- Email: [email protected]
IVOX, OXYGEN BALANCE AND THE HEART
920
BRITISH JOURNAL OF ANAESTHESIA
O. A. MERETOJA K. T. OLKKOLA
University of Helsinki Helsinki 1. Meretoja OA, Olkkola KT. Pharmacodynamics of mivacurium in children, using a computer-controlled infusion. British Journal of Anaesthesia 1993; 71: 232-237. 2. Alifimoff JK, Goudsouzian NG. Continuous infusion of mivacurium in children. British Journal of Anaesthesia 1989; 63: 520-524. 3. Brandom BW, Sarner JB, Woelfel SK, Dong ML, Horn MC, Borland LM, Cook DR, Foster VJ, McNulty BF, Weakly JN. Mivacurium infusion requirement in pediatric surgical patients during nitrous oxide-halothane and during nitrous oxide-narcotic anesthesia. Anesthesia and Analgesia 1990; 71: 16-22. 4. Meretoja OA. Neuromuscular blocking agents in paediatric patients: influence of age on the response. Anaesthesia and Intensive Care 1990; 18: 440-148. 5. Brandom BW. Neuromuscular blocking drugs. Anesthesiology Clinics of North America 1991; 9: 781-800. 6. Meretoja OA, Wirtavuori K. Influence of age on the dose-response relationship of atracurium in paediatric patients. Acta Anaesthesiologica Scandinavica 1988; 32: 614-618. 7. Anderson RH, Macartney FJ, Shinebourne EA, Tynan M. Fetal circulation and circulatory changes at birth. In: Anderson RH, Macartney FJ, Shinebourne EA, Tynan M, eds. Paediatric Cardiology, Vol. I. Edinburgh: Churchill Livingstone, 1987; 113.
FENTANYL AND THE EXTRADURAL SIEVE Sir,—Dr Tehan's report [1] prompts us to describe a similar event. A 35-yr-old primipara presented at term in spontaneous labour. She had previously undergone excision of a small subserous uterine myoma, but was not felt to have decreased uterine integrity as a result. After 8 h of effective contractions, she had reached 4 cm of cervical dilatation. An extradural catheter was
placed at the L2-3 interspace, and good analgesia obtained with 0.25% bupivacaine 9 ml and fentanyl 100 ug. A continuous infusion of 0.2 % bupivacaine with fentanyl 2 ug ml"1 at 9 ml h"1 was commenced and provided good analgesia for the next 8 h. Fourteen hours into labour, the fetal heart rate began to show intermittent late decelerations and lack of baseline variability. The infant was delivered by urgent Caesarean section. At no time after placement of the extradural did the mother experience breakthrough pain. At laparotomy, a complete placental abruption and rupture of the uterus into the left broad ligament were found. The extent to which the opioid component of this patient's extradural analgesia masked the clinical signs of the uterine rupture and placental abruption is a matter for speculation. Our patient received extradural fentanyl from the outset, and we were thus unable to demonstrate any "extradural sieve" that might have been present. We agree that the safety of extradural opioids in labouring patients with a scarred uterus would provide an interesting subject for further study. o. ixASHIQ L. J. HUSTON
University of Alberta Hospitals Edmonton, Alberta, Canada 1. Tehan B. Abolition of the extradural sieve by addition of fentanyl to extradural bupivacaine. British Journal of Anaesthesia 1992; 69: 520-521.
IVOX, OXYGEN BALANCE AND THE HEART Sir,—The Editorial on IVOX [1] referred to some of the potential gains from this technique in acute pulmonary failure. They include a reduction in mechanical lung damage, a reduction in the minimum inspired oxygen concentration and, perhaps most significant, an improvement in gas exchange. At this early stage of its development it seems important to differentiate clearly between hypoxia and hypoxaemia and, ideally, to use oxygen balance as the yardstick to assess any benefit from an increase in arterial oxygen content. Unfortunately, there is still no reliable, sensitive and real-time method for determining global and regional oxygen debt, and intramucosal gastric pH (pHi) monitoring [2] has yet to be firmly established as a more reliable guide to global oxygen balance than the uncertainties of blood lactate. Despite these limitations, it seems fair to require a swift improvement in total oxygen uptake, lactic acidosis, or a small pHi, as certain evidence of a beneficial effect on oxygen balance from an increase in arterial oxygenation with the use of IVOX in critically ill patients, although an attributable reversal of depressed organ function [3] (for example improved haemodynamics and urine output) may currently be the only favourable sign when lactate and pHi are normal. Theoretically, the oxygen content of jugular venous blood might also be helpful in this context. An increase with IVOX would signify a reduction in oxygen extraction and suggest the patient was hypoxaemic rather than hypoxic, his oxygen demand now being satisfied partially by an increase in the oxygen content of mixed venous blood. Where oxygen extraction is defective and fixed, as in pathological supply dependency from sepsis or the adult respiratory distress syndrome, there might be no change in jugular venous oxygen content, whatever the impact of IVOX on oxygen balance. Besides improving its oxygenation, enhancement of oxygen uptake and correction of hypoxaemia might also have an indirect beneficial effect on the heart by reducing its workload, an especially valuable effect if the workload is made greater by an increased oxygen demand and the heart is depressed by disease or sepsis. However, if the patient remains hypoxic and supplydependent despite correction of hypoxaemia, it seems unlikely that cardiac workload would decline. Measuring such benefit would obviously demand special monitoring. At the moment, most reports on IVOX [4-7] do not seem to address these considerations, but unless we approach IVOX in this critical way we may continue to confuse hypoxaemia with hypoxia and generate the same unjustified enthusiasm for its impact on oxygen balance in acute respiratory failure as many clinicians still have for positive end-expiratory pressure (PEEP) [8]. A. GILSTON
20 Hocroft Avenue London NW2 2EH
Downloaded from http://bja.oxfordjournals.org/ at University of Bath Library & Learning Centre on June 7, 2015
to remove a single value which is outside 3SD from the mean, provided that one describes the "outlier" in the results. However, we demonstrated that both the infusion requirement of mivacurium to maintain a 50% neuromuscular block and the onset time of mivacurium were dependent on age [1]. Dr Goodman suggests that we should have removed one patient from both of these data sets. We felt that this was not justified as we could not identify a reason why these patients were exceptional in their response, and because statistically significant regression persisted after these two patients were removed from the regression lines, as noticed also by Dr Goodman. Furthermore, we felt that our findings are clinically important. Clinicians should appreciate that the infusion requirement of mivacurium is even greater in younger children than in older children, and onset time is shortest in youngest children. Our regression lines should not be used to predict any individual's response because of the wide individual variation in the responses. Variability in infusion rate was found also in other paediatric studies [2, 3]. It is of interest that one of these studies analysed age-dependence of infusion requirement of mivacurium in children, and also found a significant negative correlation [3]. Also, in our new series of 40 children aged 1—15 yr, a negative correlation existed between age and infusion requirement of mivacurium to maintain a 90% neuromuscular block (I ss 90 = -34xAge+1000ugkg- 1 lr 1 ; r = 0.514; P = 0.0007) [Meretoja and colleagues, unpublished data]. Thus it seems that our conclusions are valid. Our data on the onset time of mivacurium are consistent with previous findings with other neuromuscular blocking drugs [4-6]. Short onset time is likely to be caused by a short circulation time, which is a physiological characteristic of children [7]. Regression analysis should not be used to make predictions outside the range of the independent variable. Dr Goodman is setting up a straw man when he estimates infusion rates and onset times far beyond the age range of our patients. It is clearly not appropriate to predict infusion rates or onset times for neonates or adults on the basis of our published data from children aged 1-15 yr[l].
CORRESPONDENCE
921
VENOUS AIR EMBOLISM WITH A WATER JET DISSECTOR Sir,—We read with interest the article by Dr Smith [1], reporting an incident of apparent major air embolism occurring during hepatic resection using a water jet dissector. We have introduced a water jet dissector (Hepatotom) [2—4] at the Clinic of Visceral and Transplantation Surgery, University of Bern, Switzerland and since 1989 have performed about 50 liver resections with it. Air embolism during hepatic resection is a result of injury to large intrahepatic veins during parenchymal dissection. Negative or low hepatic i.v. pressure results in air being aspirated into these veins. Embolism might also result from cutting more superficially situated veins and presumably this could occur even with injury to smaller veins. Interestingly, Mastragelopulos and colleagues [5] reported on argon gas embolism occurring during laparoscopic cholecystectomy and coagulation of the liver bed with the Argon Beam Coagulator. The authors recommended caution and suggested keeping the instrument a reasonable distance from vessels and allowing for possible vessel anomalies, with relatively large vessels being close to the liver surface. The same precautions are necessary during use of the water jet dissector. The report by Dr Smith stated specifically that during parenchymal transsection their patient was in a supine position with a 30° head-up tilt. This is potentially dangerous during liver resection of any kind and we recommend that during the parenchymal phase of the operation the patient should be in a Trendelenburg position of about 15°. This mandatory central venous hypotension is used during parenchymal transsection. This technique averts the risk of air or gas embolism. Using this approach, the danger of air embolism during parenchymal dissection is minimal and one of us (L.H.B.) has performed 150 major hepatic resections without such an incident. H. U. BAER L. H. BLUMGART
University of Bern Bern, Switzerland 1. Smith JAS. Possible venous air embolism with a new water jet dissector. British Journal of Anaesthesia 1993; 70:466-467. 2. Baer HU, Maddern GJ, Blumgart LH. New water-jet dissector: initial experience in hepatic surgery. British Journal of Surgery 1991; 78: 502-503. 3. Baer HU, Maddern GJ, Blumgart LH. Hepatic surgery facilitated by a new jet dissector. Hepatobiliary Surgery 1991; 4: 146. 4. Baer HU, Stain SC, Guastella T, Maddern GJ, Blumgart LH. Hepatic resection using a water jet dissector. Hepatobiliary Surgery 1993; 6: 189-198. 5. Mastragelopulos N, Sarkar MR, Kaissling G, Bahr R, Daub D. Argongas-Embolie wahrend laparoskopischer Cholecystektomie mit dem Argon-Beam-One-Coagulator. Chirurgie 1992; 63: 1053-1054.
Sir,—Thank you for the opportunity to reply to the comments of Drs Baer and Blumgart relating to my Case Report of probable venous air embolism (VAE). Their letter raises an interesting issue regarding optimum positioning for hepatic surgery. A head-up tilt has the advantage of providing excellent surgical access to the liver and, with appropriate volume loading to counter postural hypotension, should maintain positive hepatic venous pressures and make passive VAE unlikely. However, any air that does entrain passes readily to the right atrium and, if sufficient, produces pulmonary hypertension and allows paradoxical systemic air embolism in event of a patent foramen ovale. Furthermore, systemic air is more likely to gain access to the innominate and carotid arteries with head-up tilt, inviting cerebral arterial gas embolism. Venous capacitance vessels are located predominantly in the pelvic and lower limb structures and consequently a head-down position increases central venous pressure as measured with a right atrial catheter. The liver lies in plane similar to that of the heart in this position, with its major veins draining into the inferior vena cava just below the right atrium, so it may be that mean hepatic venous pressure also increases with head-down tilt compared with horizontal or head-up positioning. This should protect against VAE and any bubbles that might be entrained should float caudally in a large-bore, low-pressure vessel such as the inferior vena cava. The disadvantages of head-down attitude include less satisfactory surgical exposure with venous congestion and bowel tending to invade the field and increased right atrial pressures perhaps tending to open a patent foramen ovale, with obvious risks if air were to reach the right atrium. Current recommendations for management of cerebral arterial gas embolism advocate horizontal rather than head-down posturing. Bubbles are thought to accumulate anteriorly in the aortic outflow tract and not enter the major neck vessels and the adverse consequences of cerebral venous congestion are avoided. The approximate 30° head-up tilt in the case reported was certainly greater than usual in our community, where positioning varies from horizontal to about 15° head-up, according to surgical preference. As far as I am aware, no specialist hepatobiliary surgeons here use a head-down tilt. The matter is clearly controversial and may remain dictated largely by surgical considerations. On balance, I feel that the horizontal position offers the best compromise, but concede that the experience of Drs Baer and Blumgart validates their preference for the head-down posture. Head-up tilt would seem the most dangerous option, even though the operative site is nominally below the heart, and I agree with your correspondents that it should be discouraged. In conclusion, however, I must reiterate the purpose of my article in drawing attention to the need for addressing the potential of VAE in association with the water jet dissector. The incident reported involved dramatic air entrainment of proportions unlikely to be of passive nature, and the warning to anaesthetists remains serious. J. A. S. SMITH
Royal Adelaide Hospital Adelaide, Australia
SPIN LABEL TECHNIQUES FOR DETECTION OF MALIGNANT HYPERTHERMIA Sir,—The differences in the results obtained by Ohnishi and colleagues [1], Halsall, Ellis and Knowles [2] and Cooper and colleagues [3] may not be solely the result of technical factors relating to EPR spectra, as suggested by Halsall, Ellis and Knowles [4]. It is unclear from the paper by Ohnishi and colleagues [1] which sodium phosphate salt was used. They referred to sodium phosphate (presumably Na3PO4). Halsall, Ellis and Knowles [2] have used sodium dihydrogen phosphate. The first invariably requires correction to pH 7.4. It is unclear what effect this has on the availability of the other salts in the buffers. Calcium has been shown to affect both protein and lipid [5] structure in erythrocytes. Both the incubation and wash buffers used by Halsall, Ellis and Knowles [2] contained calcium. Methods involving lipid membrane labels usually involve a short incubation period, allowing interpolation of the membrane probe in the bilayer associated with evaporation (or dilution) of the solvent. Alternatively, exchange from Bovine Serum Albumen
Downloaded from http://bja.oxfordjournals.org/ at University of Bath Library & Learning Centre on June 7, 2015
1. Skoyles J, Pepperman M. I VOX. British Journal of Anaesthesia 1993; 70: 603-604. 2. Fiddian-Green RG. Tonometry: theory and applications. Intensive Care World 1992; 9: 60-65. 3. Von Segesser LK, Schaffher A, Stocker R. Extended (29 days) use of intravascular gas exchanger. Lancet 1992; 339: 1536. 4. Kallis P, Al-Saady NM, Bennett D, Treasure T. Clinical use of intravascular oxygenation. Lancet 1993; 337: 549. 5. High KM, Snider MT, Richard R. Clinical trials of an intravenous oxygenator in patients with adult respiratory distress syndrome. Anesthesiology 1992; 77: 856-863. 6. Schmidt H, Bohrer H, Motsch J, Tanzeem A. Intravenous oxygenator use in adult respiratory distress syndrome. Anesthesiology 1993; 78: 1193-1194. 7. Kallis P, Al-Saady NM, Bennett ED, Treasure T. Intravascular oxygen supplementation. Thorax 1992; 47: 215P-216P. 8. Gilston A. PEEP and oxygen balance. Where are the Emperor's clothes? Intensive and Critical Care Digest 1990; 9: 7-13.
BRITISH JOURNAL OF ANAESTHESIA
O. A. MERETOJA K. T. OLKKOLA
University of Helsinki Helsinki 1. Meretoja OA, Olkkola KT. Pharmacodynamics of mivacurium in children, using a computer-controlled infusion. British Journal of Anaesthesia 1993; 71: 232-237. 2. Alifimoff JK, Goudsouzian NG. Continuous infusion of mivacurium in children. British Journal of Anaesthesia 1989; 63: 520-524. 3. Brandom BW, Sarner JB, Woelfel SK, Dong ML, Horn MC, Borland LM, Cook DR, Foster VJ, McNulty BF, Weakly JN. Mivacurium infusion requirement in pediatric surgical patients during nitrous oxide-halothane and during nitrous oxide-narcotic anesthesia. Anesthesia and Analgesia 1990; 71: 16-22. 4. Meretoja OA. Neuromuscular blocking agents in paediatric patients: influence of age on the response. Anaesthesia and Intensive Care 1990; 18: 440-148. 5. Brandom BW. Neuromuscular blocking drugs. Anesthesiology Clinics of North America 1991; 9: 781-800. 6. Meretoja OA, Wirtavuori K. Influence of age on the dose-response relationship of atracurium in paediatric patients. Acta Anaesthesiologica Scandinavica 1988; 32: 614-618. 7. Anderson RH, Macartney FJ, Shinebourne EA, Tynan M. Fetal circulation and circulatory changes at birth. In: Anderson RH, Macartney FJ, Shinebourne EA, Tynan M, eds. Paediatric Cardiology, Vol. I. Edinburgh: Churchill Livingstone, 1987; 113.
FENTANYL AND THE EXTRADURAL SIEVE Sir,—Dr Tehan's report [1] prompts us to describe a similar event. A 35-yr-old primipara presented at term in spontaneous labour. She had previously undergone excision of a small subserous uterine myoma, but was not felt to have decreased uterine integrity as a result. After 8 h of effective contractions, she had reached 4 cm of cervical dilatation. An extradural catheter was
placed at the L2-3 interspace, and good analgesia obtained with 0.25% bupivacaine 9 ml and fentanyl 100 ug. A continuous infusion of 0.2 % bupivacaine with fentanyl 2 ug ml"1 at 9 ml h"1 was commenced and provided good analgesia for the next 8 h. Fourteen hours into labour, the fetal heart rate began to show intermittent late decelerations and lack of baseline variability. The infant was delivered by urgent Caesarean section. At no time after placement of the extradural did the mother experience breakthrough pain. At laparotomy, a complete placental abruption and rupture of the uterus into the left broad ligament were found. The extent to which the opioid component of this patient's extradural analgesia masked the clinical signs of the uterine rupture and placental abruption is a matter for speculation. Our patient received extradural fentanyl from the outset, and we were thus unable to demonstrate any "extradural sieve" that might have been present. We agree that the safety of extradural opioids in labouring patients with a scarred uterus would provide an interesting subject for further study. o. ixASHIQ L. J. HUSTON
University of Alberta Hospitals Edmonton, Alberta, Canada 1. Tehan B. Abolition of the extradural sieve by addition of fentanyl to extradural bupivacaine. British Journal of Anaesthesia 1992; 69: 520-521.
IVOX, OXYGEN BALANCE AND THE HEART Sir,—The Editorial on IVOX [1] referred to some of the potential gains from this technique in acute pulmonary failure. They include a reduction in mechanical lung damage, a reduction in the minimum inspired oxygen concentration and, perhaps most significant, an improvement in gas exchange. At this early stage of its development it seems important to differentiate clearly between hypoxia and hypoxaemia and, ideally, to use oxygen balance as the yardstick to assess any benefit from an increase in arterial oxygen content. Unfortunately, there is still no reliable, sensitive and real-time method for determining global and regional oxygen debt, and intramucosal gastric pH (pHi) monitoring [2] has yet to be firmly established as a more reliable guide to global oxygen balance than the uncertainties of blood lactate. Despite these limitations, it seems fair to require a swift improvement in total oxygen uptake, lactic acidosis, or a small pHi, as certain evidence of a beneficial effect on oxygen balance from an increase in arterial oxygenation with the use of IVOX in critically ill patients, although an attributable reversal of depressed organ function [3] (for example improved haemodynamics and urine output) may currently be the only favourable sign when lactate and pHi are normal. Theoretically, the oxygen content of jugular venous blood might also be helpful in this context. An increase with IVOX would signify a reduction in oxygen extraction and suggest the patient was hypoxaemic rather than hypoxic, his oxygen demand now being satisfied partially by an increase in the oxygen content of mixed venous blood. Where oxygen extraction is defective and fixed, as in pathological supply dependency from sepsis or the adult respiratory distress syndrome, there might be no change in jugular venous oxygen content, whatever the impact of IVOX on oxygen balance. Besides improving its oxygenation, enhancement of oxygen uptake and correction of hypoxaemia might also have an indirect beneficial effect on the heart by reducing its workload, an especially valuable effect if the workload is made greater by an increased oxygen demand and the heart is depressed by disease or sepsis. However, if the patient remains hypoxic and supplydependent despite correction of hypoxaemia, it seems unlikely that cardiac workload would decline. Measuring such benefit would obviously demand special monitoring. At the moment, most reports on IVOX [4-7] do not seem to address these considerations, but unless we approach IVOX in this critical way we may continue to confuse hypoxaemia with hypoxia and generate the same unjustified enthusiasm for its impact on oxygen balance in acute respiratory failure as many clinicians still have for positive end-expiratory pressure (PEEP) [8]. A. GILSTON
20 Hocroft Avenue London NW2 2EH
Downloaded from http://bja.oxfordjournals.org/ at University of Bath Library & Learning Centre on June 7, 2015
to remove a single value which is outside 3SD from the mean, provided that one describes the "outlier" in the results. However, we demonstrated that both the infusion requirement of mivacurium to maintain a 50% neuromuscular block and the onset time of mivacurium were dependent on age [1]. Dr Goodman suggests that we should have removed one patient from both of these data sets. We felt that this was not justified as we could not identify a reason why these patients were exceptional in their response, and because statistically significant regression persisted after these two patients were removed from the regression lines, as noticed also by Dr Goodman. Furthermore, we felt that our findings are clinically important. Clinicians should appreciate that the infusion requirement of mivacurium is even greater in younger children than in older children, and onset time is shortest in youngest children. Our regression lines should not be used to predict any individual's response because of the wide individual variation in the responses. Variability in infusion rate was found also in other paediatric studies [2, 3]. It is of interest that one of these studies analysed age-dependence of infusion requirement of mivacurium in children, and also found a significant negative correlation [3]. Also, in our new series of 40 children aged 1—15 yr, a negative correlation existed between age and infusion requirement of mivacurium to maintain a 90% neuromuscular block (I ss 90 = -34xAge+1000ugkg- 1 lr 1 ; r = 0.514; P = 0.0007) [Meretoja and colleagues, unpublished data]. Thus it seems that our conclusions are valid. Our data on the onset time of mivacurium are consistent with previous findings with other neuromuscular blocking drugs [4-6]. Short onset time is likely to be caused by a short circulation time, which is a physiological characteristic of children [7]. Regression analysis should not be used to make predictions outside the range of the independent variable. Dr Goodman is setting up a straw man when he estimates infusion rates and onset times far beyond the age range of our patients. It is clearly not appropriate to predict infusion rates or onset times for neonates or adults on the basis of our published data from children aged 1-15 yr[l].
CORRESPONDENCE
921
VENOUS AIR EMBOLISM WITH A WATER JET DISSECTOR Sir,—We read with interest the article by Dr Smith [1], reporting an incident of apparent major air embolism occurring during hepatic resection using a water jet dissector. We have introduced a water jet dissector (Hepatotom) [2—4] at the Clinic of Visceral and Transplantation Surgery, University of Bern, Switzerland and since 1989 have performed about 50 liver resections with it. Air embolism during hepatic resection is a result of injury to large intrahepatic veins during parenchymal dissection. Negative or low hepatic i.v. pressure results in air being aspirated into these veins. Embolism might also result from cutting more superficially situated veins and presumably this could occur even with injury to smaller veins. Interestingly, Mastragelopulos and colleagues [5] reported on argon gas embolism occurring during laparoscopic cholecystectomy and coagulation of the liver bed with the Argon Beam Coagulator. The authors recommended caution and suggested keeping the instrument a reasonable distance from vessels and allowing for possible vessel anomalies, with relatively large vessels being close to the liver surface. The same precautions are necessary during use of the water jet dissector. The report by Dr Smith stated specifically that during parenchymal transsection their patient was in a supine position with a 30° head-up tilt. This is potentially dangerous during liver resection of any kind and we recommend that during the parenchymal phase of the operation the patient should be in a Trendelenburg position of about 15°. This mandatory central venous hypotension is used during parenchymal transsection. This technique averts the risk of air or gas embolism. Using this approach, the danger of air embolism during parenchymal dissection is minimal and one of us (L.H.B.) has performed 150 major hepatic resections without such an incident. H. U. BAER L. H. BLUMGART
University of Bern Bern, Switzerland 1. Smith JAS. Possible venous air embolism with a new water jet dissector. British Journal of Anaesthesia 1993; 70:466-467. 2. Baer HU, Maddern GJ, Blumgart LH. New water-jet dissector: initial experience in hepatic surgery. British Journal of Surgery 1991; 78: 502-503. 3. Baer HU, Maddern GJ, Blumgart LH. Hepatic surgery facilitated by a new jet dissector. Hepatobiliary Surgery 1991; 4: 146. 4. Baer HU, Stain SC, Guastella T, Maddern GJ, Blumgart LH. Hepatic resection using a water jet dissector. Hepatobiliary Surgery 1993; 6: 189-198. 5. Mastragelopulos N, Sarkar MR, Kaissling G, Bahr R, Daub D. Argongas-Embolie wahrend laparoskopischer Cholecystektomie mit dem Argon-Beam-One-Coagulator. Chirurgie 1992; 63: 1053-1054.
Sir,—Thank you for the opportunity to reply to the comments of Drs Baer and Blumgart relating to my Case Report of probable venous air embolism (VAE). Their letter raises an interesting issue regarding optimum positioning for hepatic surgery. A head-up tilt has the advantage of providing excellent surgical access to the liver and, with appropriate volume loading to counter postural hypotension, should maintain positive hepatic venous pressures and make passive VAE unlikely. However, any air that does entrain passes readily to the right atrium and, if sufficient, produces pulmonary hypertension and allows paradoxical systemic air embolism in event of a patent foramen ovale. Furthermore, systemic air is more likely to gain access to the innominate and carotid arteries with head-up tilt, inviting cerebral arterial gas embolism. Venous capacitance vessels are located predominantly in the pelvic and lower limb structures and consequently a head-down position increases central venous pressure as measured with a right atrial catheter. The liver lies in plane similar to that of the heart in this position, with its major veins draining into the inferior vena cava just below the right atrium, so it may be that mean hepatic venous pressure also increases with head-down tilt compared with horizontal or head-up positioning. This should protect against VAE and any bubbles that might be entrained should float caudally in a large-bore, low-pressure vessel such as the inferior vena cava. The disadvantages of head-down attitude include less satisfactory surgical exposure with venous congestion and bowel tending to invade the field and increased right atrial pressures perhaps tending to open a patent foramen ovale, with obvious risks if air were to reach the right atrium. Current recommendations for management of cerebral arterial gas embolism advocate horizontal rather than head-down posturing. Bubbles are thought to accumulate anteriorly in the aortic outflow tract and not enter the major neck vessels and the adverse consequences of cerebral venous congestion are avoided. The approximate 30° head-up tilt in the case reported was certainly greater than usual in our community, where positioning varies from horizontal to about 15° head-up, according to surgical preference. As far as I am aware, no specialist hepatobiliary surgeons here use a head-down tilt. The matter is clearly controversial and may remain dictated largely by surgical considerations. On balance, I feel that the horizontal position offers the best compromise, but concede that the experience of Drs Baer and Blumgart validates their preference for the head-down posture. Head-up tilt would seem the most dangerous option, even though the operative site is nominally below the heart, and I agree with your correspondents that it should be discouraged. In conclusion, however, I must reiterate the purpose of my article in drawing attention to the need for addressing the potential of VAE in association with the water jet dissector. The incident reported involved dramatic air entrainment of proportions unlikely to be of passive nature, and the warning to anaesthetists remains serious. J. A. S. SMITH
Royal Adelaide Hospital Adelaide, Australia
SPIN LABEL TECHNIQUES FOR DETECTION OF MALIGNANT HYPERTHERMIA Sir,—The differences in the results obtained by Ohnishi and colleagues [1], Halsall, Ellis and Knowles [2] and Cooper and colleagues [3] may not be solely the result of technical factors relating to EPR spectra, as suggested by Halsall, Ellis and Knowles [4]. It is unclear from the paper by Ohnishi and colleagues [1] which sodium phosphate salt was used. They referred to sodium phosphate (presumably Na3PO4). Halsall, Ellis and Knowles [2] have used sodium dihydrogen phosphate. The first invariably requires correction to pH 7.4. It is unclear what effect this has on the availability of the other salts in the buffers. Calcium has been shown to affect both protein and lipid [5] structure in erythrocytes. Both the incubation and wash buffers used by Halsall, Ellis and Knowles [2] contained calcium. Methods involving lipid membrane labels usually involve a short incubation period, allowing interpolation of the membrane probe in the bilayer associated with evaporation (or dilution) of the solvent. Alternatively, exchange from Bovine Serum Albumen
Downloaded from http://bja.oxfordjournals.org/ at University of Bath Library & Learning Centre on June 7, 2015
1. Skoyles J, Pepperman M. I VOX. British Journal of Anaesthesia 1993; 70: 603-604. 2. Fiddian-Green RG. Tonometry: theory and applications. Intensive Care World 1992; 9: 60-65. 3. Von Segesser LK, Schaffher A, Stocker R. Extended (29 days) use of intravascular gas exchanger. Lancet 1992; 339: 1536. 4. Kallis P, Al-Saady NM, Bennett D, Treasure T. Clinical use of intravascular oxygenation. Lancet 1993; 337: 549. 5. High KM, Snider MT, Richard R. Clinical trials of an intravenous oxygenator in patients with adult respiratory distress syndrome. Anesthesiology 1992; 77: 856-863. 6. Schmidt H, Bohrer H, Motsch J, Tanzeem A. Intravenous oxygenator use in adult respiratory distress syndrome. Anesthesiology 1993; 78: 1193-1194. 7. Kallis P, Al-Saady NM, Bennett ED, Treasure T. Intravascular oxygen supplementation. Thorax 1992; 47: 215P-216P. 8. Gilston A. PEEP and oxygen balance. Where are the Emperor's clothes? Intensive and Critical Care Digest 1990; 9: 7-13.