To compare the differences between noninvasive and invasive blood pressure in critical ill patients.

1072 pairs of invasive blood pressure by radial/ulnar artery catheter and noninvasive blood pressure by cuff on the same arm were collected from March 2016 to September 2016. The systolic blood pressure, diastolic blood pressure, pulse pressure and mean arterial pressure were analyzed by paired t-test. In addition, the same comparison was performed in subgroups, high blood pressure (MAP≥97 mmHg) (1 mmHg=0.133 kPa), normal blood pressure (70 mmHg≤MAP<97 mmHg) and low blood pressure (MAP<70 mmHg) according to the noninvasive mean blood pressure. Two-tailed P-values<0.05 were considered statistically significant.

There was a significant statistically difference between invasive systolic blood pressure and the noninvasive systolic pressure [(128.08±35.48) mmHg vs (122.56±24.84) mmHg, t=7.896, P<0.01). Similar results were found between invasive diastolic blood pressure and the noninvasive diastolic blood pressure [(65.66±13.69) mmHg vs (67.98±13.31) mmHg, t=-8.294, P<0.01], also the invasive pulse pressure and the noninvasive pulse pressure [(62.42±28.93) mmHg vs (54.58±20.00) mmHg, t=11.697, P<0.01]. However, no significant difference was found between the invasive and noninvasive mean arterial pressure [(86.47±18.94) mmHg vs (86.17±15.33) mmHg, t=0.867, P=0.386]. Subgroups analysis showed that in the high blood pressure subgroup (n=254), a significant statistically difference exists between invasive systolic blood pressure and the noninvasive systolic pressure [(163.75±33.93) mmHg vs (152.16±16.78) mmHg, t=6.52, P<0.001] as well as invasive diastolic blood pressure and the noninvasive diastolic blood pressure [(79.17±11.03) mmHg vs (83.69±9.50) mmHg, t=-6.85, P<0.001], similarly results were also found in invasive pulse pressure and the noninvasive pulse pressure [(84.57±31.50) mmHg vs (68.47±20.72) mmHg, t=9.76, P<0.001]; in normal blood subgroup (n=687), a statistically difference exists in systolic blood pressure, diastolic blood pressure and pulse pressure, respectively between invasive and non-invasive measurements [(122.66±24.74) mmHg vs (118.70±15.14) mmHg, t=5.071, P<0.001; (63.97±10.34) mmHg vs (65.60±8.49) mmHg, t=-5.049, P<0.001; (58.69±23.05) mmHg vs (53.10±11.90) mmHg, t=7.682, P<0.001]; by contrast, in low blood pressure subgroup (n=131), no significant statistically difference was found in invasive systolic blood pressure and the noninvasive systolic pressure [(87.35±24.33) mmHg vs (85.41±11.99) mmHg, t=1.109, P=0.269], significant difference was found between invasive diastolic blood pressure and the noninvasive diastolic blood pressure [(48.32±8.27) mmHg vs (49.98±8.06) mmHg, t=-2.073, P=0.040], as well as invasive pulse pressure and the noninvasive pulse pressure [(39.03±24.00) mmHg vs (35.43±13.97) mmHg, t=1.806, P<0.001].

The invasive systolic blood pressure is higher than the noninvasive systolic blood pressure while the invasive diastolic blood pressure is lower than the noninvasive diastolic blood pressure and the invasive pulse pressure is higher than the noninvasive pulse pressure. No difference is found between the invasive and the noninvasive mean arterial pressure. The mean blood pressure may eliminate the differences between invasive and noninvasive blood pressure.