Background This study investigated the different blood pressure patterns that were evaluated by ambulatory blood pressure monitoring (ABPM) among elderly patients and explored the effect of pressure patterns on cognitive impairment and mortality. Methods A total of 305 elderly participants aged >= 65 years were divided into the cognitive impairment group (CI, n = 130) and the non-cognitive impairment group (NCI, n = 175) according to the MMSE score. All participants underwent ABPM to evaluate possible hypertensive disorder and cerebral MRI for the evaluation of cerebral small vessel disease. Follow-up was performed by telephone or medical records. The primary outcome was all-cause mortality. Secondary endpoints were major adverse cardiac and cerebrovascular events (MACCE). Results Among 305 participants, 130 (42.6%) were identified with cognitive impairment (CI), with average systolic blood pressure (BP) of 127 mmHg and diastolic BP of 66 mmHg. According to ABPM, only 13.1% had a dipper pattern, 45.6% had a nocturnal BP rise, while 41.3% had a non-dipper pattern. Compared with NCI patients, the CI group had significantly higher night-time systolic BP (130.0 +/- 18.2 vs. 123.9 +/- 15.1, p = 0.011), and more participants had nocturnal BP rise (52.3% vs. 40.6%, p = 0.042). Nocturnal BP rise was associated with greater white matter hyperintensities (WMH) (p = 0.013). After 2.03 years of follow-up, there were 35 all-cause deaths and 33 cases of major adverse cardiac and cerebrovascular events (MACCE). CI was independently associated with all-cause mortality during long-term observation (p < 0.01). Nocturnal BP rise had no significant predictive ability for all-cause mortality in elderly patients (p = 0.178). Conclusions Nocturnal BP rise contributed to greater cognitive impairment in elderly patients. Not nocturnal BP rise, but CI could significantly increase all-cause mortality. Controlling BP based on ABPM is critical for preventing the progression of cognitive dysfunction.