Background Body composition plays an important role in predicting treatment outcomes in adults with cancer. Using existing computed tomographic (CT) cross-sectional imaging and readily available software, the assessment of skeletal muscle mass to evaluate sarcopenia has become simplified. We performed a systematic review and meta-analysis to quantify the prognostic value of skeletal muscle index (SMI) obtained from cross-sectional CT imaging on clinical outcomes in non-haematologic solid tumours. Methods We searched PubMed and the American Society Clinical Oncology online database of meeting abstracts up to October 2015 for relevant studies. We included studies assessing the prognostic impact of pre-treatment SMI on clinical outcomes in patients with non-haematologic solid tumours. The primary outcome was overall survival (OS) and the secondary outcomes included cancer-specific survival (CSS), disease-free survival (DFS), and progression-free survival (PFS). The summary hazard ratio (HR) and 95% confidence interval (CI) were calculated. Results A total of 7843 patients from 38 studies were included. SMI lower than the cut-off was associated with poor OS (HR = 1.44, 95% CI = 1.32-1.56, p < 0.001). The effect of SMI on OS was observed among various tumour types and across disease stages. Worse CSS was also associated with low SMI (HR = 1.93, 95% CI = 1.38-2.70, p < 0.001) as well as DFS (HR = 1.16, 95% CI = 1.00-1.30, p = 0.014), but not PFS (HR = 1.54, 95% CI = 0.90-2.64, p = 0.117). Conclusions This meta-analysis demonstrates that low SMI at cancer diagnosis is associated with worse survival in patients with solid tumours. Further research into understanding and mitigating the negative effects of sarcopenia in adults with cancer is needed.