Objective To explore the correlation between MR findings [optic nerve sheath diameter (ONSD) and peripheral cerebrospinal fluid volume/intracranial volume ratio (PCSFV/ICV)] and cerebrospinal fluid pressure (CSFP) in patients with leptomeningeal metastases (LM) from non-small-cell lung cancer (NSCLC). And to assess the diagnostic accuracy of combined ONSD and PCSFV/ICV in diagnosing elevated CSFP. Methods: A total of 63 patients with NSCLC-LM were enrolled. Both optic nerve and cranial MRI scans were performed to measure bilateral ONSD, cranial PCSFV and ICV. Subsequently, CSFP was measured in all patients through lumbar puncture, and intrathecal chemotherapy was administered regularly. Pearson correlation analysis was used to analyze the correlation between CSFP and MRI findings. The accuracy of combined ONSD and PCSFV/ICV in diagnosing elevated CSFP was evaluated through receiver operating characteristic (ROC) curve analysis. Results: In patients with NSCLC-LM, the average ONSD of bilateral eyes was positively correlated with CSFP (r = 0.567, p<0.001). PCSFV/ICV was negatively correlated with CSFP (r = -0.365, p=0.003). ROC curve analysis showed that the performance of ONSD, PCSFV/ICV, and combined ONSD with PCSFV/ICV in diagnosing intracranial hypertension (CSFP>15mmHg) were 0.8099 (95% confidence interval: 0.71–0.91), 0.7617 (95% confidence interval: 0.64–0.89), and 0.9136 (95% confidence interval: 0.84–0.99), respectively. The performance of ONSD, PCSFV/ICV, and combined ONSD with PCSFV/ICV in diagnosing intracranial hypertension (CSFP>20mmHg) were 0.6985 (95% confidence interval: 0.58–0.82), 0.7239 (95% confidence interval: 0.57–0.88), and 0.8137 (95% confidence interval: 0.70–0.93), respectively. Conclusion: The utilization of MRI for measuring ONSD and PCSFV/ICV provides a valuable means to assess variations in CSFP among patients diagnosed with NSCLC-LM. Additionally, the combination of these two measurements enhances the diagnostic accuracy for evaluating CSFP, demonstrating its potential clinical significance.