Systolic Blood Pressure Visit-to-Visit Variability and Major Adverse Outcomes in Atrial Fibrillation: The AFFIRM Study (Atrial Fibrillation Follow-Up Investigation of Rhythm Management)

Literature - Proietti M, Romiti GF, Olshansky B, et al. - Hypertension 2017, published online ahead of print

Background

Although in the general population, greater BP variability, defined as SBP-visit-to-visit variability (VVV), is related to a higher risk of CV events and death, the clinical impact of SBP-VVV in AF patients is not known [1]. Moreover, no data are available about the relationship between SBP-VVV and quality of anticoagulation control. The latter is a strong predictor of major adverse events in these patients, and is expressed as individual time in therapeutic range (TTR) or percentage of international normalized ratio (INR) in range (PINRR) [2,3].

In this post hoc analysis of the AFFIRM trial, the relationship between SBP-VVV and quality of anticoagulation control was investigated, and the association between SBP-VVV and major adverse clinical outcomes in AF patients was analysed. The AFFIRM trial compared a rate-control versus a rhythm-control strategy for the clinical management of AF patients, with a mean follow-up of 3.5 years [4].

For the current analyses, all patients with available data on SBP at baseline and with at least 4 other measurements during follow-up were selected (n=3843; 94.7% of all AFFIRM patients). SBP-VVV was defined according to the SD of mean SBP during follow-up for every patient. According to the SBP-SD, patients were categorized in quartiles (1stQ: <10.09; 2ndQ: 10.09–13.85; 3rdQ: 13.86–17.33; 4thQ: ≥17.34 mmHg).

The thromboembolic risk was defined according to the CHA2DS2-VASc risk score, and the quality of anticoagulation control was assessed according to TTR and PINRR calculated from INR values. The outcomes of interest were stroke, major bleeding, CV death, all-cause death, and a composite of stroke/major bleeding/CV death.

Main results

SBP-VVV and quality of anticoagulation control

• The final adjusted linear regression model found that SBP-SD was inversely associated with TTR (standardized beta: -0.073; t: -4.142; P<0.001), and with PINRR (standardized beta: -0.070; t: -3.959; P<0.001).
• In the final logistic regression model, compared with patients in the 1stQ, those in the 4thQ had a significant inverse association with TTR >70% (OR: 0.75; 95% CI: 0.61–0.92), whereas patients in the 3rdQ and 4thQ were significantly associated with TTR ≤65% (P=0.014 for the 3rdQ and P<0.001 for the 4thQ), and with TTR ≤60% (both P<0.001).
• Continuous SBP-SD was inversely associated with TTR >70% (OR: 0.98; 95% CI: 0.96–0.99 per mm Hg) and directly associated with TTR ≤65% and TTR ≤60% (both P<0.001). Similar results were found for PINRR.

Follow-Up and Survival Analysis

• According to SBP-SD Qs, the rates of stroke and major bleeding progressively increased from the 1stQ to the 4thQ (both P<0.001). Patients in the 3rdQ and the 4thQ had the highest rate for CV death (P=0.005) and all-cause death (P<0.001).
• The composite outcome of stroke/major bleeding/CV death progressively increased from the 1stQ (10.8%) to 4thQ (20.8%; P<0.001).
• According to the Kaplan–Meier curves, patients in the 3rdQ and the 4thQ had a progressively higher risk of stroke (log-rank: 20.024; P<0.001), major bleeding (log-rank: 48.322; P<0.001), and CV death (log-rank: 11.985; P=0.007).
• Patients in the 4thQ had the highest risk for all-cause death (log-rank: 21.333; P<0.001). Patients in the 3rdQ and the 4thQ had the highest risk for the composite outcome (log-rank: 49.929; P<0.001).

Multivariate Analyses

• A final forward Cox multivariate regression analysis confirmed that patients in the 3rdQ and 4thQ had an independently increased risk for stroke (HR: 1.85; 95% CI: 1.02–3.35 and HR: 2.33; 95% CI: 1.30–4.16, respectively), and major bleeding (HR: 1.92; 95% CI: 1.18–3.15 and HR: 2.88; 95% CI: 1.79–4.61, respectively). SBP-SD as a continuous variable was associated with an increased risk for both outcomes (P=0.002 for stroke and P<0.001 for major bleeding).
• Despite an increase in risk of CV death for SBP-SD as a continuous variable (P=0.010), SBP-SD Qs were not independently associated with CV death. However, patients in 4thQ had an independent increase in risk for all-cause death (HR: 1.38; 95% CI: 1.00–1.91), which was also evident for SBP-SD as a continuous variable (P<0.001).

Conclusion

References

Find this article online at Hypertension