Does Mixed Venous Oxygen Saturation Reflect the Changes in Cardiac Output During Liver Transplantation? F. Acosta, T. Sansano, C.G. Palenciano, V. Roque ´ s, N. Clavel, P. Gonza ´ lez, R. Robles, F.S. Bueno, P. Ramı´rez, and P. Parrilla I T HAS BEEN SUGGESTED that mixed venous oxygen saturation (SvO 2 ) in the absence of hypoxemia and anemia could be useful in managing hemodynamically unstable patients. 1 Therefore its sensitivity and specificity in the assessment of cardiac output (CO) are controversial. 2,3 Our aim was to determine whether the constant recording of SvO 2 during liver transplantation (LT) might anticipate the frequent changes in CO. PATIENTS AND METHOD We studied 23 consecutive cirrhotic patients treated with LT using the piggyback technique. We performed a complete hemodynamic profile using a pulmonary artery catheter with a fast-response thermistor for right ventricular rejection fraction and oxymetry (RVEF, Baxter-Edwards, Calif). The variables studied were car- diac index (CI), cardiac preload measured by right ventricular end diastolic index (RVEDVI), and SvO 2 , at the following time points: beginning and end of the preanhepatic (A 1 ,A 2 ), anhepatic (B 1 ,B 2 ), and neohepatic (C 1 ,C 2 ) phases. Statistics Linear regression analysis was used with significance set at P .05. RESULTS The correlation between SvO 2 and CI was poor (r 2 = 0.36, P .001), and at certain time points during the operation, did not even exist. The same occurred with the correlation between SvO 2 and RVEDVI (r 2 = 0.34, P .001) (Table 1). DISCUSSION Jaugh et al 2 in critically ill patients found that changes in SvO 2 do not predict anticipated changes in CI. In LT we were also unable to show that the constant measurement of SvO 2 serves to anticipate the changes in cardiac preload or output. The reason may be the complex curvilinear rela- tionship between SvO 2 and CI preventing straightforward interpretation. 3 In fact, according to this relationship, when CI exceeds 4 L/min per m 2 the changes in SvO 2 are minimal. In cirrhotic patients treated with LT, CI usually exceeds this value. REFERENCES 1. Norfleet E, Watson C: J Clin Monit 1:4, 1985 2. Vaughn S, Puri V: Crit Care Med 16:495, 1988 3. Vincent J: Crit Care Clin 12:995, 1996 From the Liver Transplant Unit, University Hospital “V.Arrix- aca”, Murcia, Spain. Address reprint requests to Dr Francisco Acosta, San Cristo ´- bal, 4 (3B), 30001-Murcia, Spain. Table 1. Relationship Between SvO 2 and Hemodynamic Variables During LT A 1 A 2 B 1 B 2 C 1 C 2 SvO2 (%) 79.6 9.8 81.7 9.5 82.0 7.1 86.1 6.6 87.0 4.9 79.1 9.6 CI (L/min/m 2 ) 4.8 1.0 6.3 2.0 5.8 1.7 6.1 2.4 8.5 2.5 7.2 2.0 r 2 0.26 0.48 0.31 0.27 0.53 0.52 P .40 .05 .20 .20 .05 .05 RVEDVI (mL/m 2 ) 129.9 27.6 134.7 32.1 122.5 31.3 132.7 35.9 156.0 36.6 138.5 34.7 r 2 0.22 0.60 0.52 0.37 0.41 0.35 P .48 .05 .05 .09 .09 .13 See text for abbreviations. Values expressed as mean SD. n = 23. r 2 : correlation coefficient; P = level of significance. © 2002 by Elsevier Science Inc. 0041-1345/02/$–see front matter 655 Avenue of the Americas, New York, NY 10010 PII S0041-1345(01)02762-2 Transplantation Proceedings, 34, 277 (2002) 277