- Email: [email protected]
Near-infrared spectroscopy during peripheral vascular surgery
PII: S0967-2109(97)00015-X
Cardiovascular Surgery, Vol. 5, No. 3, pp. 304–308, 1997 1997 The International Society for Cardiovascular Research Published by Elsevier Science Ltd. Printed in Great Britain 0967–2109/97 $17.00 + 0.00
Near-infrared spectroscopy during peripheral vascular surgery J. P. Eiberg*, T. V. Schroeder*, K. C. Vogt* and N. H. Secher† *Department of Vascular Surgery, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark and †Department of Anaesthesia, the Copenhagen Muscle Research Center, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark Near-infrared spectroscopy was performed perioperatively on the dorsum of the foot in 14 patients who underwent infrainguinal bypass surgery using a prosthesis or the greater saphenous vein. Dual-wavelength continuous light spectroscopy was used to assess changes in tissue saturation before, during and after the operation. Following the use of peripheral vascular grafts an immediate postoperative increase in tissue saturation of median 28 (range 210 to + 81) arbitrary units was noted (P , 0.01), Following distal clamping of the common femoral artery, maximal ischaemia corresponding to a median reduction in tissue saturation of 61 (range 6–94) units was reached after 26 (range 8–95) min (P , 0.01). The maximal tissue saturation following declamping was median 27 (range 216 to +100) units higher than the preoperative level (P , 0.01) and was reached after median 42 (range 8–125) min, The results indicate that near-infrared spectroscopy is appropriate for perioperative monitoring during vascular grafting. 1997 The International Society for Cardiovascular Surgery Keywords: near-infrared spectroscopy, vascular disease, vascular bypass surgery, perioperative oxymetry
Near-infrared spectroscopy was introduced for measuring cerebral oxygenation non-invasively [1, 2], but is also used when estimating changes in tissue oxygenation of human skeletal muscles, after application of a tourniquet and during exercise and recovery [3, 5]. Recently, the technique was performed to monitor oxygen consumption measurement in the calf, showing a reduced oxygen consumption among patients with peripheral vascular disease, compared with that in normal controls [6]. The aim of this study was to evaluate the prospects of near-infrared spectroscopy for continuous perioperative monitoring of vascular grafting, using peripheral tissue oxygenation as an indicator of the success of revascularization. Many failures of the femoropopliteal bypass technique occur within the first 30 days following operation, with a reported rate
Correspondence to: Dr J.P. Eiberg, c/o T.V. Schroeder, Department of Vascular Surgery, Rigshospitalet, University of Copenhagen, DK2100 Copenhagen Ø., Denmark
304
of early occlusion from 3% to 21% [7–11]. Some of these grafts could be salvaged before they thrombose, providing early objective evidence of the failing graft. Near-infrared spectroscopy has so far not been investigated on patients undergoing vascular surgery. During vascular bypass surgery, a correlation between tissue oxygenation and reproducible events such as distal clamping, ischaemia and reperfusion, during the insertion of peripheral bypass grafts, was evaluated.
Patients and methods Patients Nine male and five female patients (median age 71, (range 51–83) years) with atherosclerotic arterial obstructive disease and undergoing infrainguinal bypass surgery with implantation of a prosthesis (five patients) or greater saphenous vein (nine patients) were studied. The indication for surgery was acute ischaemia, with symptoms of less than 7 days in CARDIOVASCULAR SURGERY
JUNE 1997 VOL 5 NO 3
Near-Infrared spectroscopy during peripheral vascular surgery: J. P. Eiberg et al.
three patients. The remaining 11 patients presented with chronic ischaemia lasting for at least 1 month, in one patient resulting in gangrene, in seven patients ischaemic ulceration, and in three patients rest-pain or claudication. The median ankle: brachial pressure index was 0.33 (range 0–0.72) preoperatively and 0.77 (range 0.35–1.09) on the first postoperative day. The proximal anastomoses were to the common femoral artery, while the distal anastomoses were to the popliteal vessels in six cases and to infrapopliteal vessels in eight cases. Four patients were operated on under general anaesthesia and 10 patients under epidural analgesia. The study was approved by the Ethics Committee of Copenhagen (Journal number KF-01-342/94) and informed consent was obtained from all patients. Near-infrared spectroscopy In vivo spectroscopy makes use of the fact that photons in the near-infrared spectrum are transmitted in tissues and mainly absorbed by oxygenated and deoxygenated haemoglobin, cytochrome aa3 and myoglobin [12, 13]. Both oxygenated and deoxygenated haemoglobin absorb near-infrared light at 800 nm, while near-infrared light at 760 nm is absorbed mainly by deoxygenated haemoglobin [4]. Beer–Lambert’s law dictates that the fate of the photons is a combination of absorption and scattering, i.e. a photon that changes direction but maintains its speed [14, 15]. Beer–Lambert’s law is expressed as: I = I0(10 2 ecL) where I and I0 represent the intensities of the emergent and incident light, respectively, e the specific absorption coefficient for the compound in question, c its concentration and L the optical pathlength of the absorbing solution [6]. The optical pathlength of light travelling in a medium depends very much upon the number of scattering events and their mutual distance. The pathlength is therefore unknown in biological tissue. The absorbance of near-infrared light in the tissue, and thereby tissue oxygenation, can be estimated only in relative terms, and not in absolute terms. A dual-wavelength spectroscope developed for clinical measurements of cerebral oxygenation was used (Invos 3100, Somanetics, Troy, Wisconsin, USA). The equipment includes a sensor and a central unit containing a display and a computer. The sensor to be placed on the patient is not sterilized and consists of two near-infrared diodes used as light sources and two silicon light detectors, all housed in an adhesive plastic holder for firm attachment to the patient’s skin, The source–detector separation was 30 and 40 mm, respectively, for the two diode– detector pairs, giving an average mean depth of photon migration into the tissue of approximately 15 and 20 mm [15]. Spectral analysis showed that the light CARDIOVASCULAR SURGERY
JUNE 1997 VOL 5 NO 3
emitted was composed of near-infrared wavelengths of 733 and 803 nm. Before anaesthesia induction, the near-infrared sensor was placed on the dorsum of the foot over the extensor digitorum brevis muscle of the ischaemic leg. Saturation was monitored continuously with the near-infrared spectroscope from the start of the anaesthesia to 45 (range 11–193) min postoperatively. Saturation was recorded at intervals of 1 to 3 min, to create a trend curve with special attention to clamping and revascularization of the graft. Reproducibility In six age-matched atherosclerotic subjects, measurements during tourniquet-induced ischaemia and recovery were performed twice in order to evaluate the reproducibility. Each measurement included reattachment of the sensor. The span was observed from minimal saturation following 4 min of tourniquet inflation, to maximal saturation during postischaemic recovery. Data processing In order to compare the relative measures of tissue oxygenation, obtained in individual patients, measurements were calibrated. The saturation determined 20 min after occlusion of the common femoral artery was regarded as zero (0 arbitrary units). The maximal tissue saturation during the hyperaemia following declamping was defined as 100 arbitrary units. Statistical evaluation included Mann–Whitney’s test to compare groups and Wilcoxon’s signed rank sum test to compare matched pairs. Data are presented as median, with the range in brackets. A P-value of 0.05 was considered significant.
Results Following proximal clamping, a drop in tissue oxygenation of 61 (6–90) units was observed (P , 0.01) (Figure 1). The time taken to reach maximal ischaemia was 26 (8–95) min. Declamping after implantation of the bypass was followed by a rapid increase in tissue oxygenation overshooting the preoperative level by 27 ( 216 to +100) units (P , 0.01). This peak was reached after 42 (8–125) min following declamping, and was not significantly different from the tissue saturation at the end of the operation (P = 0.05). Thus, our major finding was an increase in tissue oxygenation following the bypass procedure of 28 ( 210 to +81) units by the end of the operation as compared with the preoperative level (P , 0.01). At 1 and 2 h postoperatively, the increase in tissue saturation was 17 ( 232 to +72) units (P , 0.03) and 1.8 ( 27 to +72) units (P , 0.02), respectively, as compared with the preoperative level. In one case, 305
Near-Infrared spectroscopy during peripheral vascular surgery: J. P. Eiberg et al.
Figure 1 Relative changes (median: solid line; range: vertical bar; inter-quartile range: horizontal bars) in tissue saturation during peripheral bypass surgery performed by near-infrared spectroscopy (n = 14). The difference between preceding tissue saturations has been tested and denoted * if significant. Tissue saturations have been denoted # if significantly different from those at the beginning of the operation (P , 0.05). Dotted lines indicate 95% confidence interval
extensive cyanosis of the forefoot caused a period of unstable measurements. There was no significant difference in tissue saturation between patients operated on under epidural analgesia as compared with general anaesthesia. Also, there was no correlation between the postoperative level of tissue oxygenation as compared with the technique of operation, nor with the duration of symptoms. During follow-up, after 26 (1–76) days, there were four graft failures, occurring between 24 hours and 9 days after operation. Among the four failing grafts it was not possible to find any significant immediate postoperative saturation pattern identifying the imminent occlusion. In the reproducibility study the saturation span ranged from 7 to 22 units, with a mean of 14 units (SD error of difference = 1.9).
Discussion With near-infrared spectroscopy the metabolic state of the tissue can be assessed non-invasively during ischaemia and revascularization [4, 12, 16]; consequently, the technique was applied to monitoring in vascular surgery. An immediate postoperative increase in tissue oxygenation was found following peripheral bypass grafting, overshooting the preoperative level by 28 units. Moreover with near-infrared spectroscopy it 306
was possible to demonstrate the correlation between tissue saturation and haemodynamic events such as proximal clamping and reperfusion. Tissue saturation did not indicate that the patients with acute lower-extremity ischaemia had a better result following grafting procedure than those patients with chronic lower extremity ischaemia. This was expected on account of the higher degree of collateral vascularization within the chronic ischaemic limbs. The reproducibility study, in which the light sensor was applied twice, showed the method to be reasonably reproducible. Total deoxygenation was not obtained during 4 min of ischaemia, but a more prolonged experimental ischaemia in order to simulate the intraoperative ischaemia without anaesthesia was unsuitable. Ideally, the near-infrared sensor should be mounted above a larger and well-defined muscle, thereby excluding scattering and near-infrared signals not originating from the haemoglobin in the muscle tissue. In the present study, performed during peripheral vascular surgery, the probe had to be mounted outside of the surgical field. Thus, the sensor was placed above the relatively small muscles on the dorsum of the foot, even though the tissue was irregular. Using a calibration routine, the saturation reflecting zero was determined over a prolonged perCARDIOVASCULAR SURGERY
JUNE 1997 VOL 5 NO 3
Near-Infrared spectroscopy during peripheral vascular surgery: J. P. Eiberg et al.
iod of total intraoperative ischaemia. Using any other routine of calibration a similar degree of ischaemia could not have been induced. By using a calibration routine, obtained over a wide range of saturations, the measurements become more sensitive to changes in the actual oxygenation and less dependent upon the properties of the underlying tissue. Using transcutaneous near-infrared spectroscopy to quantify, muscular oxy- and deoxyhaemoglobin, the assumption must be made that skin and subcutaneous fat contribute a negligible amount to the signal. Germon and colleagues [17] found that scalp ischaemia affects the Invos Cerebral Oxymeter when applied to the frontal area, suggesting that skin perfusion could contribute to the signal by 18%. Piantadosi and co-workers [16] found that the response from the skin during skin- and muscle-hypoxia contributed less than 5% using a six-wavelength spectrometer. In the present study, one included case with extensive cyanosis of the foot during distal as well as proximal clamping, caused unstable measurements following the colour shift of the skin. The spectroscope used in the present study detects the intracranial changes in oxygen saturation when applied over the forehead [17–19]. By the use of two wavelengths and different interoptode distances, it is assumed that the absorption caused by the superficial and extracranial compartments may largely be eliminated [17]. Using the apparatus on the foot, the same assumptions can be made, excluding the eventual contributions from the skin and subucutaneous fat from the signal. As no cases of graft failure occurred during the near-infrared monitoring, the authors would not suggest that the technique is the appropriate tool for postoperative monitoring, but would only suggest that the demonstrated immediate increase in tissue oxygenation following revascularization implies that a significant decrease in tissue oxygenation might occur as evidence of an early graft occlusion. An instrument which is easy to use and allows continuous measurements of tissue oxygenation could be a valuable supplement to the present methods of detecting graft failure, such as loss of pedal pulses and decreasing ankle:brachial pressure index. In order to assess this applicability, the near-infrared monitoring period must be expanded to include the first postoperative days. Two advantages of the instrument used are the simplicity of the sensor placement and the feasibility of recording the tissue saturation, without additional calculations. These two conditions must be fulfilled for a more widespread use of near-infrared spectroscopy as routine continuous patient monitoring.
CARDIOVASCULAR SURGERY
JUNE 1997 VOL 5 NO 3
Acknowledgements This work was supported by grants from the P. Carl Petersens Foundation. We wish to thank senior lecturer E. Veje, H.C. Ørsted Institute, University of Copenhagen, for useful technical assistance and E. Riis Andersen, Vingmed Denmark, for providing the Somanetic Invos 3100 Cerebral Oxymeter.
References 1. Wyatt, J. S., Cope, M., Delpy, D. T., Wray, S. and Reynolds, E. O. R., Quantification of cerebral oxygenation and haemodynamics in sick newborn infants by near-infrared spectrophotometry. Lancet, 1986, ii, 1063–1065. 2. Wyatt, J. S., Cope, M., Delpy, D. T., Richardson, C. E., Edwards, A. D., Wray, S. and Reynolds, E. O. R., Quantitation of cerebral blood volume in human infants by near-infrared spectroscopy. Journal of Applied Physiology, 1990, 68(3), 1086–1091. 3. Hampson, N. B. and Piantdosi, C. A., Near infrared monitoring of human skeletal muscle oxygenation during forearm ischaemia. Journal of Applied Physiology, 1988, 64(6), 2449–2457. 4. Mancini, D. M., Ferraro, N., Nazzaro, D., Chance, B. and Wilson, J. R., Respiratory muscle deoxygenation during exercise in patients with heart failure demonstrated with near-infrared spectroscopy. Journal of the American College of Cardiology, 1991, 18, 492–498. 5. Hamaoka, T., Albani, C., Chance, B. and Iwane, H., A new method for the evaluation of muscle aerobic capacity in relation to physical activity measured by near-infrared spectroscopy. In Integration of Medical and Sports Sciences, Vol. 37, ed. Y. Sato, J. Poortmans, I. Hasimoto and Y. Oshida. Karger, Basel, 1992, pp. 421–429. 6. Cheatle, T. R., Potter, L. A., Cope, M., Delpy, D. T., Coleridge Smith, P. D. and Scurr, J. H., Near-infrared spectroscopy in peripheral vascular disease. British Journal of Surgery, 1991, 78, 405–408. 7. Bandyk, D. F., Kaebnick, H. W., Stewart, G. W. and Towne, J. B., Durability of the in situ saphenous vein arterial bypass: a comparison of primary and secondary patency. Journal of Vascular Surgery, 1987, 5, 256–268. 8. Nielsen, O. M., Perko, M., Crone, P. O., Gravgaard, E., Fasting, H., Jensen, B. M., Lass, P., Pedersen, I. K., Reimer, E. and Schroeder, T., Femoro-popliteal bypass-surgery in Denmark 1983–1987. Ugeskr laeger, 1990, 152, 1980–1982. 9. Shoenfeld, N. A., O’Donnell, T. F., Bush, H. L., Mackey, W. C. and Callow, A. D., The management of early in situ. saphenous vein bypass occlusion. Archives of Surgery, 1987, 122, 871–875. 10. Leather, R. P., Shah, D. M., Chang, B. B. and Kaufmann, J. L., Resurrection of the in situ vein bypass: 1000 cases later. Annals of Surgery, 1988, 208, 435–442. 11. Carney, W. I., Balko, A. and Barrett, M. S., In situ femoropopliteal and infrapopliteal bypass; two-year experience. Archives of Surgery, 1985, 120, 812–816. 12. Jo¨bsis, F. F., Noninvasive, infrared monitoring of cerebral and myocardial oxygen sufficiency and circulatory parameters. Science, 1977, 198, 1264–1267. 13. Wray, S., Cope, M., Delpy, D. T., Wyatt, J. S. and Reynolds, E. O. R., Characterization of the near-infrared absorption spectra of cytocrome aa3 and haemoglobin for the non-invasive monitoring of cerebral oxygenation. Biochimica et Biophysica Acta, 1988, 933, 184–192. 14. Selvick, E. M., Chance, B., Leigh, J., Nioka, S. and Maris, M., Quantitation of time and frequency-resolved optical spectra for the determination of tissue oxygenation. Analytical Biochemistry, 1991, 195, 330–351. 15. Chance, B., Continuous light, pulsed light, phase modulation and phased array. In Brain Lesions In The Newborn, Alfred Benzon
307
Near-Infrared spectroscopy during peripheral vascular surgery: J. P. Eiberg et al. Symposium, Vol. 37, ed. H. C. Lou, G. Greisen and J. F. Larsen. Munksgaard, Copenhagen, 1994, pp. 373–392. 16. Piantadosi, C. A., Hemstreet, T. M. and Jo¨bsis-Vandervliet, F. F., Near-infrared spectrophotometric monitoring of oxygen distribution to intact brain and skeletal muscle tissues. Critical Care Medicine, 1986, 14(8), 698–706. 17. Germon, T. J., Kane, N. M., Manara, A. R. and Nelson, R. J., Near-infrared spectroscopy in adults: effects of extracranial ischaemia and intracranial hypoxia on estimation of cerebral oxygenation. British Journal of Anaesthesiology, 1994, 73, 503–506.
308
18. Williams, I. M., Picton, A., Farrell, A., Mead, G. E., Mortimer, A. J. and McCollum, C. N., Light-reflective cerebral oxymetry and jugular bulb venous oxygen saturation during carotid endarterectomy. British Journal of Surgery, 1994, 81, 1291–1295. 19. Williams, I. M., Picton, A. J., Hardy, S. C., Mortimer, A. J. and McCollum, C. N., Cerebral hypoxia detected by near infrared spectroscopy. Anaesthesia, 1994, 49, 762–766. Paper accepted 22 January 1997
CARDIOVASCULAR SURGERY
JUNE 1997 VOL 5 NO 3
Cardiovascular Surgery, Vol. 5, No. 3, pp. 304–308, 1997 1997 The International Society for Cardiovascular Research Published by Elsevier Science Ltd. Printed in Great Britain 0967–2109/97 $17.00 + 0.00
Near-infrared spectroscopy during peripheral vascular surgery J. P. Eiberg*, T. V. Schroeder*, K. C. Vogt* and N. H. Secher† *Department of Vascular Surgery, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark and †Department of Anaesthesia, the Copenhagen Muscle Research Center, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark Near-infrared spectroscopy was performed perioperatively on the dorsum of the foot in 14 patients who underwent infrainguinal bypass surgery using a prosthesis or the greater saphenous vein. Dual-wavelength continuous light spectroscopy was used to assess changes in tissue saturation before, during and after the operation. Following the use of peripheral vascular grafts an immediate postoperative increase in tissue saturation of median 28 (range 210 to + 81) arbitrary units was noted (P , 0.01), Following distal clamping of the common femoral artery, maximal ischaemia corresponding to a median reduction in tissue saturation of 61 (range 6–94) units was reached after 26 (range 8–95) min (P , 0.01). The maximal tissue saturation following declamping was median 27 (range 216 to +100) units higher than the preoperative level (P , 0.01) and was reached after median 42 (range 8–125) min, The results indicate that near-infrared spectroscopy is appropriate for perioperative monitoring during vascular grafting. 1997 The International Society for Cardiovascular Surgery Keywords: near-infrared spectroscopy, vascular disease, vascular bypass surgery, perioperative oxymetry
Near-infrared spectroscopy was introduced for measuring cerebral oxygenation non-invasively [1, 2], but is also used when estimating changes in tissue oxygenation of human skeletal muscles, after application of a tourniquet and during exercise and recovery [3, 5]. Recently, the technique was performed to monitor oxygen consumption measurement in the calf, showing a reduced oxygen consumption among patients with peripheral vascular disease, compared with that in normal controls [6]. The aim of this study was to evaluate the prospects of near-infrared spectroscopy for continuous perioperative monitoring of vascular grafting, using peripheral tissue oxygenation as an indicator of the success of revascularization. Many failures of the femoropopliteal bypass technique occur within the first 30 days following operation, with a reported rate
Correspondence to: Dr J.P. Eiberg, c/o T.V. Schroeder, Department of Vascular Surgery, Rigshospitalet, University of Copenhagen, DK2100 Copenhagen Ø., Denmark
304
of early occlusion from 3% to 21% [7–11]. Some of these grafts could be salvaged before they thrombose, providing early objective evidence of the failing graft. Near-infrared spectroscopy has so far not been investigated on patients undergoing vascular surgery. During vascular bypass surgery, a correlation between tissue oxygenation and reproducible events such as distal clamping, ischaemia and reperfusion, during the insertion of peripheral bypass grafts, was evaluated.
Patients and methods Patients Nine male and five female patients (median age 71, (range 51–83) years) with atherosclerotic arterial obstructive disease and undergoing infrainguinal bypass surgery with implantation of a prosthesis (five patients) or greater saphenous vein (nine patients) were studied. The indication for surgery was acute ischaemia, with symptoms of less than 7 days in CARDIOVASCULAR SURGERY
JUNE 1997 VOL 5 NO 3
Near-Infrared spectroscopy during peripheral vascular surgery: J. P. Eiberg et al.
three patients. The remaining 11 patients presented with chronic ischaemia lasting for at least 1 month, in one patient resulting in gangrene, in seven patients ischaemic ulceration, and in three patients rest-pain or claudication. The median ankle: brachial pressure index was 0.33 (range 0–0.72) preoperatively and 0.77 (range 0.35–1.09) on the first postoperative day. The proximal anastomoses were to the common femoral artery, while the distal anastomoses were to the popliteal vessels in six cases and to infrapopliteal vessels in eight cases. Four patients were operated on under general anaesthesia and 10 patients under epidural analgesia. The study was approved by the Ethics Committee of Copenhagen (Journal number KF-01-342/94) and informed consent was obtained from all patients. Near-infrared spectroscopy In vivo spectroscopy makes use of the fact that photons in the near-infrared spectrum are transmitted in tissues and mainly absorbed by oxygenated and deoxygenated haemoglobin, cytochrome aa3 and myoglobin [12, 13]. Both oxygenated and deoxygenated haemoglobin absorb near-infrared light at 800 nm, while near-infrared light at 760 nm is absorbed mainly by deoxygenated haemoglobin [4]. Beer–Lambert’s law dictates that the fate of the photons is a combination of absorption and scattering, i.e. a photon that changes direction but maintains its speed [14, 15]. Beer–Lambert’s law is expressed as: I = I0(10 2 ecL) where I and I0 represent the intensities of the emergent and incident light, respectively, e the specific absorption coefficient for the compound in question, c its concentration and L the optical pathlength of the absorbing solution [6]. The optical pathlength of light travelling in a medium depends very much upon the number of scattering events and their mutual distance. The pathlength is therefore unknown in biological tissue. The absorbance of near-infrared light in the tissue, and thereby tissue oxygenation, can be estimated only in relative terms, and not in absolute terms. A dual-wavelength spectroscope developed for clinical measurements of cerebral oxygenation was used (Invos 3100, Somanetics, Troy, Wisconsin, USA). The equipment includes a sensor and a central unit containing a display and a computer. The sensor to be placed on the patient is not sterilized and consists of two near-infrared diodes used as light sources and two silicon light detectors, all housed in an adhesive plastic holder for firm attachment to the patient’s skin, The source–detector separation was 30 and 40 mm, respectively, for the two diode– detector pairs, giving an average mean depth of photon migration into the tissue of approximately 15 and 20 mm [15]. Spectral analysis showed that the light CARDIOVASCULAR SURGERY
JUNE 1997 VOL 5 NO 3
emitted was composed of near-infrared wavelengths of 733 and 803 nm. Before anaesthesia induction, the near-infrared sensor was placed on the dorsum of the foot over the extensor digitorum brevis muscle of the ischaemic leg. Saturation was monitored continuously with the near-infrared spectroscope from the start of the anaesthesia to 45 (range 11–193) min postoperatively. Saturation was recorded at intervals of 1 to 3 min, to create a trend curve with special attention to clamping and revascularization of the graft. Reproducibility In six age-matched atherosclerotic subjects, measurements during tourniquet-induced ischaemia and recovery were performed twice in order to evaluate the reproducibility. Each measurement included reattachment of the sensor. The span was observed from minimal saturation following 4 min of tourniquet inflation, to maximal saturation during postischaemic recovery. Data processing In order to compare the relative measures of tissue oxygenation, obtained in individual patients, measurements were calibrated. The saturation determined 20 min after occlusion of the common femoral artery was regarded as zero (0 arbitrary units). The maximal tissue saturation during the hyperaemia following declamping was defined as 100 arbitrary units. Statistical evaluation included Mann–Whitney’s test to compare groups and Wilcoxon’s signed rank sum test to compare matched pairs. Data are presented as median, with the range in brackets. A P-value of 0.05 was considered significant.
Results Following proximal clamping, a drop in tissue oxygenation of 61 (6–90) units was observed (P , 0.01) (Figure 1). The time taken to reach maximal ischaemia was 26 (8–95) min. Declamping after implantation of the bypass was followed by a rapid increase in tissue oxygenation overshooting the preoperative level by 27 ( 216 to +100) units (P , 0.01). This peak was reached after 42 (8–125) min following declamping, and was not significantly different from the tissue saturation at the end of the operation (P = 0.05). Thus, our major finding was an increase in tissue oxygenation following the bypass procedure of 28 ( 210 to +81) units by the end of the operation as compared with the preoperative level (P , 0.01). At 1 and 2 h postoperatively, the increase in tissue saturation was 17 ( 232 to +72) units (P , 0.03) and 1.8 ( 27 to +72) units (P , 0.02), respectively, as compared with the preoperative level. In one case, 305
Near-Infrared spectroscopy during peripheral vascular surgery: J. P. Eiberg et al.
Figure 1 Relative changes (median: solid line; range: vertical bar; inter-quartile range: horizontal bars) in tissue saturation during peripheral bypass surgery performed by near-infrared spectroscopy (n = 14). The difference between preceding tissue saturations has been tested and denoted * if significant. Tissue saturations have been denoted # if significantly different from those at the beginning of the operation (P , 0.05). Dotted lines indicate 95% confidence interval
extensive cyanosis of the forefoot caused a period of unstable measurements. There was no significant difference in tissue saturation between patients operated on under epidural analgesia as compared with general anaesthesia. Also, there was no correlation between the postoperative level of tissue oxygenation as compared with the technique of operation, nor with the duration of symptoms. During follow-up, after 26 (1–76) days, there were four graft failures, occurring between 24 hours and 9 days after operation. Among the four failing grafts it was not possible to find any significant immediate postoperative saturation pattern identifying the imminent occlusion. In the reproducibility study the saturation span ranged from 7 to 22 units, with a mean of 14 units (SD error of difference = 1.9).
Discussion With near-infrared spectroscopy the metabolic state of the tissue can be assessed non-invasively during ischaemia and revascularization [4, 12, 16]; consequently, the technique was applied to monitoring in vascular surgery. An immediate postoperative increase in tissue oxygenation was found following peripheral bypass grafting, overshooting the preoperative level by 28 units. Moreover with near-infrared spectroscopy it 306
was possible to demonstrate the correlation between tissue saturation and haemodynamic events such as proximal clamping and reperfusion. Tissue saturation did not indicate that the patients with acute lower-extremity ischaemia had a better result following grafting procedure than those patients with chronic lower extremity ischaemia. This was expected on account of the higher degree of collateral vascularization within the chronic ischaemic limbs. The reproducibility study, in which the light sensor was applied twice, showed the method to be reasonably reproducible. Total deoxygenation was not obtained during 4 min of ischaemia, but a more prolonged experimental ischaemia in order to simulate the intraoperative ischaemia without anaesthesia was unsuitable. Ideally, the near-infrared sensor should be mounted above a larger and well-defined muscle, thereby excluding scattering and near-infrared signals not originating from the haemoglobin in the muscle tissue. In the present study, performed during peripheral vascular surgery, the probe had to be mounted outside of the surgical field. Thus, the sensor was placed above the relatively small muscles on the dorsum of the foot, even though the tissue was irregular. Using a calibration routine, the saturation reflecting zero was determined over a prolonged perCARDIOVASCULAR SURGERY
JUNE 1997 VOL 5 NO 3
Near-Infrared spectroscopy during peripheral vascular surgery: J. P. Eiberg et al.
iod of total intraoperative ischaemia. Using any other routine of calibration a similar degree of ischaemia could not have been induced. By using a calibration routine, obtained over a wide range of saturations, the measurements become more sensitive to changes in the actual oxygenation and less dependent upon the properties of the underlying tissue. Using transcutaneous near-infrared spectroscopy to quantify, muscular oxy- and deoxyhaemoglobin, the assumption must be made that skin and subcutaneous fat contribute a negligible amount to the signal. Germon and colleagues [17] found that scalp ischaemia affects the Invos Cerebral Oxymeter when applied to the frontal area, suggesting that skin perfusion could contribute to the signal by 18%. Piantadosi and co-workers [16] found that the response from the skin during skin- and muscle-hypoxia contributed less than 5% using a six-wavelength spectrometer. In the present study, one included case with extensive cyanosis of the foot during distal as well as proximal clamping, caused unstable measurements following the colour shift of the skin. The spectroscope used in the present study detects the intracranial changes in oxygen saturation when applied over the forehead [17–19]. By the use of two wavelengths and different interoptode distances, it is assumed that the absorption caused by the superficial and extracranial compartments may largely be eliminated [17]. Using the apparatus on the foot, the same assumptions can be made, excluding the eventual contributions from the skin and subucutaneous fat from the signal. As no cases of graft failure occurred during the near-infrared monitoring, the authors would not suggest that the technique is the appropriate tool for postoperative monitoring, but would only suggest that the demonstrated immediate increase in tissue oxygenation following revascularization implies that a significant decrease in tissue oxygenation might occur as evidence of an early graft occlusion. An instrument which is easy to use and allows continuous measurements of tissue oxygenation could be a valuable supplement to the present methods of detecting graft failure, such as loss of pedal pulses and decreasing ankle:brachial pressure index. In order to assess this applicability, the near-infrared monitoring period must be expanded to include the first postoperative days. Two advantages of the instrument used are the simplicity of the sensor placement and the feasibility of recording the tissue saturation, without additional calculations. These two conditions must be fulfilled for a more widespread use of near-infrared spectroscopy as routine continuous patient monitoring.
CARDIOVASCULAR SURGERY
JUNE 1997 VOL 5 NO 3
Acknowledgements This work was supported by grants from the P. Carl Petersens Foundation. We wish to thank senior lecturer E. Veje, H.C. Ørsted Institute, University of Copenhagen, for useful technical assistance and E. Riis Andersen, Vingmed Denmark, for providing the Somanetic Invos 3100 Cerebral Oxymeter.
References 1. Wyatt, J. S., Cope, M., Delpy, D. T., Wray, S. and Reynolds, E. O. R., Quantification of cerebral oxygenation and haemodynamics in sick newborn infants by near-infrared spectrophotometry. Lancet, 1986, ii, 1063–1065. 2. Wyatt, J. S., Cope, M., Delpy, D. T., Richardson, C. E., Edwards, A. D., Wray, S. and Reynolds, E. O. R., Quantitation of cerebral blood volume in human infants by near-infrared spectroscopy. Journal of Applied Physiology, 1990, 68(3), 1086–1091. 3. Hampson, N. B. and Piantdosi, C. A., Near infrared monitoring of human skeletal muscle oxygenation during forearm ischaemia. Journal of Applied Physiology, 1988, 64(6), 2449–2457. 4. Mancini, D. M., Ferraro, N., Nazzaro, D., Chance, B. and Wilson, J. R., Respiratory muscle deoxygenation during exercise in patients with heart failure demonstrated with near-infrared spectroscopy. Journal of the American College of Cardiology, 1991, 18, 492–498. 5. Hamaoka, T., Albani, C., Chance, B. and Iwane, H., A new method for the evaluation of muscle aerobic capacity in relation to physical activity measured by near-infrared spectroscopy. In Integration of Medical and Sports Sciences, Vol. 37, ed. Y. Sato, J. Poortmans, I. Hasimoto and Y. Oshida. Karger, Basel, 1992, pp. 421–429. 6. Cheatle, T. R., Potter, L. A., Cope, M., Delpy, D. T., Coleridge Smith, P. D. and Scurr, J. H., Near-infrared spectroscopy in peripheral vascular disease. British Journal of Surgery, 1991, 78, 405–408. 7. Bandyk, D. F., Kaebnick, H. W., Stewart, G. W. and Towne, J. B., Durability of the in situ saphenous vein arterial bypass: a comparison of primary and secondary patency. Journal of Vascular Surgery, 1987, 5, 256–268. 8. Nielsen, O. M., Perko, M., Crone, P. O., Gravgaard, E., Fasting, H., Jensen, B. M., Lass, P., Pedersen, I. K., Reimer, E. and Schroeder, T., Femoro-popliteal bypass-surgery in Denmark 1983–1987. Ugeskr laeger, 1990, 152, 1980–1982. 9. Shoenfeld, N. A., O’Donnell, T. F., Bush, H. L., Mackey, W. C. and Callow, A. D., The management of early in situ. saphenous vein bypass occlusion. Archives of Surgery, 1987, 122, 871–875. 10. Leather, R. P., Shah, D. M., Chang, B. B. and Kaufmann, J. L., Resurrection of the in situ vein bypass: 1000 cases later. Annals of Surgery, 1988, 208, 435–442. 11. Carney, W. I., Balko, A. and Barrett, M. S., In situ femoropopliteal and infrapopliteal bypass; two-year experience. Archives of Surgery, 1985, 120, 812–816. 12. Jo¨bsis, F. F., Noninvasive, infrared monitoring of cerebral and myocardial oxygen sufficiency and circulatory parameters. Science, 1977, 198, 1264–1267. 13. Wray, S., Cope, M., Delpy, D. T., Wyatt, J. S. and Reynolds, E. O. R., Characterization of the near-infrared absorption spectra of cytocrome aa3 and haemoglobin for the non-invasive monitoring of cerebral oxygenation. Biochimica et Biophysica Acta, 1988, 933, 184–192. 14. Selvick, E. M., Chance, B., Leigh, J., Nioka, S. and Maris, M., Quantitation of time and frequency-resolved optical spectra for the determination of tissue oxygenation. Analytical Biochemistry, 1991, 195, 330–351. 15. Chance, B., Continuous light, pulsed light, phase modulation and phased array. In Brain Lesions In The Newborn, Alfred Benzon
307
Near-Infrared spectroscopy during peripheral vascular surgery: J. P. Eiberg et al. Symposium, Vol. 37, ed. H. C. Lou, G. Greisen and J. F. Larsen. Munksgaard, Copenhagen, 1994, pp. 373–392. 16. Piantadosi, C. A., Hemstreet, T. M. and Jo¨bsis-Vandervliet, F. F., Near-infrared spectrophotometric monitoring of oxygen distribution to intact brain and skeletal muscle tissues. Critical Care Medicine, 1986, 14(8), 698–706. 17. Germon, T. J., Kane, N. M., Manara, A. R. and Nelson, R. J., Near-infrared spectroscopy in adults: effects of extracranial ischaemia and intracranial hypoxia on estimation of cerebral oxygenation. British Journal of Anaesthesiology, 1994, 73, 503–506.
308
18. Williams, I. M., Picton, A., Farrell, A., Mead, G. E., Mortimer, A. J. and McCollum, C. N., Light-reflective cerebral oxymetry and jugular bulb venous oxygen saturation during carotid endarterectomy. British Journal of Surgery, 1994, 81, 1291–1295. 19. Williams, I. M., Picton, A. J., Hardy, S. C., Mortimer, A. J. and McCollum, C. N., Cerebral hypoxia detected by near infrared spectroscopy. Anaesthesia, 1994, 49, 762–766. Paper accepted 22 January 1997
CARDIOVASCULAR SURGERY
JUNE 1997 VOL 5 NO 3