To the Editor: Measuring cardiac output (CO) in hemodialysis (HD) patients is of potentially great value for studying the pathophysiology of intradialytic hypotension. However, only one device currently available—the Transonic monitor (TRS; Transonic Systems, Ithaca, NY)—has gone through an appropriate validation. The TRS is generally regarded as the reference method for intradialytic CO measurements. Unfortunately, being an indicator dilution method, it cannot give continuous readings, and its use is restricted to patients with AV fistulas. Kitzler et al1 performed a validation study on 12 patients, examining a novel impedance cardiography (ICG) device [Task Force monitor (TFM), CN Systems, Graz, Austria] with the TRS as the reference method. The authors reported good agreement between the two methods (r2 = 0.64, SD 0.6 L/min on Bland-Altman analysis). However, their approach was to measure CO simultaneously with both methods 3 to 7 times during each session and then compare the 12 overall session mean values from each method. Thus, their publication neither revealed the precision of individual CO measurements with the TFM nor, most importantly, whether intradialytic CO trends are reliably tracked with the TFM. We addressed these last questions in a validation study on another ICG device, the BioZ (BZ; SonoSite Inc., Washington,). Twelve maintenance HD patients with AV fistulas were studied during one session each; the study was approved by the regional ethics committee and all participants gave informed consent. Demographic data of the participants are listed in Table 1.Table 1: Demographic Data of the ParticipantsA simultaneous CO measurement was performed with the BZ and the TRS in the first hour of the dialysis session and repeated in the last hour. The measurements of each method are presented in Table 2. Bland-Altman analysis revealed systematic underestimation of CO by the BZ (bias −15.9%, SD 26.4%, Figure 1). Regression analysis of difference versus average for proportional bias was nonsignificant (Figure 1).Table 2: Intradialytic CO Measurements with Each MethodFigure 1.: Bland-Altman plot of paired measurements, n = 24. The difference between measurements is presented as the relative (percentage) difference from the average. The dotted lines represent the limits of agreement. BZ, BioZ; TRS, Transonic.Regression analysis of relative changes showed no correlation between pairs (r2 = 0.04, Figure 2). A concordance analysis2 with the cutoff set at 15% change yielded a concordance rate of 20% (1/5 pairs).Figure 2.: Correlation plot of the measured relative (percentage) intradialytic change in cardiac output, n = 12. CO, cardiac output; BZ, BioZ; TRS, Transonic.Our results raise serious concerns about use of the BZ for intradialytic hemodynamic assessment. Although the BZ and other ICG monitors have in fact been used for this purpose,3,4 to our knowledge, the method has in this setting only been validated against transthoracic echocardiography-derived CO5—a technique whose operator-dependency and variability interferes with its suitability as reference method—and in the aforementioned study1 with its relative obscurity of data. In conclusion, we find that use of ICG for research in dialysis hemodynamics remains controversial until rigorous validation studies support its reliability. Joakim Cordtz Soeren D. Ladefoged Department of Nephrology Copenhagen University Hospital Copenhagen, Denmark