Can clinicians reliably estimate cardiac output in children?
© Current Science Ltd 1997
Published: 1 March 1997
Background and objectives
The acquisition of a reliable, repeatable, relatively non-invasive method for measuring cardiac output (CO) and hence cardiac index (CI) in children remains elusive. Because of this many clinicians involved in the provision of paediatric intensive care tend to rely on clinical and basic haemodynamic parameters when caring for the critically ill child. At our institution, we routinely use femoral artery thermodilution to measure cardiac output in critically ill children. We wished to evaluate different clinicians' abilities to estimate Cl in children and to see if estimates improved with time when results of objective measurements were known.
Patients and methods
One hundred and twelve estimates of Cl were made by 27 clinicians on 36 patients, median age 34.5 months (interquartile range 4.8–90.8 months). Clinicians originated from a variety of paediatric sub-specialities, including intensive care, cardiology and anaesthetics, and ranged from consultant to senior house officer (SHO) level. Before estimating cardiac index, clinicians were exposed to all clinical and laboratory data available on the patients, and were allowed to make a physical examination. They were initially asked to estimate the category of cardiac index (high ≥ 5 l/min/m2, high-normal 4.0–4.9, low-normal 3.0–3.9 and low < 3.0), and then the absolute value. Concurrently, five consecutive measurements were made, and then averaged using femoral artery thermodilution (COLD® Z-021, Pulsion Medical Systems, Munich). One clinician, who made all thermodilution measurements, estimated CI prior to measurement. This was to test if positive feedback improved estimation ability over time.
Measured CI ranged from 1.39-6.84 l/min/m2. Overall, there was poor correlation between absolute measured and estimated Cl (r = 0.24). There were slight differences between levels of seniority and accuracy of estimation: Consultants r = 0.19; Fellows r = 0.38; SHOs r = 0.04. For categorical estimation of Cl, the kappa statistic was 0.09 and weighted kappa 0.169, both indicate a poor strength of agreement. Furthermore, the clinician who was aware of previous Cl measurements showed no improvement in his ability to estimate Cl with time.
Clinical estimation of cardiac output is unreliable in paediatric practice. This unreliability is spread across disciplines and levels of seniority.