Automated blood pressure measurement in atrial fibrillation: validation process modification and evaluation of a novel professional device which detects atrial fibrillation and adapts its blood pressure measurement algorithm

J Hypertens. 2021 Apr 1;39(4):614-620. doi: 10.1097/HJH.0000000000002684.


Objectives: Blood pressure (BP) measurement in atrial fibrillation (AF) patients is problematic and automated monitors are regarded as inaccurate. The optimal procedure for validating BP monitors in AF is questionable. This study evaluated the accuracy of a novel professional oscillometric upper-arm cuff device (Microlife WatchBP Office), which has an algorithm for detecting AF and then applies an AF-specific BP measurement algorithm. BP variability, which is inherently increased in AF patients, was considered in the analysis.

Methods: Subjects with sustained AF were included in a validation study using the same arm sequential measurement method of the Universal Standard (ISO 81060-2:2018) for special populations. Analysis was performed in all subjects and separately in those with and without high reference BP variability (>12/8 mmHg SBP/DBP).

Results: Thirty-five subjects with 105 paired test/reference BP measurements were included (mean age 76.3 ± 8.4 years, reference SBP/DBP 128.2 ± 19.5/72.5 ± 12.1 mmHg, pulse rate 68.3 ± 14.9 bpm). Validation Criterion 1 (mean difference ± SD) was 0.0 ± 7.7/0.2 ± 7.0 mmHg in all 105 BP pairs (threshold ≤5 ± 8 mmHg). Criterion 1 was 0.5 ± 6.1/-0.2 ± 6.8 mmHg in 18 subjects (54 BP pairs) with low reference BP variability and -0.6 ± 9.2/0.6 ± 7.3 mmHg in 17 (51 pairs) with high variability. Criterion 1 did not differ in pulse rate < 70 vs. ≥ 70 bpm Validation Criterion 2 (SD of differences for 35 individuals) was 5.38/6.20 mmHg (SBP/DBP; threshold ≤6.95/6.95).

Conclusion: A technology which detects AF and activates an AF-specific BP measurement algorithm introduces a challenging solution for clinical practice. Validation of BP monitors in AF patients should not ignore their inherently high BP variability.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Aged
  • Algorithms
  • Atrial Fibrillation* / diagnosis
  • Blood Pressure
  • Blood Pressure Determination
  • Humans
  • Oscillometry