| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
INNOVATIVE METHODOLOGY
1Department of Cardiovascular Dynamics, National Cardiovascular Center Research Institute, Suita 565-8565; 2Japan Space Forum, Tokyo 105-0013; 3Organization of Pharmaceutical Safety and Research, Tokyo 100-0013; and 4Department of Cardiovascular Control, Kochi Medical School, Nankoku 783-8505, Japan
Submitted 15 January 2004 ; accepted in final form 28 July 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONAPPENDIXGRANTSREFERENCES |
|---|
conductance catheter; serial reproducibility; volumetric accuracy; dual-frequency method; Bluetooth
In the present study, we have developed a new telemetry system to measure LV volume, pressure, and electrocardiogram (ECG) in conscious, freely moving rats. In this system, the LV pressure-volume relation was obtained from a pressure-conductance catheter chronically implanted in the rat LV. To calibrate the conductance signal and obtain absolute LV volume, measurements of blood resistivity (
) and parallel conductance (Gp) are required (3, 4). These calibration procedures require blood sampling and hypertonic saline infusion, but such ex vivo procedures are not applicable to conscious, freely moving small animals. To circumvent such ex vivo procedures in our new telemetry system (29), we adopted a self-calibrating method for the LV volume measurement, as reported in our previous study (28). The aim of the present study was therefore to develop a telemetry system and evaluate its performance. Our results indicate that we succeeded in measuring the LV pressure-volume relation in conscious, untethered rats with reasonable accuracy and reproducibility.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONAPPENDIXGRANTSREFERENCES |
|---|
Figure 1A illustrates a newly developed pressure-volume telemetry system for rats; it consists of a pressure-conductance catheter, an analog processor-transmitter (weight = 14 g, volume = 11 ml), and a battery unit (lithium battery; weight = 12 g, volume = 10 ml).
|
, four smaller platinum electrodes (0.1 mm wide, 0.2 mm between centers of adjacent electrodes, 0.6 mm between centers of excitation electrodes) were placed near the pressure transducer (Fig. 1B, inset). Constant excitation current was applied to the two outer electrodes while the voltage signal associated with was measured from the two inner electrodes.
Analog processor transmitter.
A block diagram of the analog processor transmitter is presented in Fig. 1C. It was equipped with several functions. First, it delivered a dual-frequency (2 and 20 kHz) constant excitation current [20 µA root mean square (RMS)] for measurements of LV conductance and . We validated the current output by injecting it into known resistors and examining the developed voltage. The resulting RMS current output was 20.4 µA (SD 0.2) and 19.3 µA (SD 0.2) at 2 and 20 kHz, respectively. These values were constant over different resistors (50–990 
). Second, it measured and processed the voltage signal from the sensing electrodes as follows: Analog signals were digitized (12 bits, 40-kHz sampling rate; model ADS7870, Texas Instruments, Dallas, TX) and then fed into a programmable logic device (model XC 2C256, Xilinx, San Jose, CA), which processed them to yield RMS digital signals corresponding to frequency components of 2 and 20 kHz and a low-frequency signal (<2 kHz; see APPENDIX). The circuit was connected to the larger or smaller electrodes in response to a command signal, so that LV conductance or could be measured. Third, the analog processor-transmitter had a bridge amplifier for the LV pressure measurement. The LV pressure signal was also digitized (12 bits, 40-kHz sampling rate). All these functions were controlled by a microcomputer (model H8S, Hitachi, Tokyo, Japan).
Bluetooth technology was used to transmit the data (18). For real-time monitoring, all processed signals were resampled at 200 Hz by the microcomputer and transmitted to an external receiver (CASIRA, CSR, Cambridge, UK) by a Bluetooth module (model LMBTB027, Murata, Tokyo, Japan). For high-precision non-real-time analysis, signals recorded at 2,000 Hz over a 6-s interval were stored in a static random access memory (model HM62V16256, Hitachi) and then transmitted to the receiver by the Bluetooth module. The external receiver detected the radio-frequency signal from the transmitter and converted it to a serial bit stream.
Self-Calibration of Ventricular Volumetry
The principles of conductance volumetry have been described previously (3, 4). Briefly, the ventricular conductance signal (G) can be converted to absolute ventricular volume (V) as follows
 | (1) |
is a volume calibration factor, L is the distance between the sensing electrodes, is blood resistivity, and Gp is parallel conductance. L was 9 mm in the present catheter design.
In a preliminary experiment, when the catheter was placed in a series of graduated syringes filled with diluted saline, conductance-derived volumes at 2 and 20 kHz were close to the true syringe volume in the volume range of interest (Fig. 2). Conductance-derived volumes at the two frequencies were essentially identical for each of the syringe volumes. Hence, was assumed to be unity in the present study (14, 23).
|
(Fig. 1B, inset). The distance between the excitation electrodes was set at 0.6 mm. In an in vitro experiment, we confirmed that the current distribution volume was confined to an 
4-mm diameter around the catheter with this electrode design (see APPENDIX). The end-diastolic LV diameter is 7–9 mm in normal rats and 9–12 mm in rats with left heart failure (17). Because the interelectrode distance between the excitation electrodes was small enough to confine the current distribution volume to within the end-diastolic ventricular blood pool in the rat LV, we estimated at end diastole (10, 28).
Gp was estimated by the dual-frequency excitation method (8, 9, 28) as follows
 | (2) |
G20–2 is the difference in ventricular conductance values between the 20- and 2-kHz excitation frequencies and 
is an experimentally derived constant. Once is determined, Gp can be estimated from G20–2, obviating the need for saline infusion. Instrumentation and Experimental Protocols
Thirty-three male Sprague-Dawley rats (350–400 g body wt) were used. Care of the animals was in strict accordance with the Guiding Principles for the Care and Use of Animals in the Field of Physiological Sciences as approved by the Physiological Society of Japan. The animals were anesthetized with pentobarbital sodium (50 mg/kg ip) and ventilated artificially. A vertical midline cervical incision was made to expose the right common carotid artery while the animal was in the supine position. The pressure-conductance catheter of the telemetry system was inserted into the LV retrogradely from the right common carotid artery. The position of the catheter was verified by monitoring the pressure-volume signal and by two-dimensional echocardiography. At the conclusion of the experiment, the animal was killed with an overdose of pentobarbital sodium, and the heart was examined to reconfirm the proper positioning of the catheter.
Group 1 (n = 23).
We evaluated the accuracy of telemetric calibration of and Gp under anesthetized, closed-chest conditions. Catheters (3-Fr) were inserted into the right and left jugular veins for blood sampling and saline injection, respectively. In 10 of the 23 rats, we compared estimated telemetrically (est) with measured from sampled blood by a conventional ex vivo cuvette method (conv). In the remaining 13 rats, we estimated Gp by the dual-frequency excitation method (Gp,est) and by the hypertonic saline method (Gp,conv). To obtain Gp,conv, we injected 20 µl of saturated saline into the right jugular vein while continuously measuring LV conductance (14, 23). To obtain Gp,est, we measured LV conductance at 2- and 20-kHz excitation frequencies and derived G20–2 by averaging the instantaneous conductance difference over 10 cardiac cycles. We randomly selected 7 of the 13 rats and determined the proportionality constant ( in Eq. 2) from the averaged ratio of Gp,conv to G20–2. Gp,est and Gp,conv were measured while the artificial ventilation was temporarily suspended at end expiration.
Group 2 (n = 4). Under anesthetized, open-chest conditions, we evaluated the accuracy of volumetry by comparing stroke volume (SV) measured by the telemetry system with SV measured by an ultrasonic flowmeter (model 2.5S273, Transonic Systems, Ithaca, NY). After median sternotomy, the aortic arch was dissected free from surrounding tissues. A flow probe was placed around the ascending aorta to measure the aortic blood flow. A string occluder was placed loosely around the inferior vena cava to decrease the LV preload and vary the SV over a wide range. We simultaneously measured the telemetric LV volume and the ultrasonic aortic blood flow while varying the preload. The measurements were done while the artificial ventilation was temporarily suspended at end expiration.
Group 3 (n = 6). Under conscious, closed-chest conditions, we evaluated the reproducibility of the telemetry on different days. Aseptic conditions were maintained throughout the surgical procedure. The telemetry system was implanted in a subcutaneous pocket made at the right upper dorsum. The skin was closed, and the animal was allowed to recover from anesthesia. On the day after implantation surgery, the LV volume, pressure, and an ECG were measured telemetrically in the fully recovered, conscious animal (study 1). Each rat underwent a second set of telemetric measurements at 1–6 days after the initial study (study 2). Ambient barometric pressure was measured simultaneously and subtracted from the telemetric LV pressure to compensate for changes in atmospheric pressure.
Data Collection
We used the real-time mode (200-Hz sampling) of the telemetry system and recorded LV conductance, LV pressure, intraventricular ECG, and on a hard disk of a dedicated laboratory computer system (model HFPA031003, Epson, Tokyo, Japan). In group 2, ultrasonic aortic blood flow was digitized at 1,000 Hz through a 12-bit analog-to-digital converter and stored on a hard disk for subsequent analyses.
Statistical Analysis
For the calculation of LV volume using Eq. 1, G and were obtained from the 20-kHz frequency component. In group 1, we used linear regression analysis to compare the telemetric and conventional measurements of (est vs. conv) and Gp (Gp,est vs. Gp,conv). In group 2, we calculated the telemetric SV from the difference between the end-diastolic volume (EDV) and end-systolic volume (ESV) in each beat. The flowmetric SV was computed from the time integral of aortic blood flow. The telemetric SV was compared with the flowmetric SV by linear regression analysis. In group 3, we compared heart rate (HR), EDV, ESV, end-diastolic pressure (EDP), and end-systolic pressure (ESP) between study 1 and study 2 for each rat. Using the pressure-volume data, we calculated ejection fraction (EF), maximal pressure increase (+dP/dtmax) or decrease (–dP/dtmax) over time, and the time constant of isovolumic relaxation (
) and compared them between study 1 and study 2 for each rat. A nonparametric multiple comparison (Wilcoxon’s signed-rank test) was used to examine the difference in each parameter between study 1 and study 2. Group data are expressed as means (SD). Differences were considered significant at P < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONAPPENDIXGRANTSREFERENCES |
|---|
Figure 3A is a representative time series showing LV pressure and at 2 and 20 kHz derived from the telemetry. The bottom of the waveform, which corresponded to end diastole, represents the time when there was sufficient blood volume around the catheter (10). The lowest values at 2 kHz (2kHz) and 20 kHz (20kHz) were very close (197 and 207 ·cm, respectively). This was the case for all the rats, indicating that was frequency independent (2kHz = 1.0820kHz – 13.8, r2 = 0.96, SE of the estimate = 6.7 ·cm) (6, 9). The lowest at 20 kHz was treated as est.
|
est and conv obtained from 10 rats in group 1. est agreed with conv reasonably well (est = 1.09conv – 11.9, r2 = 0.88, SE of the estimate = 10.7 ·cm). The ratio of SE of the estimate to the mean of est was 0.046, indicating small variability around the regression line. Telemetric Calibration of Gp
Figure 4A illustrates a representative time series of telemetrically measured ECG, LV conductance signals at 2 and 20 kHz, and LV pressure. In this animal, G20–2 was 0.56 mS and Gp,conv was 3.27 mS. Therefore, was calculated to be 5.79 from Eq. 2 in this animal. The averaged from seven randomly selected rats was 5.14, which we used as the experimentally derived constant to obtain Gp,est for all rats.
|
Accuracy of the Televolumetry
Figure 5A depicts LV pressure and volume measured by telemetry and aortic blood flow measured by the ultrasonic flowmeter. Vena caval occlusion decreased LV pressure, volume, and aortic blood flow.
|
Reproducibility of the Telemetry
Individual data obtained by the telemetry system for all six rats in group 3 are provided in Tables 1 and 2. The overall variability between repeated measurements in the same rat was reasonably small. There were no significant differences in repeated measurements of HR, EDV, ESV, EDP, and ESP between study 1 and study 2 (Table 1). There were no significant differences in repeated measurements of EF, +dP/dtmax, –dP/dtmax, and between study 1 and study 2 (Table 2).
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONAPPENDIXGRANTSREFERENCES |
|---|
Self-Calibrating Volumetry
In our conductance volumetric system, and Gp were estimated using the telemetric signals alone (Figs. 3 and 4). We will be able to use the empirical constant (=5.14), determined in this study, in the future application of our telemetry system to rats. The self-calibrating feature made it possible to measure the LV pressure-volume relation in rats without tethering them for ex vivo calibration procedures, such as blood sampling and hypertonic saline infusion. Besides their impracticality in conscious, small animals, these procedures can alter hemodynamic conditions (6, 9). Frequent blood sampling can induce anemia. Concentrated saline injection depresses myocardial contractility and has volume-loading effects (6, 9). Our telemetry system is free of these problems.
The current used for resistivity measurements was distributed in a 2-mm radius around the catheter (see APPENDIX). The ratio of the radius (i.e., penetration depth) to the distance between the excitation electrodes was 3 (
2/0.6). This ratio is at odds with previously reported values, which were around or less than unity (6, 10, 26). Penetration depth is affected by the relation between the resistivity of the target tissue and that of the surrounding structure (26). This relation in our study was different from those in previous studies, which would be one reason for the discrepancy. Difference in shape and arrangement of the electrodes between our system and those previous studies would be another reason. Because the electrodes were placed very closely, stray capacitance between connecting wires could be a problem (31). The fact that resistivity values at 2 and 20 kHz were very close indicated that our titration method effectively removed the problem of stray capacitance (see APPENDIX). However, it might be better to incorporate techniques such as capacitance neutralization to completely prevent the problem, in case the capacitance were to significantly affect our titration accuracy in future long-term use, e.g., with increases in electrode impedance (31).
We used the dual-frequency excitation method previously described by Gawne et al. (8). Feldman et al. (6) combined measured resistivity of the myocardium with an analytic approach and estimated Gp from the conductance signals at 10 and 100 kHz. Although their method was completely independent of saline injection, it required measurement of myocardial resistivity with an additional four-electrode sensor.
Volumetric Accuracy and Reproducibility
We have verified the volumetric accuracy of our telemetry system by comparing SVtele with SVflow during inferior vena cava occlusions (Fig. 5A). The volumetric accuracy of the conductance catheter technique in the rat heart has been examined using a similar comparison (14, 23). Ito et al. (14) reported a very high and linear correlation (r = 0.97–0.99) between conductance-derived SV and SV measured by an electromagnetic flowmeter in rats. We also obtained a similar highly linear relation between SVtele and SVflow (Fig. 5B).
The reproducibility of our telemetry system (Table 1) is good enough for many applications, such as the study of LV remodeling in rats. This is because EDV has been reported to increase to 200% of the control value in rats with ischemic heart failure and in heart failure-prone rats (2, 7, 12).
Applications of the Telemetry System
The developed telemetry system enables detailed evaluation of cardiac function in small animals by eliminating the effects of anesthesia and acute surgical intervention (13, 22, 30). By using a single-beat estimation method to determine the ESPVR, our system would enable evaluation of the load-independent contractile index in conscious animals (24, 25). We validated pressure-volume signals only under control conditions in this study. The stability of the acquired data and the capacity of our system to monitor altered hemodynamics remain to be evaluated.
Our telemetry system is potentially useful for the long-term monitoring of LV function. We confirmed that our system was viable for up to 8 days in this study. However, further studies are required to definitively evaluate the longevity of the implants over a longer period of time (19). Thrombosis and infection would affect the morbidity and mortality associated with the chronic implantation of our system. Coating of the pressure-conductance catheter with anticoagulants and further miniaturization of the implant are under development to ameliorate such problems.
We adopted Bluetooth technology for telecommunication. Bluetooth is a wireless technology designed to allow low-cost, short-range radio links between mobile personal computers and other portable devices (18). While point-to-point connections are supported, Bluetooth technology allows up to seven simultaneous connections to be established and maintained by a single receiver (18). This unique feature of Bluetooth technology should be beneficial in experimental settings where a large population of animals in a single cage must be evaluated (16).
Limitations
The volume calibration factor was assumed to be unity on the basis of the preliminary experiment, where the conductance-derived volume was close to true syringe volume in the normal operating range for rats (Fig. 2). Georgakopoulos and Kass (9) noted that the relation was quite linear when the volume range was limited to the physiological operating range for mice. Hettrick et al. (11) also noted that conductance-derived volume was close to true syringe volume and was unity in a volume range. However, both groups and others noted that the relation was nonlinear when considered over a wider volume range (1, 9, 11, 20). In addition, the syringes have no Gp, whereas the rat heart does. It has been shown that Gp has significant effects on (11). Taken together, these findings suggest that it will be necessary to recalibrate when we apply our system to the rat LV in heart failure or other cardiac disorders, where drastic changes in ventricular volume and changes in the electrical properties of surrounding structures, i.e., change in Gp, are probable (1).
Values of EF in Table 2 are low for normal rats (5, 7). Other parameters of LV function are, however, within the normal range (5, 7) (Table 2). Dual-frequency derived Gp values from the rats in group 3 ranged from 1.8 to 3.3 mS (mean 2.3 ± 0.4 mS). The dual-frequency method slightly underestimated Gp in that range compared with the saline injection method (Fig. 4B). This might result in an apparent reduction of EF. To settle the discrepancy between EF and other functional parameters, it is necessary to compare the telemetric EF with the EF determined by other independent methods, such as echocardiography.
We were able to estimate in the LV cavity in normal-sized rats with the present catheter design (Fig. 1B, inset). However, the catheter design may not be applicable to smaller rats or mice, where the current distribution volume probably distributes outside the LV cavity. To apply our system to these small animals, further reduction of the interelectrode distance is required for measurement of .
Conclusion
A novel telemetry system was developed for measurements of LV pressure, volume, and ECG in conscious, freely moving rats. The system enabled us to accurately measure the LV pressure-volume relation with good reproducibility and without the harmful effects of anesthesia or acute surgical trauma in rats.
|
APPENDIX |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONAPPENDIXGRANTSREFERENCES |
|---|
We extracted frequency components of 20 kHz, 2 kHz, and low frequency (<2 kHz) by logical processing of digital signals. The analog signals were converted to digital signals at a sampling rate of 40 kHz. Twenty serial digital values were processed simultaneously (Fig. 7). We obtained the 20-kHz component on the basis of the difference between even- and odd-numbered digital values. We calculated an average of every 10 digital values. We obtained the 2-kHz component on the basis of the difference between the two averaged values of the former half and the latter half (average of 10 values each). We obtained the low-frequency component by averaging all 20 digital values. All this logical processing was performed by the programmable logic device (Fig. 1C).
|
.
First, we experimentally determined the current distribution volume of the four small electrodes for estimation of . We placed our pressure-conductance catheter at the center of plastic syringes of various sizes filled with diluted saline. Saline resistivity was matched to that of the blood (122 ·cm). We injected a constant current (20 kHz, 20 µA RMS) into the excitation electrodes (0.6 mm apart; Fig. 1B, inset) and measured voltage via the sensing electrodes. We present the relation between the measured voltage and the syringe diameter in Fig. 8. As demonstrated, with increasing syringe diameter, the voltage signal decreased and reached a minimum at a syringe diameter of 4 mm. This implied that most (>95%) of the current was confined to within the cylindrical diameter at which the voltage reached a minimum. From these data, we concluded that the current distribution volume was confined to within a 4-mm diameter around the catheter.
|
by a conversion formula. We determined the conversion formula experimentally by placing the catheter at the center of a plastic syringe with a diameter of 9 mm. Syringes were filled with diluted saline solutions with known resistivities in the range of those expected in rat blood (122 and 244 ·cm). Constant currents (20 and 2 kHz, 20 µA RMS) were injected between the excitation electrodes. We linearly related the measured RMS voltage to saline resistivity and used this relation as a conversion formula (Fig. 9).
|
|
GRANTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONAPPENDIXGRANTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONAPPENDIXGRANTSREFERENCES |
|---|
. Circ Res 82: 416–423, 1998.This article has been cited by other articles:
|
|
 |
|
  M. J. Faber, M. Dalinghaus, I. M. Lankhuizen, P. Steendijk, W. C. Hop, R. G. Schoemaker, D. J. Duncker, J. M. J. Lamers, and W. A. Helbing Right and left ventricular function after chronic pulmonary artery banding in rats assessed with biventricular pressure-volume loops Am J Physiol Heart Circ Physiol, October 1, 2006; 291(4): H1580 - H1586. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
