Cardiac output calculation using the Liljestrand and Zander formula: is this method applicable during immediate transition after birth? — A post hoc analysis

The immediate neonatal transition after birth is associated with complex cardio-circulatory changes and is one of the most vulnerable periods in life [1]. Consequently, the clinical evaluation of the neonate’s condition after birth is crucial and consistently includes Apgar scoring. The Apgar score is designed to assess for signs of hemodynamic compromise. However, due to its inter-observer variability and the limitations regarding preterm neonates, its reliability is debatable [2, 3]. Recent resuscitation guidelines recommend the use of pulse oximetry and electrocardiogram (ECG) for initial assessment of neonates to identify those in need for medical support [4, 5]. The routinely monitored heart rate (HR) alone may not always reflect failure of cardio-circulatory transition sufficiently. Assessment of arterial blood pressure might provide additional information. Even though our research group established reference ranges for arterial blood pressure during this period after birth, its value is not clear yet [6, 7]. For a more thorough understanding of cardio-circulatory changes during neonatal transition, cardiac output (CO) assessment might add important information. The gold standard in terms of non-invasive evaluation of CO is echocardiography, which, however, is not always feasible during this immediate transition period [8]. Non-invasive CO measurement (CO-bioimpedance) using electrical bioimpedance emerges as an alternative; hence, its practicability within the initial minutes after birth poses challenges but its accuracy appears acceptable [9, 10]. An alternative tool to gain information on CO is the Liljestrand and Zander (LaZ) formula [11] which has been described to be one of the most adequate pulse pressure wave analysis (PPWA) [12]. The application of this formula to estimate CO might be a promising alternative to other more complex methods of CO assessment during the immediate transition in neonates.

The aim of this study was to assess the adjusted estimated CO values calculated by the LaZ formula (COest/adj LaZ) and compare these with CO values non-invasively measured by electrical bioimpedance (CO-bioimpedance) during immediate transition in preterm and term neonates.

Between December 2018 and November 2019, 80 preterm and term neonates were included in the prospective observational study and eligible for analysis. Forty-eight neonates were excluded because CO-bioimpedance data showed a SQI < 80%. No additional neonates were excluded for any other reasons.

The demographic and clinical data presented in Table 1 are within normal ranges. Within the cohort, there were two neonates classified as small for gestational age (SGA). Additionally, the favorable Apgar scores and arterial pH values indicate stability among the newborn neonates.

No significant difference was observed between the two CO measurement methods (Table 1).

A correlation analysis of CO-bioimpedance and COest/adj LaZ revealed a significant positive correlation between the CO values (Table 1).

In the Bland–Altman plot (Fig. 1), the mean difference between CO-bioimpedance and COest/adj LaZ was 0.015, with a standard deviation (SD) of 0.15, respectively. The upper limit was 0.31 and a lower limit −0.28.

The present study demonstrates comparable results between COest/adj LaZ and CO-bioimpedance during the immediate transition period, indicating that the COest/adj LaZ may be a proper alternative with an acceptable precision. The results are also in accordance with previous findings after the immediate transition [12].

Before comparison of CO-bioimpedance with the COest/adj LaZ, the COest LaZ estimated value underwent adjustment through multiplication by a calibration factor (k) as proposed by König and Sun [13]. This adjustment resulting in COest/adj LaZ improved in accordance with the CO-bioimpedance.

Currently, CO-bioimpedance stands as the sole non-invasive method for continuous CO monitoring, demonstrating acceptable precision. However, its main challenge lies in signal quality. Noori et al. compared echocardiography with CO-bioimpedance and described comparable accuracy and precision in CO measurements when the SQI of CO-bioimpedance was equal to or exceeds 80% [14]. Considering this, CO-bioimpedance encounters difficulties, especially during the immediate transition period due to low SQI caused by movements and manipulations, as highlighted in several studies [9, 10]. During the immediate transition, therefore, calculating COest/adj using the LaZ method may prove beneficial for clinicians and researchers, since obtaining COest/adj LaZ based on blood pressure and heart rate with satisfactory quality appears more feasible than with CO-bioimpedance measurement. For instance, in our study, approximately 60% of neonates were excluded due to poor SQI of CO-bioimpedance which is consistent with other studies [9]. In contrast, no neonate was excluded due to missing blood pressure or heart rate values. Despite the high exclusion rate due to the high SQI may be debated, this stringent standard of data quality of CO-bioimpedance bolsters the precision of the comparison of CO-bioimpedance with COest/adj LaZ in the present study.

Besides that, Boet et al. [15] demonstrated a tendency for CO-bioimpedance to overestimate echocardiography readings when CO exceeds 0.4 l per minute in preterm neonates with a mean gestational age of 31 weeks. Our study, however, included neonates with the mean gestational age of 37 weeks that makes comparison to the data of Boet et al. [15] difficult and might explain differences in findings. In contrast, the study conducted by Noori et al. [14] demonstrated in a comparable cohort in neonates with a mean gestational age of 39.2 weeks that CO-bioimpedance readings align closely with echocardiography even for CO values exceeding 0.4 l per minute. Furthermore, it has been noted that the precision of CO-bioimpedance remains uncertain when applied to sick neonates [10]. However, in our cohort, the included neonates after caesarean section did not exhibit severe illness, suggesting further reliability in the CO-bioimpedance measurements and allowing comparison with COest/adj LaZ.

The high comparability between COest/adj LaZ and another CO measurement method is in accordance with PPWA trial by Sun et al. which demonstrated that the LaZ calculation emerged as the most suitable PPWA method [12]. Furthermore, Koenig et al. showed commendable precision of COest/adj LaZ when comparing it to CO derived from the Finometer in adults [13]. The Finometer, a Modelflow device used for CO estimation, has consistently shown, in various adult studies, values akin to those obtained through thermodilution — a more invasive technique considered the gold standard for CO measurement [16].

In our study, we explored a straightforward formula for estimating CO. Our results indicate that calculated COest/adj using the LaZ formula was comparable and correlated significantly with CO-bioimpedance in healthy neonates during immediate neonatal transition after birth. These findings have significant clinical implications suggesting that CO can easily be derived in a cost-effective and feasible manner, particularly when other methods are not available.