- Email: [email protected]
Profound systemic hypothermia and release of neurotransmitter amino acids
406
Letters to the Editor
whether the fiN case of MPLCs wa~ described by Beyreuther in 1924, as mentioned by Okada and colleagues, I or whether the first case was descnbed in 1928, as pOinted out by Chaudhun. 5 Bedrettin Yildt::.eli. MD Mustafa Yiiksel, MD Sina Ercclll, MD Hasan F Bat/rei, MD Marmara Unil'ersity Faculty of Medicine Department (){ Thoracic Surgery Istanbul. Turkev REFERENCES I. Okada M, Tsubota N, Ym,himura M, Miyamoto Y. Operative
2. 3. 4.
5.
approach for multiple primary lung carcinoma,. J Thorac Cardiovase Surg 1998: 115.836-40. Antakli T, Schaefer RE Rutherford JE, Read RC. Second prImary lung cancers. Ann Thorac Surg 1995:59:863-7. MartinI N, Melamed MR. Multiple prImary lung cancers. J Thorac CardlOvasc Surg I975,7{l'606-12. Richardson GE, Tucl..er MA, Venzon DJ, et al. SmokIng cessation after successful treatment of small-cell lung cancer is assOCiated With fewer smokIng-related second primary cancers. Ann Intern Med 1993:119:383-90. Chaudhuri MR Independent bilateral primary bronchial carCInomas. Thorax 1971 :26.476-80. 12/8/94818
Reply to the Editor: It is our pleasure to address the points raised by YI1dlzeli and associates in their letter regardmg our recently published article, 1 We had already asses~ed the characteristics of small non-small cell lung cancers with DNA ploidy. This study consisted of 39% aneuploidy and 61 % diplOidy tumors and suggested that DNA ploidy was a prognostic factor and that aneuploidy possessed occult malignancy, 2 However, we do not think that the DNA ploidy IS the definitive answer to the dilemma whether a tumor IS a second primary or a recurrence. When the 2 tumors had completely different patterns of DNA ploidy, they were defined as second primaries. However, as shown in many previous reports,3.5 all specimens had normal (diploid) peaks, which were made by tumor cells and normal epithelial or stromal cells (or both) within tumors. Even with cytofluorometric evaluation of lung cancer, which can theoretically select only tumor cells, most tumors contained normal peaks." Therefore we have speculated on the sensitivity and specificity of the ploidy results regarding the differentiation between second primaries and metastases. We have followed the policy that lung-saving procedures such as bronchoplasty for a primary leSIOn must always be kept in mind. Of 671 patients who underwent operations for primary lung cancer over a period of 6 years. we con~idered 39 patients to have pT2 NI MO disease. None was subjected to pneumonectomy, and 14 (36'7c) underwent sleeve lobectomy instead. We succeeded in preserving lung function with a reasonable prognosis and a better postoperative quality of life, which was proven by spirometry and treadmill exercise test 7 .8 Needless to say, pneumonectomy decreases the chance
The Jou mal of ThoraCIC and Cardiovascular Surgery February 1999
of a second pnmary sigmfICantly compared wilh lobectomy, because removal of more lung tissue with the first tumor ',>,ould leave the patient With les~ lung tissue in which a second pnmary could develop later. Preservation of lung function, however, is much more important than the risk of second primary lesions. A smoking index (cigarettes/day x years) of more than 600 at the time of diagnosis of the second tumor was recorded in 43 patient~ (75%), suggesting a remarkable incidence of smoking among patients with multiple primary tumors. This probably accounts for the high proportion of squamous cell carcinoma~ in our series. Chaudhuri,9 like US,l mentioned that Beyreuther descnbed the first case of multiple primary lung cancer in 1924. Morihito Okada, MD Nortaki Tsubota, MD Department o.{ Thoracic Surge I)' Hyogo Medical Center for Adults Kitaohji-cllO 13-70, Akashi City Hyof{o, 673 Japan REFERENCES I. Okada M. Tsubota N, Yoshimura M, Miyamoto Y. Operative approach for multIple primary lung carCInomas. J Thorac Cardlllvasc Surg 1998,115:836-40. 2 Sakamoto T. Tsubota N. Yoshimura M, Kubota M, MurotanI A. Flow cytometric DNA analysis in smallleslom, of non-small cell lung cancer. Halgan 1993:33.517-23 3. Zimmerman PV, Hawson GAT. BInt MH. Parsons PG. PlOidy as a prognostic determInant In surgically treated lung cancer. Lancet 1987:330:530-3. 4. Tmndelh-Danesl D, Teodon L Mauro E et al. PrognostIC signifIcance of flow cytometry III lung cancer. a 5-year study. Cancer 1987:60.844-51. 5. Salvati F, Teodori L. Gagliardi L, Signora M, AqUlIIni M. Stormello G. DNA flow cytometric ,tudies ot 66 human lung tumors analyzed before treatment. Chest 1989.96: 1092-8. 6. Asamura H, NaI-.aJlma T, Mukai K, Shlmo,ato Y. Nuclear DNA content by cytofluorometry of ;,tate I adenocarcInoma of the lung In relation to postoperative recurrence. Chest 1989:96 312-8. 7. Tsubota N. Yo,hlmura M, Murotam A, Miyamoto Y. Matoha Y. One hundred and one ca,e, of bronchopla;,ty for pnmary lung cancer Surg Today 1994.24:978-81. 8 Tsuhota N, Yoshimura M, Murotam A, Miyamoto Y, Matoba Y. Atypical bronchoplasty to lung cancer and benign bronchial disease. Tohoku J Exp Med 1994,172:209-20. 9 Chaudhun MR. Independent bIlateral pnmary bronchial carCInOmas. Thorax 1971:26:476-80 12/8/94817
Profound systemic hypothermia and release of neurotransmitter amino acids To the Editor: We read with great interest the article titled "Profound Systemic Hypothermia Inhibits the Release of Neurotransmitter Ammo Acids in Spinal Cord Ischemia" by Rokkas and associates (1 Thorac Cardiovasc Surg 1995:110:27-35). Although considerable information IS provided, we believe It needs to be placed in the proper perspective to lead to a cor-
The Journal of ThoracIc and
Letters to the Editor 407
Cardiovascular Surgery Volume 117, Number 2
rect interpretation. It is unfortunate that only 2 sham animals were operated on for each experimental group (that would preclude using that information for any statistical analysi'J, and the data are not reported but merely mentioned in the text: "The control sham animals demonstrated stable baseline values of amino acid concentrations for at least 150 minutes. The hypothermia sham animals (hypothermia without aortic crossclamping) exhibited time-course changes in dialysate amino acid levels that were indistinguishable from the changes observed in group 2 animals (hypothermia and aortic crossclamping)." Although the general time course might have had the same trend, the values must have been different. If they were not different, that knowledge would have been helpful and informative for the readership. Had the data been reported, we believe differences would have been noticed between the clamped (group 2) and the sham (nonclamped) hypothennic animals. It is also unfortunate that the main thrust was to compare group 1 (nonnothennic ischemia) and group 2 (hypothermic Ischemia), with less weight given to comparison against their own baselines. The baseline measurements for group 2 were taken only before cooling, which we think is inadequate unless enough sham group data are provided to assess the effects of hypothermia per se. Had enough consideration to comparison again,t their baselines been given, some obvious differences would have been noticed that might have helped in the proper interpretation of the data shown in Table I. Perhaps the most important informatIOn in this article concerned the significant differences in adenosine concentrations between groups I and 2 during the ischemic period. However, because attention was focused on the excitatory amino acid (glutamate), the adenosine data were not properly discussed. Although Table I appears to show that the differences from the baseline values were not Significant, our analysis proved otherwise. Adenosine increased more than 4-fold during normothennlc ischemia, which is anticipated and in agreement with reported facts in normothermic ischemia. but it was significantly lower in the hypothermic group during ischemia, as well as during each of the reperfusion times. With such an increase in adenosine in the normothermic group, an increase in taurine is expected, because adenosine releases taurine. I-3 Because only the baseline data for hypothermic conditions are supplied (ie, just before clamping in group 2) and because the sham group was not large enough to supply sufficient data, adequate comparison is not possible. Hypothermia will decrease the rate of all processes, taurine relem,e Illcluded, and it would be improper to make comparisons agaillst the baseline values measured under normothermiC conditions. However, disregarding the effect of hypothermia per se on the release of the various amino acids, adenosllle levels in group 2 are significantly lower than the normothermic baseline, low enough to explalll at least partially why the taurine concentrations were also lower than the normothermic baseline value. The decreased adenosine and glutamate levels in group 2 during ischemia might indeed represent the protective effect of hypothermia, which prevented the degradation of adenosine triphosphate to adenosine and the ischemiainduced depolarization with its concomitant release of gluta-
mate. It can be speculated that glutamate, y-aminobutyric acid, and taurine metabolic production during hypothennia were decreased. Apparently taurine is a co-product formed during adenosine triphosphate degradation, which involves interaction with the homocysteine-methionine cycle producing taurine as an end co-product from decarboxylatIOn of cysteic acid. 4 .5 On the basis of the mentioned conSiderations, and although the authors ,tate "the role of taurine in the spinal cord IS not clear but It does not appear that taurine plays any significant role in modulating the ischemic Illjury," the findings of the study do not ,ubstantiate that assertion. Furthermore, the study was not designed to prove or disprove the protective effects of taurine. In their study taurine was never administered to evaluate its effects. The release of taurine is part of the natural defensive mechanism set in motion when excess adenosine is formed, such as in ischemia or trauma to any area of the central nervous system,6.7 spinal cord included. The fact that taurine levels were found to be low does not indicate that taunne does not have a protective role. What is actually indicating IS that the concomitantly used protective strategy (in this case profound levels of hypothermia) was effective in preventing the activation of mechanisms that release taurine. We even hypothesize that the levels of taurine could actually be used as an indicator of whether protection was achieved when using protective strategies not involving taurine administration. In fact, our recent experimental studies in rabbits have shown that exogenous taurine enhances the protective effects of hypothermia and confers significant protection in animals with spinal cord normothermic ischemia as well. If the temporal course of y-aminobutyric acid and glutamate in the post-rewarming phase of a large enough sham group of hypothermic animals is similar to that of group 2, the effect of hypothermia is not limited to the release of glutamate during the ischemic period only. Since it is known that mild or moderate degrees of hypothermia are effective in decreasing the release of glutamate during ischemia, it would be of interest to see whether such late effects are also observed with mJld or moderate degrees of hypothermia rather than profound hypothermia. Tadaomi-A. Miyamoto, MD Research Department Kokura Memorial Hospital Kitakyushu, Japan KoJw-J. Miyamoto. MD, PhD Assistant Professor II Department of Physiology University of Ryukius School of Medicine Okinawa, Japan
REFERENCES I. Madelian V. Silliman S, Sham W. Adenosme stimulates cAMPmediated taurine release from LRM55 glial cells J Neurosci Res 1988;20: 176-81. 2. Miyamoto TA, MIyamoto KJ. Effects of adenosme on taurine release m the central nervoU> system. J Jpn PhyslOl Soc 1996; 46(SuppIJ: 179. 3. Lupica CR, Cass WA. Zahniser NR. Dunwiddie TV. Effects of
408
4.
5.
6.
7.
The Journal of ThoraCIC and Cardiovascular Surgery February 1999
Letters to the Editor
the selective adeno'Ine A2 receptor agonist CGS 216XO on In vitro electrophy,iology, cAMP formatIOn and dopamine release in rat hlppocampu, and stnatum, 1 Pharmacol Exp Ther 1990; 252. 1134-41. Schrader 1. Formation and metabolism of adenOSIne and adenine nucleotIdes in cardiac tissue. In, Phillis 1W, edItor AdenOSIne and adenIne nucleotides m, regulators of cellular functIOn. Boca Raton [FL): CRC Press; 1991. p, 56-65. Fruton 1S, Simmonds S. AmIno aCIds as structural Units of proteins, In: Fruton 1S, Simmonds S, editors. General biochemIstry. London: John Wiley; Chapman & Hall; 1953. p, 58, Benveniste H, Drejer 1, Schousboe A, DIemer NH. ElevatIon of extracellular concentrations of glutamate and aspartate In rat hIppocampus during transIent cerebral Ischemia mOnitored by Intracerebral microdlalysis, 1 Neurochem 1984;43: 1369-74. Hillered L, Hallstrom A, Segersvad S, Persson L, Ungerstedt U. Dynamics of extracellular metabolItes in the stnatum after middle cerebral artery occlusion In the rat monitored by intracerebral mlcrodialysis, 1 Cereb Blood Flow Metab 1989,9'60716. 12/8/95024
Reply to the Editor: It is somewhat surprising and quite unusual to receive a letter about an article that we published in the Journal more than 3 years ago. The fast-paced developments in neuroscience make It somewhat difficult to put the findings of an older study in perspective, In our article titled "Profound Systemic Hypothermia Inhibits the Release of Neurotransmitter Amino Acids in Spinal Cord Ischemia" (J Thorac Cardiovasc Surg 1995;1 10:27-35), we simply described the findings of a study designed to clarify the role of excitotoxicity in spinal cord ischemia and a possible mechanism for the protective effect of hypothermia, For the first time in the spinal cord, amino acid levels were measured under conditions simulating the clmical situation, and several of the findings broke new ground in this field. More recent studies put those findings in a proper perspective, describing mechanbms that qUIte simply were unknown at the time of publication of our article. Our paper is criticized for not discussing references that appeared in the literature after the paper was published. The authors of the letter also claim that they re-analyzed statistically our data and reached different results, mdirectly implying that our statistical methods were flawed. We wonder. however, how they were able to perform statistical analysis Without havmg the raw data. In the data table, we provided mean values and standard deviations for the various expenmental groups. The statistical analysis was performed by taking into conSideration the results from individual experiments. Any analysis that compares only mean values of different groups is simply not accurate. We also wish to point out that sham animals served for validation of methodology only, without per se constituting an experimental group. It is a commonly accepted practice to use baseline values as controL so long as they do not differ statistically among the experimental groups. The field of ischemia and neuronal cell death has expanded the horizons of neuroscience tremendously in recent years.
We believe our article provided some msights into the pathophysiology of spinal cord ischemia. Chris K. Rokkas. MD Nicholas T. Kouchoukos, MD Cardiac, Thoracic & Vascular Surgery Inc. 3009 North Ballas Rd Suite 266 C St Louis, MO 63131 12/8/95023
Left subclavian artery as a site of proximal aortic perfusion for hypothermic repair of thoracic and thoracoabdominal aneurysms To the Editor: Retrograde aortic perfusion through the femoral artery with deep hypothermic circulatory arrest is a valuable adjunct for thoracic and thoracoabdominal aneurysms. I However, retrograde perfusion of the brain through an atheromatous or dissected aorta carries the risk of cerebral embolism or malperfusion. To avoid these fatal complications, Westaby and Katsumata2 proposed a proximal aortic perfusion through the ascending aorta or aortic arch via an extended left thoracotomy. However, the possibility exists that cerebral emboli may be produced even by external manipulation of an atheromatous aorta or aneurysm. Hence we have recently adopted a proximal aortic perfusion technique usmg a prosthetic graft attached to the left subclavian artery via a left posterolateral thoracotomy for hypothermic thoracic and thoraco-abdominal aneurysm repair (Fig I). The left subclavian artery with easy access from a thoracotomy has been used for a systemic-pulmonary arterial shunt using a prosthetic graft in neonates or small infants. 3 Moreover, the axillary artery is also widely known as an alternative site for proximal arterial
Fig 1. Left subclavian artery (LSA) cannulation technique through a posterolateral thoracotomy. After proximal and distal control of the LSA is obtained, an 8-mm sealed woven Dacron graft is attached to the LSA. followed by cannulation with a polyvinyl arch cannula. Venous drainage is through a straight venous cannula placed in the main pulmonary artery (PA).
Letters to the Editor
whether the fiN case of MPLCs wa~ described by Beyreuther in 1924, as mentioned by Okada and colleagues, I or whether the first case was descnbed in 1928, as pOinted out by Chaudhun. 5 Bedrettin Yildt::.eli. MD Mustafa Yiiksel, MD Sina Ercclll, MD Hasan F Bat/rei, MD Marmara Unil'ersity Faculty of Medicine Department (){ Thoracic Surgery Istanbul. Turkev REFERENCES I. Okada M, Tsubota N, Ym,himura M, Miyamoto Y. Operative
2. 3. 4.
5.
approach for multiple primary lung carcinoma,. J Thorac Cardiovase Surg 1998: 115.836-40. Antakli T, Schaefer RE Rutherford JE, Read RC. Second prImary lung cancers. Ann Thorac Surg 1995:59:863-7. MartinI N, Melamed MR. Multiple prImary lung cancers. J Thorac CardlOvasc Surg I975,7{l'606-12. Richardson GE, Tucl..er MA, Venzon DJ, et al. SmokIng cessation after successful treatment of small-cell lung cancer is assOCiated With fewer smokIng-related second primary cancers. Ann Intern Med 1993:119:383-90. Chaudhuri MR Independent bilateral primary bronchial carCInomas. Thorax 1971 :26.476-80. 12/8/94818
Reply to the Editor: It is our pleasure to address the points raised by YI1dlzeli and associates in their letter regardmg our recently published article, 1 We had already asses~ed the characteristics of small non-small cell lung cancers with DNA ploidy. This study consisted of 39% aneuploidy and 61 % diplOidy tumors and suggested that DNA ploidy was a prognostic factor and that aneuploidy possessed occult malignancy, 2 However, we do not think that the DNA ploidy IS the definitive answer to the dilemma whether a tumor IS a second primary or a recurrence. When the 2 tumors had completely different patterns of DNA ploidy, they were defined as second primaries. However, as shown in many previous reports,3.5 all specimens had normal (diploid) peaks, which were made by tumor cells and normal epithelial or stromal cells (or both) within tumors. Even with cytofluorometric evaluation of lung cancer, which can theoretically select only tumor cells, most tumors contained normal peaks." Therefore we have speculated on the sensitivity and specificity of the ploidy results regarding the differentiation between second primaries and metastases. We have followed the policy that lung-saving procedures such as bronchoplasty for a primary leSIOn must always be kept in mind. Of 671 patients who underwent operations for primary lung cancer over a period of 6 years. we con~idered 39 patients to have pT2 NI MO disease. None was subjected to pneumonectomy, and 14 (36'7c) underwent sleeve lobectomy instead. We succeeded in preserving lung function with a reasonable prognosis and a better postoperative quality of life, which was proven by spirometry and treadmill exercise test 7 .8 Needless to say, pneumonectomy decreases the chance
The Jou mal of ThoraCIC and Cardiovascular Surgery February 1999
of a second pnmary sigmfICantly compared wilh lobectomy, because removal of more lung tissue with the first tumor ',>,ould leave the patient With les~ lung tissue in which a second pnmary could develop later. Preservation of lung function, however, is much more important than the risk of second primary lesions. A smoking index (cigarettes/day x years) of more than 600 at the time of diagnosis of the second tumor was recorded in 43 patient~ (75%), suggesting a remarkable incidence of smoking among patients with multiple primary tumors. This probably accounts for the high proportion of squamous cell carcinoma~ in our series. Chaudhuri,9 like US,l mentioned that Beyreuther descnbed the first case of multiple primary lung cancer in 1924. Morihito Okada, MD Nortaki Tsubota, MD Department o.{ Thoracic Surge I)' Hyogo Medical Center for Adults Kitaohji-cllO 13-70, Akashi City Hyof{o, 673 Japan REFERENCES I. Okada M. Tsubota N, Yoshimura M, Miyamoto Y. Operative approach for multIple primary lung carCInomas. J Thorac Cardlllvasc Surg 1998,115:836-40. 2 Sakamoto T. Tsubota N. Yoshimura M, Kubota M, MurotanI A. Flow cytometric DNA analysis in smallleslom, of non-small cell lung cancer. Halgan 1993:33.517-23 3. Zimmerman PV, Hawson GAT. BInt MH. Parsons PG. PlOidy as a prognostic determInant In surgically treated lung cancer. Lancet 1987:330:530-3. 4. Tmndelh-Danesl D, Teodon L Mauro E et al. PrognostIC signifIcance of flow cytometry III lung cancer. a 5-year study. Cancer 1987:60.844-51. 5. Salvati F, Teodori L. Gagliardi L, Signora M, AqUlIIni M. Stormello G. DNA flow cytometric ,tudies ot 66 human lung tumors analyzed before treatment. Chest 1989.96: 1092-8. 6. Asamura H, NaI-.aJlma T, Mukai K, Shlmo,ato Y. Nuclear DNA content by cytofluorometry of ;,tate I adenocarcInoma of the lung In relation to postoperative recurrence. Chest 1989:96 312-8. 7. Tsubota N. Yo,hlmura M, Murotam A, Miyamoto Y. Matoha Y. One hundred and one ca,e, of bronchopla;,ty for pnmary lung cancer Surg Today 1994.24:978-81. 8 Tsuhota N, Yoshimura M, Murotam A, Miyamoto Y, Matoba Y. Atypical bronchoplasty to lung cancer and benign bronchial disease. Tohoku J Exp Med 1994,172:209-20. 9 Chaudhun MR. Independent bIlateral pnmary bronchial carCInOmas. Thorax 1971:26:476-80 12/8/94817
Profound systemic hypothermia and release of neurotransmitter amino acids To the Editor: We read with great interest the article titled "Profound Systemic Hypothermia Inhibits the Release of Neurotransmitter Ammo Acids in Spinal Cord Ischemia" by Rokkas and associates (1 Thorac Cardiovasc Surg 1995:110:27-35). Although considerable information IS provided, we believe It needs to be placed in the proper perspective to lead to a cor-
The Journal of ThoracIc and
Letters to the Editor 407
Cardiovascular Surgery Volume 117, Number 2
rect interpretation. It is unfortunate that only 2 sham animals were operated on for each experimental group (that would preclude using that information for any statistical analysi'J, and the data are not reported but merely mentioned in the text: "The control sham animals demonstrated stable baseline values of amino acid concentrations for at least 150 minutes. The hypothermia sham animals (hypothermia without aortic crossclamping) exhibited time-course changes in dialysate amino acid levels that were indistinguishable from the changes observed in group 2 animals (hypothermia and aortic crossclamping)." Although the general time course might have had the same trend, the values must have been different. If they were not different, that knowledge would have been helpful and informative for the readership. Had the data been reported, we believe differences would have been noticed between the clamped (group 2) and the sham (nonclamped) hypothennic animals. It is also unfortunate that the main thrust was to compare group 1 (nonnothennic ischemia) and group 2 (hypothermic Ischemia), with less weight given to comparison against their own baselines. The baseline measurements for group 2 were taken only before cooling, which we think is inadequate unless enough sham group data are provided to assess the effects of hypothermia per se. Had enough consideration to comparison again,t their baselines been given, some obvious differences would have been noticed that might have helped in the proper interpretation of the data shown in Table I. Perhaps the most important informatIOn in this article concerned the significant differences in adenosine concentrations between groups I and 2 during the ischemic period. However, because attention was focused on the excitatory amino acid (glutamate), the adenosine data were not properly discussed. Although Table I appears to show that the differences from the baseline values were not Significant, our analysis proved otherwise. Adenosine increased more than 4-fold during normothennlc ischemia, which is anticipated and in agreement with reported facts in normothermic ischemia. but it was significantly lower in the hypothermic group during ischemia, as well as during each of the reperfusion times. With such an increase in adenosine in the normothermic group, an increase in taurine is expected, because adenosine releases taurine. I-3 Because only the baseline data for hypothermic conditions are supplied (ie, just before clamping in group 2) and because the sham group was not large enough to supply sufficient data, adequate comparison is not possible. Hypothermia will decrease the rate of all processes, taurine relem,e Illcluded, and it would be improper to make comparisons agaillst the baseline values measured under normothermiC conditions. However, disregarding the effect of hypothermia per se on the release of the various amino acids, adenosllle levels in group 2 are significantly lower than the normothermic baseline, low enough to explalll at least partially why the taurine concentrations were also lower than the normothermic baseline value. The decreased adenosine and glutamate levels in group 2 during ischemia might indeed represent the protective effect of hypothermia, which prevented the degradation of adenosine triphosphate to adenosine and the ischemiainduced depolarization with its concomitant release of gluta-
mate. It can be speculated that glutamate, y-aminobutyric acid, and taurine metabolic production during hypothennia were decreased. Apparently taurine is a co-product formed during adenosine triphosphate degradation, which involves interaction with the homocysteine-methionine cycle producing taurine as an end co-product from decarboxylatIOn of cysteic acid. 4 .5 On the basis of the mentioned conSiderations, and although the authors ,tate "the role of taurine in the spinal cord IS not clear but It does not appear that taurine plays any significant role in modulating the ischemic Illjury," the findings of the study do not ,ubstantiate that assertion. Furthermore, the study was not designed to prove or disprove the protective effects of taurine. In their study taurine was never administered to evaluate its effects. The release of taurine is part of the natural defensive mechanism set in motion when excess adenosine is formed, such as in ischemia or trauma to any area of the central nervous system,6.7 spinal cord included. The fact that taurine levels were found to be low does not indicate that taunne does not have a protective role. What is actually indicating IS that the concomitantly used protective strategy (in this case profound levels of hypothermia) was effective in preventing the activation of mechanisms that release taurine. We even hypothesize that the levels of taurine could actually be used as an indicator of whether protection was achieved when using protective strategies not involving taurine administration. In fact, our recent experimental studies in rabbits have shown that exogenous taurine enhances the protective effects of hypothermia and confers significant protection in animals with spinal cord normothermic ischemia as well. If the temporal course of y-aminobutyric acid and glutamate in the post-rewarming phase of a large enough sham group of hypothermic animals is similar to that of group 2, the effect of hypothermia is not limited to the release of glutamate during the ischemic period only. Since it is known that mild or moderate degrees of hypothermia are effective in decreasing the release of glutamate during ischemia, it would be of interest to see whether such late effects are also observed with mJld or moderate degrees of hypothermia rather than profound hypothermia. Tadaomi-A. Miyamoto, MD Research Department Kokura Memorial Hospital Kitakyushu, Japan KoJw-J. Miyamoto. MD, PhD Assistant Professor II Department of Physiology University of Ryukius School of Medicine Okinawa, Japan
REFERENCES I. Madelian V. Silliman S, Sham W. Adenosme stimulates cAMPmediated taurine release from LRM55 glial cells J Neurosci Res 1988;20: 176-81. 2. Miyamoto TA, MIyamoto KJ. Effects of adenosme on taurine release m the central nervoU> system. J Jpn PhyslOl Soc 1996; 46(SuppIJ: 179. 3. Lupica CR, Cass WA. Zahniser NR. Dunwiddie TV. Effects of
408
4.
5.
6.
7.
The Journal of ThoraCIC and Cardiovascular Surgery February 1999
Letters to the Editor
the selective adeno'Ine A2 receptor agonist CGS 216XO on In vitro electrophy,iology, cAMP formatIOn and dopamine release in rat hlppocampu, and stnatum, 1 Pharmacol Exp Ther 1990; 252. 1134-41. Schrader 1. Formation and metabolism of adenOSIne and adenine nucleotIdes in cardiac tissue. In, Phillis 1W, edItor AdenOSIne and adenIne nucleotides m, regulators of cellular functIOn. Boca Raton [FL): CRC Press; 1991. p, 56-65. Fruton 1S, Simmonds S. AmIno aCIds as structural Units of proteins, In: Fruton 1S, Simmonds S, editors. General biochemIstry. London: John Wiley; Chapman & Hall; 1953. p, 58, Benveniste H, Drejer 1, Schousboe A, DIemer NH. ElevatIon of extracellular concentrations of glutamate and aspartate In rat hIppocampus during transIent cerebral Ischemia mOnitored by Intracerebral microdlalysis, 1 Neurochem 1984;43: 1369-74. Hillered L, Hallstrom A, Segersvad S, Persson L, Ungerstedt U. Dynamics of extracellular metabolItes in the stnatum after middle cerebral artery occlusion In the rat monitored by intracerebral mlcrodialysis, 1 Cereb Blood Flow Metab 1989,9'60716. 12/8/95024
Reply to the Editor: It is somewhat surprising and quite unusual to receive a letter about an article that we published in the Journal more than 3 years ago. The fast-paced developments in neuroscience make It somewhat difficult to put the findings of an older study in perspective, In our article titled "Profound Systemic Hypothermia Inhibits the Release of Neurotransmitter Amino Acids in Spinal Cord Ischemia" (J Thorac Cardiovasc Surg 1995;1 10:27-35), we simply described the findings of a study designed to clarify the role of excitotoxicity in spinal cord ischemia and a possible mechanism for the protective effect of hypothermia, For the first time in the spinal cord, amino acid levels were measured under conditions simulating the clmical situation, and several of the findings broke new ground in this field. More recent studies put those findings in a proper perspective, describing mechanbms that qUIte simply were unknown at the time of publication of our article. Our paper is criticized for not discussing references that appeared in the literature after the paper was published. The authors of the letter also claim that they re-analyzed statistically our data and reached different results, mdirectly implying that our statistical methods were flawed. We wonder. however, how they were able to perform statistical analysis Without havmg the raw data. In the data table, we provided mean values and standard deviations for the various expenmental groups. The statistical analysis was performed by taking into conSideration the results from individual experiments. Any analysis that compares only mean values of different groups is simply not accurate. We also wish to point out that sham animals served for validation of methodology only, without per se constituting an experimental group. It is a commonly accepted practice to use baseline values as controL so long as they do not differ statistically among the experimental groups. The field of ischemia and neuronal cell death has expanded the horizons of neuroscience tremendously in recent years.
We believe our article provided some msights into the pathophysiology of spinal cord ischemia. Chris K. Rokkas. MD Nicholas T. Kouchoukos, MD Cardiac, Thoracic & Vascular Surgery Inc. 3009 North Ballas Rd Suite 266 C St Louis, MO 63131 12/8/95023
Left subclavian artery as a site of proximal aortic perfusion for hypothermic repair of thoracic and thoracoabdominal aneurysms To the Editor: Retrograde aortic perfusion through the femoral artery with deep hypothermic circulatory arrest is a valuable adjunct for thoracic and thoracoabdominal aneurysms. I However, retrograde perfusion of the brain through an atheromatous or dissected aorta carries the risk of cerebral embolism or malperfusion. To avoid these fatal complications, Westaby and Katsumata2 proposed a proximal aortic perfusion through the ascending aorta or aortic arch via an extended left thoracotomy. However, the possibility exists that cerebral emboli may be produced even by external manipulation of an atheromatous aorta or aneurysm. Hence we have recently adopted a proximal aortic perfusion technique usmg a prosthetic graft attached to the left subclavian artery via a left posterolateral thoracotomy for hypothermic thoracic and thoraco-abdominal aneurysm repair (Fig I). The left subclavian artery with easy access from a thoracotomy has been used for a systemic-pulmonary arterial shunt using a prosthetic graft in neonates or small infants. 3 Moreover, the axillary artery is also widely known as an alternative site for proximal arterial
Fig 1. Left subclavian artery (LSA) cannulation technique through a posterolateral thoracotomy. After proximal and distal control of the LSA is obtained, an 8-mm sealed woven Dacron graft is attached to the LSA. followed by cannulation with a polyvinyl arch cannula. Venous drainage is through a straight venous cannula placed in the main pulmonary artery (PA).