INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

ISCHEMIC PRECONDITIONING (IP) consists of one or more ischemia-reperfusion cycles of a given organ or tissue (23). This procedure has been shown to improve myocardial tolerance to a subsequent ischemic stress under experimental (28) and clinical (33) conditions. The effects of IP were also demonstrated in skeletal muscles of animals, where it appears to reduce the size of experimental infarcts (25, 26) and to enhance postischemic function (18).

The mechanism by which IP operates has not yet been elucidated. Several hypotheses have been put forward. These are mainly based on the release by the tissues of substances such as bradykinin, adenosine, nitric oxide, opioids, and free radicals. However, none of the latter appears to explain the effects of IP. Among the hypothesized mechanisms by which repeated ischemia could operate, changes in energy metabolism (28) and/or localized edema (34), were proposed. With regard to the latter, cellular stretch could trigger the release of active substances, hence the effects of IP (34).

To our knowledge, the effects of repeated ischemia-reperfusion cycles have not been assessed in human skeletal muscle. The purposes of the present study were to investigate possible changes of energy metabolism and interstitial fluid shifts in human gastrocnemius induced by IP. The novelty of the present approach is that the above mechanisms, possibly underlying IP, could be assessed by combining near-infrared spectroscopy (NIRS) (15, 20) with ³¹P-nuclear magnetic resonance spectroscopy (NMRS) and NIRS with ²³Na-NMRS, respectively.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Subjects

Group 1. Resting baseline absolute concentration of deoxyhemoglobin ([Hb]_b) and oxyhemoglobin ([HbO₂]_b) were determined in the gastrocnemius of 17 men [age 33.5 ± 8.5 (SD) yr, body weight (BW) 75.2 ± 8.5 kg] and of 20 women (age 29.7 ± 3.6 yr; BW 56.1 ± 7.3 kg) by a phase-modulated NIR photometer. Adipose tissue thickness (ATT; fat plus skin layers) was measured by a skinfold caliper.

Group 2. Simultaneous ³¹P-NMRS and concentration changes in Hb ([Hb]) and HbO₂ ([HbO₂]) were obtained throughout the adopted IP experimental protocol (see below) in a group of eight male subjects (age 38.0 ± 9.4 yr; BW 78.5 ± 16.7 kg) by a 1.5-T magnetic resonance spectrometer and by a continuous-wave NIR photometer, respectively.

Group 3. ²³Na-NMRS data were obtained in five different male subjects (age 37.4 ± 12.0 yr, BW 73.4 ± 15.9 kg) during an IP protocol the same as was adopted for group 2. In group 3, a control ²³Na-NMRS experiment was also carried out.

NIRS Measurements

[Hb] and [HbO₂] were obtained by a NIRO-500 (Hamamatsu Photonics, Hamamatsu City, Japan) continuous-wave photometer, which may be used in an NMR magnet. This device operates at four wavelengths: 773, 828, 853, and 914 nm. Optical densities (OD) for the four wavelengths were acquired with a sampling time of 2 s. [Hb] and [HbO₂] were then calculated from the experimental OD values by means of dedicated NIRO-500 software (11). The algorithm used is based on the modified Beer's law: A =  · c · d · B + G, where A is the measured attenuation in OD,  is the specific extinction coefficient of the absorbing compound measured in µM/cm, c is the concentration of the absorbing compound measured in µM, d is the distance between the optodes on the skin surface (cm), B is the differential pathlength factor, and G is a factor mainly dependent on the scattering parameter of the tissue, which, in the NIRO-500 software, is assumed to be constant. By the latter assumption, by using a system of four modified Beer equations, one for each wavelength, it is possible to obtain, by multilinear regression, [HbO₂] and [Hb]. The differential pathlength factor value adopted for the calf muscles was 5 (12).

Tissue ₂ Measurements

(1)

(2)

(3)

Thus Eq. 1 turns out to be

(4)

Mean ₂ (_{2 mean}) was calculated over 4-min intervals

(5)

The maximum error in _{2 mean} calculation has been assessed for the whole range of experimental values obtained in the present study by means of a mathematical simulation (unpublished data) and found to be <10%.

³¹P-NMRS Measurements

²³Na-NMRS Measurements

Relative concentration of total Na ([Na]; intracellular and extracellular) was detected with a standard one-pulse NMRS sequence. The transmitter-pulse duration and amplitude were chosen so as to maximize the in vivo ²³Na signal-to-noise ratio from the target sample. Acquisition parameters were 120 µs for the radio-frequency pulse duration, 512 points per FID, a spectral width of 5,000 Hz, and a 164-ms repetition time. A single spectrum consisted of the sum of 80 FIDs, resulting in an acquisition time of 13.12 s per spectrum. The FIDs were first filtered with an exponential function (exponential coefficient 3 Hz) to improve the signal-to-noise ratio.

Analysis of the spectra (²³Na-NMR spectra are characterized by only one peak) was performed by using the same software as for ³¹P-NMRS.

Experimental Protocol

Group 1 (n = 37). Absolute [HbO₂]_b and [Hb]_b measurements were performed in the right gastrocnemius muscle while the subject was seated with the lower legs down. The probe was placed on the medial line of the upper third of the gastrocnemius medialis. The data were collected over a time interval of 5 min after an initial 5-min readjustment period. The tracings were averaged to improve the signal-to-noise ratio. To make the measured values comparable among subjects, the photometer was tested and calibrated in a phantom before each measurement. An additional tracing, in the same phantom, was obtained at the end of each measurement to check the stability of the equipment. ATT was measured at the same location.

Group 2 (n = 8). After the control values at rest (5 min) were recorded, a sequence of three cycles, each consisting of 5-min ischemia followed by 5-min reperfusion, was performed in the right leg of the subject. The subject was lying supine inside the NMR magnet, with the calf placed over the ³¹P-NMR radio-frequency coil. The right leg was tightly fixed to the sliding board to avoid motion artifacts. The transmitting and receiving optodes of the NIR photometer were placed on the gastrocnemius medialis 2 cm proximal to the radio-frequency coil (10). Their relative distance (3 cm) was ensured by a rigid 3-mm-thick custom-made holder (29). Measurements were performed by a pair of optical fibers (length 6 m). Ischemia was induced by inflating a nonmagnetic cuff (width 15 cm) up to 350 mmHg around the leg, 1 cm distal to the groin. Continuous acquisition of ³¹P-NMRS and NIRS data was carried out simultaneously.

Group 3 (n = 5). The same protocol as for group 2 was repeated for ²³Na-NMRS. In addition, control ²³Na-NMRS measurements were carried out in separate sessions before the IP protocol. In both cases, the subjects were asked not to lie supine within 2 h before the experiment. The initial ²³Na-NMR spectra were acquired 10 min after the lying position was assumed.

Statistics

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

NIRS: Aerobic Metabolism

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Resting baseline absolute concentration of deoxyhemoglobin ([Hb]b; A) and oxyhemoglobin ([HbO2]b; B) in human gastrocnemius as a function of adipose tissue thickness (fat plus skin). , Men; , women.

Typical time courses of [Hb], [HbO₂], and [Hb]_tot during the IP protocol are shown in Fig. 2 for a typical subject. The horizontal solid bars represent the ischemic periods (ISPE). As expected, a constant decrease in [HbO₂] is accompanied by an increase in [Hb]. Individual tracings have similar patterns, but [Hb]_tot during ISPE (i.e., the intervals during which _{2 mean} was calculated) varies considerably among subjects. The largest observed [Hb]_tot was on the order of 10 µM, being within the concentration range (5 to 10 µM). The values are well within the limits imposed for the simulation. The [Hb]_tot values recorded at baseline level and at the end (after 35 min) of the IP protocol are essentially the same. These results, combined with the ²³Na-NMRS data, are required to assess interstitial fluid shifts (see DISCUSSION).

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Time course of absolute changes () in [Hb] (A), [HbO2] (B), and total hemoglobin concentration ([Hb]tot; C), and from baseline during three 5-min ischemia-reperfusion cycles (solid bars).

_{2 mean} in the muscle region of interest, as a function of ATT, calculated by Eq. 5, is shown in Fig. 3. Circles, stars, and crosses correspond to the first, second, and third ISPE, respectively. No _{2 mean} differences were found among the three ischemic cycles. The data on the left of the curve reflect _{2 mean} of mainly muscle, which could be estimated at ~8 µmol · 100 g¹ · min¹.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Mean rate of tissue O2 consumption (2 mean) as a function of adipose tissue thickness. , First ischemic period; *, 2nd ischemic period; ×, 3rd ischemic period.

³¹P-NMRS: Anaerobic Metabolism

View larger version (18K): [in this window] [in a new window]   Fig. 4.   A: phosphocreatine (PCr; ) and adenosinetriphosphate (ATP; ) concentrations (relative units, 100% = PCr baseline level). B: muscle intracellular pH as a function of time during three 5-min ischemia-reperfusion cycles (solid bars).

Muscle intramuscular Mg²⁺, as calculated from ³¹P-NMR spectra, appears to be constant throughout the experiment (0.52 ± 0.05 mM; not shown).

²³Na-NMRS: Assessment of Fluid Shifts

View larger version (26K): [in this window] [in a new window]   Fig. 5.   Muscle Na concentration ([Na]; relative units, 100% = baseline level) as a function of time. Subject's posture was shifted from standing to supine 10 min before acquisition of 1st spectrum. A: during three 5-min ischemia-reperfusion cycles (solid bars). B: control conditions.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

As indicated at the beginning of this study, one purpose of the study was to investigate possible total energy changes, both aerobic and anaerobic (lactic and alactic), in the human gastrocnemius during the IP protocol.

NIRS Measurements: ₂

1

1

Muscle myoglobin (Mb_tot) has an absorption spectrum similar to that of hemoglobin. Therefore, Mb_tot does interfere with the NIRS measurement of muscle. On the basis of values in human gastrocnemius, i.e., on average, 4 mg/g (24) or 41.9 µM, one can estimate its weight as a confounding factor at ~5-10% of the whole hemoglobin signal, an amount that may be considered negligible for the present calculation. However, the presence of myoglobin, whatever its concentration, does not influence ₂ calculation because the transformation of four molecules of oxygenated myoglobin to deoxygenated myoglobin is equivalent, from the optical standpoint, to the transformation of one molecule of HbO₂ to Hb. This leads in any case to the utilization of four molecules of O₂, and Eqs. 1 and 5 remain valid.

The use of the phase-modulated OMNIA photometer for the determination of [Hb]_b and [HbO₂]_b in group 1 would require that the tissue within the region of interest be homogenous. As appears in Fig. 1A, [Hb]_b values in two homogeneous regions of interest (mainly fat or mainly muscle, respectively) are practically identical. A possible optical artifact due to the heterogeneity of the investigated tissue layers would necessarily lead to average [Hb]_b levels different from the values found for mainly fat vs. mainly muscle. This rules out tissue heterogeneity as a possible confounding factor for [Hb]_b measurements. It necessarily follows that, if [Hb]_b values are reliable, then, on the basis of the OMNIA algorithm, [HbO₂]_b values must also be correct. This leads us to conclude that the measured [HbO₂]_b and [Hb]_b values reflect actual physiological conditions.

Absolute [Hb]_b and [HbO₂]_b values are required to calculate "true" _{2 mean} during IP by Eqs. 1 and 5 (see MATERIALS AND METHODS).

[Hb]_b and [HbO₂]_b cannot be determined by continuous-wave photometers of the type of NIRO-500 used for the present joint NIRS-NMRS protocols. However, even a device such as the phase-modulated OMNIA photometer, by which absolute [HbO₂]_b and [Hb]_b can be assessed, cannot monitor the fractional [Hb] and [HbO₂] due to blood volume shifts ([Hb]_tot) and due to variations in the Hb-to-HbO₂ ratio for any given [Hb]_tot in the region of interest. These changes could introduce an error in the _{2 mean} calculation. This is the reason a simulation was made. By the simulation it was proven that the use of Eqs. 4 and 5 yields reliable _{2 mean} data even in the presence of sizeable [Hb]_tot and Hb-to-HbO₂ ratio changes. Moreover, it was proven that Eqs. 4 and 5 are not highly dependent on [HbO₂]_b and [Hb]_b and that, in any case, the error in estimating _{2 mean} is <10% for the whole range of experimental values.

It is noteworthy that, apart from extreme ATT values (i.e., nearly mainly muscle or mainly fat), _{2 mean} takes into account different fractions of muscle and adipose tissue. However, in the present case, the fact that during IP _{2 mean} is the same at each ATT value proves that _{2 mean} also does not change for each of the two tissues under consideration. In fact, it would be highly improbable that muscle ₂ could totally compensate for adipose tissue ₂ at each ATT.

NMRS Measurements: Energy Metabolism

(6)

NMRS Measurements: Fluid Shifts

It necessarily follows that the observed decrease in Na at the end of the experiment (Fig. 5) is attributable only to a loss of interstitial fluid. This shift is probably due to the fluid redistribution to the upper part of the body when the subject is tilted from standing to supine (3). The behavior of the fluid content as a function of time in control conditions (Fig. 5B) is similar to that found by Berg et al. (3), even though the time constant is somewhat shorter: 26.4 vs. up to 37 min. This difference might be methodological. In fact, Berg et al. measured muscle volume changes by plethysmography and bioelectrical impedance analysis. In the present case, Na values represent only the fluid compartments, whereas the surrounding tissues are neglected.

Should the water shift in the gastrocnemius be only due to gravity, then Na level at the end of the IP protocol (35 min; Fig 5A) would be equal to the level found 20 min after the onset of the control measurement (Fig. 5B). In fact, during the ISPE the water cannot leave the legs because of the cuff, and the time available for the fluid shift would be 35  15 min = 20 min. The actual Na level after 20 min (Fig. 5B) was 86.35 ± 0.71%. This is significantly lower (P < 0.05) than the final level in the IP protocol recorded in Fig. 5A. This difference becomes significant only after the third ISPE. Thus, after IP, less water is lost than expected. A mechanism other than gravity must therefore affect muscle fluid redistribution, which leads to a relative increase, or more appropriately, to a lesser decrease, in the interstitial water content. This is at variance with a previous hypothetical interpretation by our group (6), whereby Na disappearance was attributable, in part, to the Na "trapping" (i.e., NMR invisible) in the interstitial space.

In conclusion, it was shown for the first time, to our knowledge, that 1) three 5-min ISPE followed by 5-min reperfusion intervals do not induce significant changes in the total energy consumption () in the human gastrocnemius; and 2) the same protocol is associated with a reduction in the postural fluid shift observed in the muscle when the body is tilted from the vertical to the supine position. The latter change obtained by combined ²³Na-NMRS and NIRS may be a precursor of edema, which is a possible determinant of IP. In addition to the above-mentioned main findings, the following results were obtained: 1) [Hb]_b values in resting muscle and adipose tissue are the same; 2) the _{2 mean} of the resting gastrocnemius has been confirmed to be on the order of ~8 µmol · 100 g¹ · min¹, whereas that of adipose tissue is ~2 µmol · 100 g¹ · min¹; 3) during repeated 5-min ischemia-reperfusion cycles, anaerobic metabolism does not contribute energy to the muscle; and 4) on assumption of the lying position, the decrease in the interstitial fluid in the gastrocnemius is characterized by a half-time of 24.6 min.