Few studies assessed modalities of invasive arterial pressure monitoring (IAPM). We evaluated effects on measured values of various combinations of transducer level, catheter access site, and patient position. Prospective observational study in consecutive adults admitted to a French intensive care unit in 2009 to 2011 and fulfilling our inclusion criteria. Four combinations (B-E) of transducer level, catheter access site, and patient position were compared with a reference combination (A) (A: patient supine with all catheters in the same plane and a single transducer level (M) for zero point reference (Z) aligned on the phlebostatic axis; B: 45° head-of-bed elevation with M and Z aligned on the phlebostatic axis; C: 45° head-of-bed elevation with M aligned on the catheter access site and Z on the phlebostatic axis; D: 45° head-of-bed elevation with M and Z aligned on the catheter access site; and E: 45° head-of-bed elevation with M aligned on the phlebostatic axis and Z on the catheter access site). We included 103 patients, 68 men and 35 women, with a median age of 69 years (interquartile range [IQR], 56-78); at inclusion, 91 (88.3%) received mechanical ventilation, 45 (43.7%) catecholamines, and 66 (64.1%) sedation. The IAPM access site was femoral in 49 (47.6%) and radial in 54 (52.4%) patients, with 62 of 103 (60.2%) catheters on the right side. Measured absolute mean arterial pressure values were significantly higher with 3 study combinations (C-E) than with the reference combination (A). After adjustment, the differences versus A (median, 83 [IQR, 74-92] mm Hg) remained significant for D (median, 91 [IQR, 85-100] mm Hg, P < 0.001) and E (median, 88 [IQR, 77-99] mm Hg, P < 0.001). The difference versus A was not significant for B (median, 85 [IQR, 76-94] mm Hg, P = 0.21) or C (median, 90 [IQR, 84-100] mm Hg, P = 0.006). Several modalities used for zeroing and/or transducer leveling during IAPM may result in statistically and clinically significant overestimation of measured mean arterial pressure values. For patients in the 45° head-of-bed elevation position, aligning the Z on the phlebostatic axis provides values that are not significantly different from those obtained using the reference supine modality.