.59; 1172 participants, low‐certainty evidence). For the following outcomes, higher numbers indicate better outcomes, with 1 being the best result for the standardised mean differences. Eight studies analysed the influence of exercise intervention on QoL. It remains unclear, whether physical exercise improves QoL (SMD 0.11; 95% CI ‐0.03 to 0.24; 1259 participants, low‐certainty evidence). There is also no evidence for a difference for the subscales physical functioning (SMD 0.15; 95% CI ‐0.01 to 0.32; 8 trials, 1329 participants, low‐certainty evidence) and anxiety (SMD 0.03; 95% CI ‐0.30 to 0.36; 6 trials, 445 participants, very low‐certainty evidence). Depression might slightly be improved by exercising (SMD 0.19; 95% CI 0.0 to 0.38; 6 trials, 445 participants, low‐certainty evidence). There is moderate‐certainty evidence that exercise probably improves fatigue (SMD 0.31; 95% CI 0.13 to 0.48; 9 trials, 826 patients). Six trials (435 participants) investigated serious adverse events. We are very uncertain, whether additional exercise leads to more serious adverse events (RR 1.39; 95% CI 0.94 to 2.06), based on very low‐certainty evidence. In addition, we are aware of four ongoing trials. However, none of these trials stated, how many patients they will recruit and when the studies will be completed, thus, potential influence of these trials for the current analyses remains unclear. AUTHORS' CONCLUSIONS: Eighteen, mostly small RCTs did not identify evidence for a difference in terms of mortality. Physical exercise added to standard care might improve fatigue and depression. Currently, there is inconclusive evidence regarding QoL, physical functioning, anxiety and SAEs . We need further trials with more participants and longer follow‐up periods to evaluate the effects of exercise intervention for people suffering from haematological malignancies. To enhance comparability of study data, development and implementation of core sets of measuring devices would be helpful.