Background Recent evidence suggests that central aortic blood pressure may be a better predictor of cardiovascular risk than peripheral blood pressure. The central SBP (cSBP) can be estimated from the late systolic shoulder of the radial pulse waveform. We compared the second systolic peak of the radial waveform (pSBP2) with the central systolic pressure derived by a generalized transfer function in a large cohort, across a wide age range, of patients from the Anglo-Cardiff Collaborative Trial. We also compared pSBP 2 with the true cSBP measured by cardiac catheterization [invasively measured cSBP (cSBPi)]. Methods Noninvasive measurements were made by applanation tonometry using the SphygmoCor device. The aortic pressure waveform was derived from the radial waveform using a validated transfer function. Invasive measures of cSBPi were carried out in a group of 38 patients undergoing diagnostic cardiac angiography, and radial artery pressure waveforms were simultaneously recorded using the SphygmoCor device. Results Overall, there was a strong correlation (r=0.99, P<0.001) and good agreement between pSBP 2 and the derived cSBP (mean difference ± SD 1 ±4 mmHg). However, there was a systematic bias with a greater difference between these measures at lower average pressures. There was also a strong correlation and good agreement between the invasively measured cSBPi and pSBP2 (r=0.92, P<0.001, mean difference 2±6mmHg). Conclusion The second systolic shoulder of the peripheral pressure waveform approximates the cSBP in a large cohort of patients across a wide age range, but this may be inaccurate at low SBP values.