Intravenous 0.9 mg/kg recombinant tissue plasminogen activator (r-tPA) is one of the most effective treatments in acute ischemic stroke patients. Practically, the dose of r-tPA is still a topic that is constantly being discussed. For this observational study, data were obtained from 537 patients who received r-tPA thrombolysis at Shanghai Sixth People's Hospital stroke center over 5 years (2014-2019). Patients were divided into two groups: a non-standard dose group (0.6 mg/kg ≤ dose < 0.9 mg/kg) and a standard dose group (0.9 mg/kg). Different outcomes were observed: efficacy: 3 months mRS 0-1 (3m-mRS0-1); safety: symptomatic intracranial hemorrhage within 24 h (24h-sICH) and 3 months mortality (3m-death). We also observed the effect of r-tPA dose coefficient on outcomes in different age groups and baseline National Institute of Health stroke scale (NIHSS) score subgroups. There were 265 patients who gave the standard dose treatment and 272 gave the nonstandard dose. There was no significant difference between the non-standard dose group and the standard dose group in 3m-mRS0-1, 3m-death, and 24h-sICH (p = 0.567, 0.327, and 0.415, respectively). The dose coefficient presents a significant negative correlation (p = 0.034, B = -4.290) with 3m-death in NIHSS < 16 sub-group. Door-to-needle time (DNT) is the most important independent outcome-influential factor (MIOIF) in the NIHSS ≥16 sub-group. The diabetes history and baseline NIHSS score were the MIOIF in the age ≥80-year sub-group. The non-standard dose group (0.6 mg/kg ≤ dose < 0.9 mg/kg) shows no difference in safety and effectiveness than the standard dose group (0.9 mg/kg) in our study. The standard dose should be considered first according to current evidence and Guidelines, but the non-standard dose (0.6 mg/kg ≤ dose < 0.9 mg/kg) might be an option in the actual diagnosis and treatment process considering the patient's clinical profile and financial condition.