Table 2

Relationship between clinical variables and procedural success in London by univariate and multivariate analysis

VariableUnivariate analysesMultivariate analyses
P valueRR (95% CI)*P value
Long-standing persistent AF (>1 year)<0.00010.40 (0.23 to 0.70)0.001
Follow-up (months)<0.00010.87 (0.79 to 0.95)†0.003
Female sex0.0510.44 (0.25 to 0.78)0.005
Age (years)0.0020.006
 <500.26 (0.11 to 0.58)0.001
 50–590.65 (0.32 to 1.32)0.234
 60–690.81 (0.43 to 1.54)0.517
 ≥701.0001.000
Thyrotoxicosis within the last year0.0060.21 (0.06 to 0.83)0.025
AT rather than AF (at redo procedure)0.1222.04 (1.10 to 3.82)0.025
Left atrial diameter (mm)0.0010.96 (0.93 to 0.996)‡0.028
Extreme comorbidity0.0010.46 (0.21 to 1.03)0.059
Previous use of amiodarone0.008
Previous antiarrhythmic drugs§0.013
Valvular heart disease0.062
Left ventricular function0.106
Diabetes mellitus0.177
Renal dysfunction0.178
Hypertension0.191
Stroke or TIA0.198
Any previous AF ablation0.327
Coronary artery disease0.374
Pulmonary disease0.429
Number previous AF ablations0.783
Body mass index0.987
  • *Relative Risk (95% CI) of success, derived from binary logistic regression.

  • †Continuous variable; negative correlation with success.

  • ‡Borderline statistical significance

  • §Class I or III agents

  • AF, indicates atrial fibrillation; AT, atrial tachycardia; TIA, transient ischaemic attack.