We hypothesized that suppression of peripheral circulation via cryotherapy may be effective in preventing paclitaxel-induced peripheral neuropathy (PIPN). Therefore, this study aimed to clarify whether self-administered cryotherapy could prevent PIPN in patients with early-stage breast cancer, using real-world data. A single-center, retrospective, observational study was conducted. Data from the electronic medical records of consecutive patients aged ≥ 20 years with early-stage breast cancer who received a regimen containing paclitaxel for 12 cycles with or without self-administered cryotherapy at the National Cancer Center Hospital from March 2018 to May 2019 were evaluated. The primary endpoint was the cumulative dose of paclitaxel until the onset of grade ≥ 2 PIPN. To compare the difference between the two groups, multivariable Cox proportional hazards models adjusted for prognostically important variables were used. Ninety Japanese patients were included in this study. The estimated incidence of grade ≥ 2 PIPN was 26.9% and 37.7% in the self-administered cryotherapy group and control group, respectively (P = 0.314). The multivariable Cox proportional hazards model showed that the self-administered cryotherapy group had a decreased risk of onset of grade ≥ 2 PIPN (hazard ratio: 0.63, 95% confidence interval: 0.25 to 1.39; P = 0.281). Sensitivity analyses using multivariable Cox proportional hazards models along with two propensity score-adjusted methods demonstrated consistent results. The findings suggest that the methods of self-administered cryotherapy may prevent PIPN and should be reinforced appropriately in clinical practice. A randomized controlled multicenter trial of self-administered cryotherapy is warranted.