Central venous pressure (CVP) can be measured readily by placing a transducer in series with any catheter in central venous circulation. CVP is equivalent to right atrial pressure (except in the uncommon setting of vena caval obstruction), and provides an estimate of right ventricular preload. Use of CVP to measure fluid responsiveness, however, is complicated by a number of factors. First, the waveform is complex, with an a wave resulting from atrial contraction, a c wave coincident with the closing of the tricuspid valve in end-diastole, and a v wave caused by atrial filling, with an x descent after the c wave and a y descent after the v wave (Fig 1). Strictly speaking, CVP should be measured in end-expiration at the base of the c wave, but if the c wave is not visible then the average of the a wave is used.20 The second, and even more daunting challenge, is that the physiologic determinants of CVP are multiple and do not remain constant. CVP is determined by interactions among venous return, which is a function of blood volume and compliance of the venous system, right ventricular function, and pulmonary arterial pressure. In critically ill patients, these are rarely static, and so it should come as little surprise that a single measurement of CVP is a poor predictor of fluid responsiveness.21 Tricuspid regurgitation, which produces large v waves, can also complicate interpretation of the CVP. Finally, use of CVP as an index of left ventricular preload implies that right-sided pressures are normal; in an ICU population with a high incidence of both acute and chronic pulmonary abnormalities, this is often not the case.