Medical Engineering & Physics 35 (2013) 178–187 Contents lists available at SciVerse ScienceDirect Medical Engineering & Physics j o ur nal homep age : www.elsevier.com/locate/medengphy Nonlinear dynamics of heart rate variability in response to orthostatism and hemodialysis in chronic renal failure patients: Recurrence analysis approach Hortensia González a , Oscar Infante b , Héctor Pérez-Grovas c , Marco V. Jose d , Claudia Lerma b,∗ a Laboratorio de Biofísica de Sistemas Excitables, Facultad de Ciencias, Universidad Nacional Autónoma de México, Mexico b Departamento de Instrumentación Electromecánica, Instituto Nacional de Cardiología, Mexico c Departamento de Nefrología, Instituto Nacional de Cardiología, Mexico d Grupo de Biología Teórica, Instituto de Investigaciones Biomédicas, Universidad Nacional Autónoma de México, Mexico a r t i c l e i n f o Article history: Received 19 September 2011 Received in revised form 24 April 2012 Accepted 28 April 2012 Keywords: Heart rate variability indexes Nonlinear dynamics Hemodialysis Renal disease Recurrence plot a b s t r a c t We studied the response of heart rate variability to hemodialysis and orthostatism using traditional linear indexes and 9 recurrence quantification analysis indexes to reveal changes in the heart rate dynam- ics. Twenty healthy subjects and 19 chronic renal failure patients treated with hemodialysis thrice a week were included. Five-minute heart rate variability time series were obtained during supine posi- tion (clinostatism) and orthostatism from each participant; recordings in renal patients were repeated after hemodialysis. Linear indexes were consistent with sympathetic predominance in response to orthostatism in the control group. Renal patients before hemodialysis showed increased sympathetic predominance in clinostatism, with further increase in orthostatism and hemodialysis. In response to orthostatism, 4 recurrence indexes changed in the control group, while in renal patients any of them changed before hemodialysis and 1 changed after hemodialysis. In clinostatism, renal patients (both before and after hemodialysis) had higher laminarity, trapping time, and recurrence time than the con- trol group. Recurrence indexes showed that the heart rate dynamics in renal patients are different from healthy subjects, suggesting loss of access to some regulatory conditions. These findings are consistent with reports of sympathetic stimulation induced by hemodialysis and active standing. © 2012 IPEM. Published by Elsevier Ltd. All rights reserved. 1. Introduction Heart rate variability (HRV) for short periods of time is asso- ciated to several cardiovascular regulatory systems including the sympathetic and parasympathetic nervous systems, as well as the renin-angiotensin system [1]. The variations in heart period can be studied by means of linear analysis in time and frequency domains if the methodological requirements, such as stationarity, are met [2]. The change from the supine to the standing posi- tion (orthostatism) is a well-described autonomic stress paradigm characterized by reduction in vagal outflow to the sinus node and increase in sympathetic nerve activity [3]. Linear HRV anal- ysis has been extensively used to evaluate autonomic regulation during orthostatism, characterized by increased mean heart rate, decreased variability and mean total power, increased mean power in the low frequency (LF) band (in normalized units), decreased mean power in the high frequency (HF) band (in normalized units), ∗ Corresponding author at: Departamento de Instrumentación Electromecánica, Instituto Nacional de Cardiología, C. Juan Badiano No. 1 Col. Sección XVI, Tlalpan, CP 14080, D.F., Mexico. Tel.: +52 55 5573 2911x1386; fax: +52 55 5573 0926. E-mail address: lermag@unam.mx (C. Lerma). and an increase in the LF/HF ratio [3–6]. However, the use of linear HRV indexes to assess autonomic regulation of the heart rate can- not be applied directly in all conditions [7] because HRV represents the integrated end-organ response to the complex nonlinear inter- action between sympathetic and parasympathetic activity among other factors [8–10]. Several nonlinear indexes of heart rate variability have been evaluated in healthy subjects and in different pathologies [11–14]. Besides describing nonlinear properties of the HRV signal, clinical applications have been shown for some indexes (e.g. independent prognosis of cardiac death in survivors of myocardial infarction) [11]. Most nonlinear time series analysis methods need long and stationary time series and their uncritical application to biological data can lead to serious pitfalls [15,16]. Therefore, new nonlinear methods with applicability to short and noisy data, as HRV, are con- tinuously developed with the aim of revealing hidden changes in the cardiovascular control system [17]. This is the case of recurrence analysis. Recurrence is a basic feature of many dynamical systems, including physiological systems, and it measures the repeated occurrence of a given state of the system. Recurrence plots are graphical representations of such dynamics [18]; they offer the pos- sibility not only of visualizing at once structures of the time series but also to estimate invariants from them. The structures exhibited 1350-4533/$ – see front matter © 2012 IPEM. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.medengphy.2012.04.013