Prediction of fluid responsiveness in severe preeclamptic patients with oliguria

Abstract

Purpose: Because severe preeclampsia (SP) may be associated with acute pulmonary oedema, fluid responsiveness needs to be accurately predicted. Passive leg raising (PLR) predicts fluid responsiveness. PLR has never been reported during pregnancy. Our first aim was to determine the percentage of SP patients with oliguria increasing their stroke volume after fluid challenge. Our second aim was to assess the accuracy of PLR to predict fluid responsiveness in those patients.

Methods: Patients with SP were prospectively included in the study. In the subgroup developing oliguria, transthoracic echocardiography was performed at baseline, during PLR and after a 500 ml fluid infusion over 15 min. Fluid responders were defined by a 15 % increase of stroke volume index. Five consecutive measurements were averaged for all parameters.

Results: Twenty-three (56 %) out of 41 patients with SP developed oliguria, 12 (52 %) out of these 23 responded to fluid challenge. During PLR, an increase of the velocity time integral of subaortic blood flow (ΔVTI) above 12 % predicted the response with a sensitivity and specificity of 75 [95 % confident interval (CI): 0.42-0.95] and 100 % (95 % CI: 0.72-1.00), respectively. An algorithm combining ΔVTI and the baseline value of VTI predicted fluid responsiveness with a sensitivity and specificity of 100 % (95 % CI: 0.74-1.00) and (95 % CI: 0.75-1.00). Urine output and respiratory variations of inferior vena cava diameter did not predict fluid responsiveness.

Conclusions: Only 52 % of oliguric patients were responders. PLR accurately predicts fluid responsiveness in the specific setting of SP. This noninvasive test should be tested in future algorithms for the management of SP.