| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Cardiovascular Therapeutic Exercise Laboratory, Faculty of Rehabilitation Medicine, and 2Division of Cardiology, University of Alberta, Edmonton, Alberta; 3Cardiovascular Physiology and Rehabilitation Laboratory, Experimental Medicine, Faculty of Medicine, University of British Columbia, Vancouver, British Columbia
Submitted 5 July 2007 ; accepted in final form 21 August 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
O2p), heart rate (HR), stroke volume (SV), cardiac output (
), and skeletal muscle deoxygenation (HHb) kinetics were assessed during moderate-intensity ULKE exercise. Peak exercise and reserve O2p, , and systemic arterial-venous oxygen difference (a-vO2diff) were 23–52% lower (P < 0.05) in HTR. The reduced and a-vO2diff reserves were associated with lower HR and HHb reserves, respectively. The phase II O2p time delay was greater (HTR: 38 ± 2 vs. CON: 25 ± 1 s, P < 0.05), while time constants for phase II O2p (HTR: 54 ± 8 vs. CON: 31 ± 3 s), (HTR: 66 ± 8 vs. CON: 28 ± 4 s), and HHb (HTR: 27 ± 5 vs. CON: 13 ± 3 s) were significantly slower in HTR. The HR half-time was slower in HTR (113 ± 21 s) vs. CON (21 ± 2 s, P < 0.05); however, no significant difference was found between groups for SV kinetics (HTR: 39 ± 8 s vs. CON 31 ± 6 s). The lower peak O2p and prolonged O2p kinetics in HTR were secondary to impairments in both cardiovascular and skeletal muscle function that result in reduced oxygen delivery and utilization by the active muscles.
O2p) during peak aerobic exercise remains 50% lower than healthy individuals (13, 17, 29). Further, the kinetics of O2p following a step transition to moderate-intensity exercise (below the gas exchange ventilatory threshold) is markedly slower in HTR compared with healthy individuals (12, 22). The reason for the abnormal O2p on-kinetics and peak O2p has been attributed to impaired cardiovascular and skeletal muscle function, which results in reduced oxygen delivery and/or utilization by the active muscles (12–14, 17, 20, 22). Currently, our understanding of the factors contributing to the exercise limitation post-heart transplantation is based on data obtained during large muscle mass (2-legged bicycle) aerobic exercise. In healthy sedentary individuals, the capacity of skeletal muscle to receive blood flow exceeds the capacity of the heart's ability to supply blood when aerobic exercise is performed with one-third or more of total muscle mass (25). Importantly, if peak skeletal muscle perfusion and cardiac output () in HTR are 250 ml·100 g–1·min–1 (1) and 12.5 l/min (20), respectively, then aerobic exercise involving 
5 kg of muscle mass will exceed the capacity of the transplanted heart to pump blood to the active muscles. Thus, aerobic exercise performed with < 5 kg of muscle mass should reduce the cardiac limitation to exercise performance in HTR.
The primary aim of this investigation was to examine peak exercise and reserve (peak exercise – rest) cardiovascular function, skeletal muscle deoxygenation (HHb), and O2p during unilateral knee extension (ULKE) exercise in HTR and healthy controls (CON). A secondary objective was to assess heart rate (HR), stroke volume (SV), , skeletal muscle HHb, and O2p on-kinetics during moderate-intensity ULKE exercise. Our primary hypothesis was that a lower peak O2p during ULKE exercise in HTR vs. CON would primarily be related to lower peak systemic arterial-venous oxygen difference (a-vO2diff) and skeletal muscle HHb, and to a lesser extent, reduced peak . Our secondary hypothesis was that a delay in O2p on-kinetics in HTR during ULKE exercise would be associated with slower skeletal muscle HHb kinetics.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
The participants for this study included five clinically stable male HTR (mean ± SE, age: 53 ± 3 years; body mass index: 27 ± 2 kg·m2; time post transplant: 6 ± 4 years) and five age-, gender-, and activity-matched healthy CON (age: 53 ± 3 years; body mass index: 28 ± 1 kg·m2). The HTR participants were clinically stable and had no biopsy or clinical evidence of rejection. This investigation received approval from the University of Alberta Health Research Ethics Board (Biomedical Panel) and informed consent was obtained prior to study participation.
Experimental Protocol
Participants reported to our exercise laboratory on two separate occasions. On the first day, an incremental ULKE test was performed to determine the gas exchange ventilation threshold (2), peak and reserve O2p, HR, SV, , a-vO2diff, and skeletal muscle HHb. The test was performed on a custom built knee-extensor ergometer as previously described (1), with the dominant limb used for exercise. Initial practice sessions were performed to allow for protocol familiarity, and to ensure that the exercising limb remained passive during the knee flexion phase by allowing the momentum of the flywheel to pull the participant's limb back to the resting position. The incremental test began with 0-watt kicking for 1 min and increased by 3–5 watts/min to volitional exhaustion, or until a cadence of 50 contractions/min was no longer attainable.
On a second day, four repetitions of a square-wave protocol were conducted with a minimum rest of 20 min or greater between each exercise bout, to ensure HR, blood pressure, and O2p reached pre-exercise baseline values. The square-wave protocol began with a 3-min, 0-watt kicking baseline, followed by an unannounced step increase in work rate corresponding to 50% peak O2p (
90% of the ventilatory threshold) for 5 min. The cadence during this test was strictly maintained at 50 contractions/min.
Measurements
Pulmonary oxygen uptake and cardiovascular function.
Expired gas analysis was obtained at rest and during exercise by means of a commercially available metabolic measurement system (Parvomedics, Salt Lake City, UT). A 12-lead electrocardiogram was monitored, and systolic (SBP) and diastolic (DBP) blood pressure (cuff sphygmomanometer) were recorded. The rate of change in thoracic bioimpedance (dZ/dt), first and second heart sounds, ejection time, and HR via an integrated electrocardiogram were sampled at 600 Hz (Minnesota Impedance Cardiograph, model 304B; Surcom, Minneapolis, MN). The individual dZ/dt waveforms were then measured offline independently by two investigators, and SV was calculated by means of Bernstein's formula (4). Cardiac output (SV x HR), mean arterial pressure [MAP = 1/3 (SBP – DBP) + DBP], systemic a-vO2diff (O2p/), and systemic vascular resistance (SVR = MAP/) were also calculated. The highest O2p over a 30-s period defined the peak score, while peak HR, SV, and were averaged over 5 cardiac cycles within the same period.
Skeletal muscle oxygenation. Skeletal muscle oxygenation was measured with a NIRO 300 (Hammamatsu Photonics, Japan) spatially resolved near-infrared oxygenation spectroscopy monitor that employed four laser diodes to pulse near-infrared light at 775, 810, 850, and 905 nm and a photomultiplier tube for near-infrared light detection (15). Emission and detection probes were placed midway between the greater trochanter and lateral epicondyle of the femur on the belly of the exercising vastus lateralis muscle. Probes were fixed in a black holder ensuring a 5-cm distance between the light source and detection probe, secured to the leg with tape, wrapped in a black cloth to restrict external light sources and loss of transmitted light, and wrapped with a tensor bandage to further secure the probes and ensure free limb movement. The intensity of incident and transmitted light signals was sampled continuously at 2 Hz. A differential pathlength factor of 3.83 was used as previously described (11). Measurement of near-infrared light attenuation provides information about the relative concentration changes of oxygenated (HbO2), HHb, and total (Hbtot) hemoglobin and/or myoglobin in the small arterioles, capillaries, and venules at the field of interrogation of the muscle. In the present investigation, tissue oxygenation index (TOI = HbO2/Hbtot) was used as a measure of tissue O2 saturation (6), while the change in vastus lateralis muscle HHb was used as an index of skeletal muscle O2 extraction (17).
Cardiopulmonary and skeletal muscle deoxygenation kinetic analysis.
Breath-by-breath O2p, beat-by-beat HR, SV, , and instantaneous near-infrared spectroscopy-derived HHb data were sampled and recorded continuously throughout exercise. Data points were removed if greater than 3 standard deviations from the local mean (16) and interpolated to 1-s intervals. Data from the four square-wave protocols were then time aligned and averaged to yield a single response profile for respective variables. These data were averaged into 5-s time bins to further clarify the response profiles.
The onset of phase II O2p kinetics was carefully determined from the phase-I-phase-II interface as previously described (30). Data for phase II O2p were then fit from the phase-I-phase-II interface to 180 s into exercise with a monoexponential equation of the form:
 |
O2p at any time (t), b is the baseline value of Y over 60 s prior to the step increase in work rate, A is the amplitude change in Y above the baseline, 
is the time to reach a 63% change in Y, and TD is the time delay prior to the exponential increase in Y. Additionally, we calculated and reported the amplitude of the phase II O2p response as the amplitude in O2p starting from the onset of the exponential curve (phase I-phase II interface) to 180 s into exercise (5).
The same monoexponential model above was used for determining kinetic parameter estimates for SV and . However, curve-fitting parameter estimates were initiated at time 0 with a time delay to model the evolution of the responses from exercise onset (i.e., the time of the step increase in work rate). Given that HR increased in a linear fashion in the HTR group, the halftime of this response was used to evaluate the time course of HR. The kinetics of near-infrared spectroscopy-derived HHb data were also determined by means of the equation above. The exponential increase in HHb was modeled following a time delay, which was defined as the first data point greater than 1 standard deviation above the mean baseline value (11).
A Levenberg-Marquardt iterative procedure was used for curve fitting, where the best fit was defined by minimization of the residual sum of squares (Origin 7.5, OriginLab, Northampton, MA). Two investigators independently determined the goodness of fit of the derived nonlinear regressions to the measured data by 1) visual inspection of the curve for appropriateness of fit, 2) visual inspection of the residuals for clustering and systematic deviations from the x-axis, 3) a sudden increase in the value, and 4) by demonstration of a local threshold in the reduced chi-squared value.
Statistical Analysis
Statistical analysis was performed with independent t-tests for between-group comparisons at rest, peak exercise, reserve function, and derived curve-fitting parameters. Data are expressed as mean ± SE, and P < 0.05 was considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Resting SV was significantly lower in HTR (60 ± 2 ml/beat) compared with CON (76 ± 4 ml/beat, Table 1). No significant difference was found between groups for any other resting measure (Table 1).
|
Peak and reserve O2p, , and a-vO2diff were 23–52% lower (P < 0.05) in HTR than in CON (Table 1 and Fig. 1). The reduced reserve was due to a lower HR reserve, as SV reserve was not different between groups (Fig. 1). Peak exercise and reserve SVR and HHb, as well as peak TOI, were not significantly different between groups (Table 1 and Fig. 1). Peak and reserve MAP were significantly different between groups (Table 1 and Fig. 1). Finally, the trend line for the HHb/O2p relationship appeared greater in HTR than in CON at any given submaximal power output (Fig. 2).
|
|
During baseline (0 watt) exercise, HR was higher (HTR: 98 ± 2 beats/min vs. CON: 81 ± 7 beats/min, P < 0.05) and SV was lower in HTR (HTR: 61 ± 3 ml/beat vs. CON: 77 ± 5 ml/beat, P < 0.05). No significant difference was found for O2p (HTR: 518 ± 54 ml/min vs. CON: 493 ± 35 ml/min), (HTR: 6.0 ± 0.3 l/min vs. CON: 6.2 ± 0.3 l/min), or HHb (HTR: 2.7 ± 0.6 µM vs. CON: 0.8 ± 1.1 µM) during the 0-watt kicking baseline.
The power output during constant load exercise was significantly lower in HTR (18 ± 2 watts) compared with CON (31 ± 3 watts). Figure 3 illustrates O2p responses and monoexponential curve fits for a representative HTR and CON. Cardiac output, O2p, and HHb kinetics were significantly slower in HTR compared with CON (Table 2). Using the formula from Lamarra et al. (16), we calculated the 95% confidence interval for the O2p time constant to be ±6 s for HTR and ±7 s for CON. The HR halftime was significantly slower in HTR (113 ± 21 s) than in CON (21 ± 2 s); however, no significant difference was found between groups for SV kinetics (HTR: 39 ± 8 s vs. CON: 31 ± 6 s). Finally, the phase II O2p amplitude was lower (P < 0.05), while the phase II O2p time delay and the O2p and mean response times were greater (P < 0.05) in HTR (Table 2).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
O2p found in HTR during ULKE exercise was the result of a lower peak , systemic a-vO2diff, and skeletal muscle O2 extraction (as measured by near-infrared spectroscopy-derived HHb). Second, the prolonged O2p kinetics during moderate-intensity ULKE exercise was associated with slower HR, , and HHb kinetics. Accordingly, the abnormal O2p on-kinetics and peak O2p during small muscle mass exercise found in heart transplant recipients are due to both an O2 delivery limitation and skeletal muscle O2 utilization limitation. Impaired Pulmonary Oxygen Uptake During Peak ULKE Exercise Post-Heart Transplantation
A characteristic feature post-heart transplantation is the persistent impairment in exercise tolerance despite normal left ventricular systolic function (13, 14, 27). Kao and colleagues (13, 14) and Mettauer et al. (20), using right heart catheterization and expired gas analysis during two-legged bicycle exercise, demonstrated that the reduced peak O2p was primarily due to a lower , and to a lesser extent, a lower a-vO2diff. In turn, the blunted peak was the result of a slower chronotropic response and lower end-diastolic volume, as peak ejection fraction was similar between HTR and CON (13, 14). Our data confirm and extend prior study findings by demonstrating that peak exercise and reserve O2p are 40–50% lower in HTR during aerobic exercise involving a small muscle mass. Our finding that reserve is secondary to a blunted HR reserve (Fig. 1) is also consistent with previous study findings (13, 14). However, in contrast to others and our own hypothesis, the lower peak O2p during ULKE exercise was the result of reductions in both peak (–23%) and a-vO2diff (–23%).
The disparity between our findings and those of others may be due to different muscle mass involvement during aerobic exercise. Specifically, Andersen and Saltin (1) demonstrated that the capacity of the skeletal muscle to accommodate blood flow exceeds the upper limit of during large muscle mass aerobic exercise. Given that peak is 40% lower in HTR than in healthy sedentary individuals (13, 14), only 5 kg of muscle mass would need to be engaged in exercise for skeletal muscle perfusion capacity to exceed the upper limit of the cardiac allograft's ability to supply blood to the systemic circulation. Consistent with this hypothesis, our peak O2p (11.5 ml·kg–1·min–1), HR (119 beats/min), SV (78 ml/beat), and (9.4 l/min) during ULKE exercise (estimated quadricep muscle mass = 1.2 kg) are similar to those reported by Kao et al. (13) for HTR during peak two-legged bicycle exercise (peak O2p: 11.1 ml·kg–1·min–1; HR: 113 beats/min; SV: 82 ml/beat; : 9.0 l/min). Thus, the central (cardiac) limit to exercise performance that occurs with large muscle mass aerobic exercise is less prominent during small muscle mass exercise, and as a result, a peripheral limitation to peak exercise performance appears to play an equally important limiting role.
Cardiopulmonary and Skeletal Muscle Deoxygenation Kinetics During Moderate-Intensity ULKE Exercise
Several prior investigators have reported that O2p kinetics during the onset to moderate-intensity exercise is delayed in HTR compared with age-matched healthy individuals (9, 10, 12, 17, 18, 21, 22). This finding has been attributed to a reduction in O2 supply to the active muscles (22) and to impaired skeletal muscle oxidative metabolism (12). Paterson et al. (22) examined the role that warm-up exercise had on O2p on-kinetics during moderate-intensity bicycle exercise. The delayed O2p kinetics found in HTR during the initial exercise on-transient was mitigated (i.e., a faster response) during the second exercise bout. Given that HR was 8% higher prior to initiating the second exercise test, the sluggish O2p kinetics were attributed in part to a reduction in O2 delivery. In a similar study, Grassi et al. (12) examined the role of warm-up exercise on O2p and (impedance cardiography) on-kinetics. Contrary to the findings by Paterson's group, the faster kinetics during the second on-transient was not associated with a change in O2p on-kinetics (12), leading the investigators to conclude that abnormalities in skeletal muscle oxidative metabolism results in delayed O2p on-kinetics in HTR (12). Our data extend prior investigations by showing that the delayed O2p kinetics in HTR vs. CON during moderate-intensity ULKE exercise was associated with a prolonged (cardiodynamic) phase II O2p time delay and slower phase II O2p kinetics, which contributed to an overall slower O2p mean response time (Table 2). However, this finding contrasts those of Borrelli et al. (5) and Matteauer et al. (21), who reported that the cardiodynamic component of O2p, and not the phase II O2p response, in HTR during cycling exercise accounted for slower overall O2p kinetics. In the present investigation during small muscle mass exercise, the delayed phase II O2p time delay was likely due to the slower HR kinetics at exercise onset, while the prolonged phase II O2p kinetics were associated with both slower and HHb on-kinetics (Table 2).
Impaired Pulmonary Oxygen Uptake On-Kinetics and Peak Pulmonary Oxygen Uptake: Role of Abnormal Cardiovascular Function
The lower peak O2p and delayed O2p kinetics that we found in HTR during ULKE exercise were due, in part, to impaired cardiovascular function. Specifically, the lower peak and slower on-kinetics are secondary to the blunted HR and chronotropic reserve associated with cardiac denervation (13, 14). Posttransplant diastolic dysfunction manifested as prolonged acceleration of left ventricular relaxation (23), and increased diastolic passive chamber and myocardial stiffness (13, 14) may also reduce preload and reserve. Indeed, Mettauer et al. (21) found that the delayed O2p phase I duration was positively related to isovolumic relaxation time in HTR. Although we did not assess diastolic function, if the absolute change in ejection fraction from rest to peak exercise was 8% (14), then the increase in estimated end-diastolic volume (+15%) would be greater than the decline in estimated end-systolic volume (–8%) during ULKE exercise. Thus, the preserved SV reserve and SV kinetics during ULKE exercise (Fig. 1 and Table 2) may be due to greater utilization of the Starling mechanism. A final reason for the blunted peak and reserve exercise is that it may be due to an increased afterload associated with pre- and posttransplant vascular dysfunction. For example, peak SVR is 45–70% higher while SVR reserve is 15–35% lower in HTR than in normal individuals during bicycle exercise (13, 14, 20). Consistent with these findings, peak SVR was higher and SVR reserve was lower in HTR than in CON during ULKE exercise. A consequence of abnormal vascular function is an associated reduction in exercise capacity, as Bussieres et al. (7) found that HTR with the lowest peak O2p also had the highest peak exercise SVR. Taken together, the lower peak O2p and O2p on-kinetics may be secondary to an impaired and SVR that result in a reduction in O2 delivery to the exercising muscles.
Impaired Pulmonary Oxygen Uptake On-Kinetics and Peak Pulmonary Oxygen Uptake: Role of Abnormal Skeletal Muscle Oxygenation
The abnormal skeletal muscle morphology, histology and metabolism associated with heart failure syndrome may not be completely reversed after heart transplantation. Bussieres et al. (8) and others (8, 24, 26) reported that the size and percent of fatigue resistance (type I) oxidative muscle fibers and oxidative enzyme activity is reduced in HTR. Stratton (27), using 31P magnetic resonance spectroscopy, found that abnormal skeletal muscle bioenergetics during submaximal forearm exercise in individuals with heart failure do not improve in the early (mean = 4 mo) or late (mean = 15 mo) period following heart transplantation. A possible consequence of the reduced oxidative fibers and enzymes and abnormal oxidative metabolism is reduced O2 utilization by the active muscles. Indeed, our finding of a lower a-vO2diff and HHb during peak ULKE exercise in HTR vs. CON is consistent with previous studies involving large muscle mass aerobic exercise (13, 14, 17, 20). We extend these prior research findings by revealing that the time course of near-infrared spectroscopy-derived HHb is significantly delayed in HTR vs. CON during the transition to moderate-intensity small muscle mass exercise. Taken together, our findings suggest that the reduced peak O2p and delayed O2p on-kinetics during ULKE are also mediated, in part, by abnormal skeletal muscle function and metabolism resulting in reduced O2 utilization by the exercising muscles.
Limitations
A limitation of this investigation is that resting and exercise SV and were indirectly determined via impedance cardiography. Moreover, the a-vO2diff was indirectly measured as O2p divided by . Belardinelli et al. (3) have shown, in individuals with normal and impaired left ventricular systolic function, that impedance cardiography determined was well correlated (r = 0.9) with values obtained from thermodilution and direct Fick methods. Despite this previous report, a limitation of impedance cardiography during exercise is movement-related artifact (28); however, we were able to eliminate this potential problem by having the participants isolate any movement to the exercising limb only. Further, the peak in our CON group was similar to that reported by Magnusson et al. (19) for healthy males during peak ULKE exercise. Also, our peak a-vO2diff values are similar to those previously reported for HTR (20). Another limitation was that resting or exercise limb blood flow was not measured. However, given that HTR had a greater impairment in peak exercise and reserve and SVR, we would expect HTR to have lower limb blood flow compared with the CON group. Indeed, a consequence of the reduced limb blood flow was that skeletal muscle O2 extraction at any given submaximal O2p was higher in HTR than CON (Fig. 2).
Summary
Heart transplant recipients have a severe and marked reduction in peak O2p and prolonged O2p on-kinetics during ULKE exercise. Unlike large muscle mass aerobic exercise, the lower peak O2p that we found during small muscle mass exercise was secondary to equal reductions in both peak exercise and a-vO2diff. Further, the abnormal peak and reserve is due to a blunted HR reserve, while the lower peak a-vO2diff is due to reduced HHb. Finally, the delayed O2p on-kinetics is associated with prolonged HR, , and near-infrared spectroscopy-derived HHb kinetics.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
|
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 |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
O2 kinetics reveal a central limitation at the onset of subthreshold exercise in heart transplant recipients. J Appl Physiol 88: 1228–1238, 2000.This article has been cited by other articles:
|
|
 |
|
  P. A. Sperandio, A. Borghi-Silva, A. Barroco, L. E. Nery, D. R. Almeida, and J. A. Neder Microvascular oxygen delivery-to-utilization mismatch at the onset of heavy-intensity exercise in optimally treated patients with CHF Am J Physiol Heart Circ Physiol, November 1, 2009; 297(5): H1720 - H1728. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Scott, B. T. A. Esch, M. J. Haykowsky, D. E. R. Warburton, M. Toma, A. Jelani, D. Taylor, I. Paterson, D. Poppe, Y. Liang, et al. Cardiovascular responses to incremental and sustained submaximal exercise in heart transplant recipients Am J Physiol Heart Circ Physiol, February 1, 2009; 296(2): H350 - H358. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Marconi, M. Marzorati, P. Cerretelli, R. J. Shephard, E. S. Prakash, B. Grassi, F. Lanfranconi, S. Porcelli, A. Ferri, E. Borrelli, et al. Re: Cardiac denervation does/does not play a major role in exercise limitation after heart transplantation. J Appl Physiol, February 1, 2008; 104(2): 565 - 566. [Full Text] [PDF]
|
|
|
|
|
|
B. Geny, R. Richard, J. Zoll, A. Charloux, and F. Piquard Last Word on Point:Counterpoint: Cardiac denervation does/does not play a major role in exercise limitation after heart transplantation J Appl Physiol, February 1, 2008; 104(2): 568 - 568. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
