OBJECTIVE The objective of this study was to assess the complicated relationship between frailty, perioperative complications, and patient-reported outcomes (PROs) in elderly patients (>= 75 years old) undergoing lumbar spine fusion (LSF). METHODS Consecutive patients who underwent LSF between March 2019 and December 2021 were recruited in this study. Frail patients (modified frailty index [mFI] score >= 2) were propensity score matched to nonfrail patients (mFI score 0-1) on the basis of age, sex, and the number of fused levels. Perioperative complications were collected and assessed according to the comprehensive complication index. Subgroups were further subdivided on the basis of the presence of major complications. The data from SF-36, Oswestry Disability Index (ODI), and North American Spine Society Satisfaction Questionnaire (NASS) at baseline and 1- and 2-year follow-up evaluations were compared between groups. Furthermore, the minimal clinically important difference (MCID) achievement rate was also compared. RESULTS The final analysis included 631 patients: 344 in the frail group and 287 in the nonfrail group. Frail patients were older (79.7 +/- 5.1 years vs 76.4 +/- 4.8 years, p < 0.001), with a higher proportion of females (68.9% vs 57.8%, p = 0.004) and those with malnutrition (17.7% vs 11.1%, p = 0.020). After propensity score matching for age, sex, and number of fused levels, 402 patients (201 in each group) were analyzed. Frail patients were more prone to have delirium (7.5% vs 3.0%, p = 0.044), blood transfusion (43.3% vs 30.3%, p = 0.007), and surgical site infection (6.0% vs 2.0%, p = 0.041). In addition, frail patients had a higher proportion of major complications (29.4%vs 16.9%, p = 0.003). Although they had worse PROs at baseline, frail patients obtained higher mean improvements and higher rates of MCID achievement by the 1- and 2-year follow-up evaluations than their nonfrail counterparts. Major complications did not seem to affect PROs in frail and nonfrail patients. CONCLUSIONS Despite being associated with worse baseline PROs, frail patients gained greater mean improvement in PROs and higher rates of MCID achievement by the 1- and 2-year follow-up evaluations than nonfrail patients. In addition, the presence of major complications did not affect PROs at the 1- and 2-year follow-ups. Although associated with major complications, elderly patients with frailty could benefit from LSF.