Controlled Hypotension Decreases Blood Transfusion Requirement: Fact or Fallacy?

INTRODUCTION



Gardner is considered as the first author to report an operation with controlled hypotension (CH) <|[1]|>. Nevertheless he has controlled hypotension with arteriotomy, which trangresses the physiological principles on which deliberate hypotension is based. In fact, the concept of induced-hypotension is to produce a reduction in peripheral vascular resistance, and thus hydrostatic pressure, while maintaining cardiac output <|[2, 3]|>. Pioneering studies on CH were performed in the 1950s by Enderby with the introduction of ganglionic blocking drugs <|[2]|>. In 1948, Griffiths and Gillis used high spinal anesthesia with postural adjustment to lower the arterial pressure and also obtained a dry field of operation <|[4]|>. But this technique does not really refer to CH and will not be addressed in this review.



The amount of blood loss during operation is mainly determined by the surgical technique and the methods used to secure hemostasis. Along with CH there are a variety of autotransfusion (ATS) and hemodilution techniques which can reduce the need for transfusion of red cells <|[5]|>. The clinical story of CH began in 1946 <|[1]|> and since this date has undergone 3 periods: the first devoted to the reduction of bleeding to improve surgical technique; the second to decrease the need in allogenic blood transfusion (ABT) in order to reduce exposure to blood and the risk of viral contamination; - the third to improve the global quality of management with a reduction of cost. Unfortunately the first period led to a major number of studies and in most of these works there is no report of the ABT, particularly in term of patients who were transfused. Also the level of hemoglobin for transfusion is rarely defined. Presently a 7 g/dL level is accepted for the intraoperative period and 9 g/dL of hemoglobin at the end of hospitalization which reduced the total blood requirement. It was recently demonstrated that such a policy in critically ill patients did not impede the prognosis. On the contrary mortality could be reduced, except for patients with cardiac disease, particularly coronary disease, for whom 9 g/dL should be preferred <|[6]|>. Such a policy was not adopted before 1990 and probably led to a more liberal transfusion.



One present question is to associate two techniques and evaluate their potential interest and risks. In any case it is capital to remember the progress in ABT which can be considered as possibly the best and most secure method of blood transfusion <|[7]|>. Therefore, the risk of combining CH with various techniques of ATS or blood salvage with retransfusion must be weighed since this combination with surgical stimuli can lead to marked decrease in SvO2 indicating impairment in tissue oxygenation <|[8]|>.

CH has several advantages <|[9]|>. It provides more technical ease for the surgeon because of the bloodless field. Incidently by reducing the amount of ligated or cauterized tissue it also prevents infection. By decreasing oozing beneath skin flaps, it can promote better healing and decrease postoperative edema in plastic operations. Finally it reduces the need for blood transfusion. This technique has been commonly used in the clinical situations summarized in the <|[table I]|> <|[2, 3, 9, 10]|>. However, it is no
longer used in many of these surgical operations particularly in neurosurgery. Inversely moderate CH always contributes to a dry field in ENT. The real advantage of CH in minimizing blood loss in operation associated with serious hemorrhage and the impact on blood transfusion requirement remains to be evaluated. Therefore this review is focused only on ABT reduction induced by CH. It is not devoted to the principles and techniques of CH which are detailed in other reviews <|[2, 3, 9, 10]|>. This review will scrutinize the studies performed in hip surgery, spinal surgery for scoliosis, neck and head surgery and miscellanous surgeries, for red cell transfusion requirements.



HIP SURGERY <|[Table II]|>



All the studies concern total hip arthroplasty. Davies et al. <|[11]|> conducted a prospective study for CH compared retrospectively to patients operated with a normotensive anesthesia. CH decreased replacement with blood and albumin by 50%. The average blood replacements during the entire hospital stay for the normotensive group and the CH group in the patients with no prior hip surgery were 2250 mL and 1150 mL respectively and 2965 mL and 1730 mL in patients with prior hip surgery. In the CH group details were given concerning ABT which represented 50% of blood replacement. Amaranath et al. <|[12]|> reported total ABT (intra and postoperative ABT) to be around 1000 mL in 29 patients who underwent CH with no result given in the normotensive group. In a subgroup of 11 patients operated on for bilateral hip replacement, total ABT was 1975 mL for the first operation (the first hip) performed without hypotension while total ABT for the second (the second hip) operated with CH was1090 mL. In the studies from the same team all patients with CH were Jehovahs witnesses and autotransfusion was performed during the intraoperative period by suction with retransfusion <|[13, 14]|>. Of course they were not transfused with homologous blood. The 5 control patients with normotensive anesthesia received 14 units of homologous blood. But no conclusion could be drawn from this experience since ATS was not used in the control group. The report by Gross <|[15]|> is interesting
because of the large series of patients, unfortunately the transfusion data is not well-analyzed. Apparently blood loss seemed to be calculated based on blood replacement which was reduced by 64% with CH: 1950 mL (range: 0-3000 mL) in the control group and 700 mL (0-5000 mL) in the CH group.

The amount of ABT was not given in Keiths study <|[16]|>, however the number of patients transfused was reported: 7 out of 8 patients in the control group, 7 out of 9 in the halothane group and 3 out of 8 in the epidural group. These results justify the interest of epidural anesthesia.

Thompson et al. <|[17]|> reported a significant reduction in the total ABT by CH: 1300200 mL in the control group versus 500120 mL and 23030 mL in the CH groups, halothane and SNP groups respectively. Only half of patients operated with CH were transfused, but no precise data was given for the control group.

Results reported by Eerola et al. <|[18]|> are biased because autotransfusion (ANH) is not used in the normotensive group. Therefore ABT was 2100200 mL in the normotensive group and 19070 mL in the CH group. In the latter group 56020 mL of autologous blood collected by ANH were transfused. Clearly there was at least 50% reduction in blood transfusion, but absence of ANH in the control group precluded any conclusion on reduction of exposure to homologous blood. In the study performed by Vazeery and Lunde no patients received blood transfusion <|[19]|>.

Barbier-Bhm et al. <|[20]|> reported an ABT of 500170 mL in the CH group and 1500200 mL in the control group, respectively 8 out of 11 and 10 out of 10 of patients received ABT. However the hemoglobin level at 10 days was 12 g/dL in the control group, in contrast to 10 g/dL in the hypotensive group. Therefore ABT seems to have been more liberal in the control group. This remark underlines the importance of a predefined acceptable level of hemoglobin for the postoperative period. In Rosberg et al. <|[21]|>, CH was compared to a slight hypotensive anesthesia with halothane and to neuroleptanalgesia (NLA). Patients in CH group needed significantly less ABT (1100150 mL) than patients in the halothane group (1700200 mL) or in the NLA group (2200460 mL). No ATS was used despite a large bleeding.

Also, the study by Qvist et al. <|[22]|> lacked ATS despite excessive bleeding. ABT was significantly lower in the CH group (720480 mL) than in the NLA group (21001300 mL). The number of patients exposed to ABT was not given. Fredin et al. <|[23]|> compared different techniques and particularly the effect of dextran for thromboprophylaxis. The choice of the colloid must be carefully considered since dextran may potentiate bleeding. In any case there were no differences for blood loss or ABT between sub-groups with or without dextran in association to CH and between the group with normotensive anesthesia. CH significantly reduced ABT, 920600 mL and 12001000 mL in the two groups compared to 15001000 mL in one normotensive group. The second normotensive group had ANH (1100800 mL) and ABT (1300950 mL). Unfortunatley ANH was not used with CH and the number of patients exposed to ABT was not given. The last comparative study evaluated two techniques of CH where ATS was not used <|[24]|>. But the number of patients exposed to ABT was reported: one and two out of 12 patients in each group.

Finally in a recent evaluation, ATS, ANH and peroperative blood salvage with retransfusion, was associated with a moderate CH <|[25]|>. This open non-randomized study confirmed the interest of combining various techniques to decrease ABT. With a triggering level of hemoglobin at 9 g/dL for transfusion, only 3 patients out of 43 received ABT.



In conclusion, CH reduces blood loss by 50% but the reduction in ABT is never evaluated specifically. In fact, it was considered as a secondary end-point in all studies and there was no strategy devoted to this aim, particularly association with predonation of blood, ANH or peroperative blood salvage. Since CH reduces blood loss when arterial pressure is lowered to 50 to 60 mmHg the authors may have feared a combination with hemodilution. The benefit in terms of reducing ABT is also largely biased, because the trigerring level of postoperative hemoglobin for transfusion is rarely defined, which clearly impedes the results in many studies. Furthermore we lack information about the exposure to ABT which is the end-point when blood transfusion is considered in surgery in the nineties. Finally it could be potentially valuable for total hip replacement to use CH combined with an ATS technique for operation in patients already operated from the hip.



SPINAL SURGERY FOR SCOLIOSIS <|[Table III]|>



The evaluation of Mc Neil et al. <|[26]|> showed a significant reduction in blood loss and ABT: 2270 mL in the control group and 1380 mL in the CH group. The triggering level of hemoglobin for transfusion was defined as 10 g/dL and transfusion of 2 units was performed for every 1 g/dL of hemoglobin under 10 g/dL, but the number of patients transfused was not given.

Grundy et al. <|[27]|> reported only the ABT during the 24 postoperative hours. There was a very small reduction in ABT: in the CH group 2.20.6 units versus 2.71 units in the normotensive group. In the study by Malcom-Smith and Mc Master <|[28]|> there was a significant reduction in ABT between the control group (4.92.3 units) and the CH group (2.20.9 units). Hemoglobin levels upon discharge from hospital were higher in the control group (1 g/dL as a mean) which accounted for a more restrictive strategy of ABT in the CH group. In the study performed by Patel et al. <|[29]|> the triggering level of hemoglobin for transfusion was clearly defined for the postoperative period, around 10 g/dL. In the CH group ABT was significantly reduced (490 mL) in comparison to the normotensive group (950 mL).

The results in the study by Broisin et al. <|[30]|> were expressed as blood loss per vertebrae fused. For 10 vertebrae fused blood loss should be around 3000 mL in the control group and 800 mL in the CH group. In a third group ANH was practised with a blood loss of around 1200 mL. Blood transfusion was not reported. In all these studies ATS was never used in combination with CH and the number of patients exposed to ABT was never given. Brodsky et al. <|[31]|> did not find any differences in blood loss between a non-Jehovah control group and a Jehovahs witness group operated with CH. Blood loss was independent of the level of arterial pressure. The main advantage of CH seemed to be the improvement of surgery leading to a reduction in operative duration. Of course no Jehovah witness patient was transfused and no information was available for the other patients. In the two last studies there was no control group and two techniques of CH were compared. In both studies predonation ATS was associated with intraoperative blood salvage by suction at the surgical site with retransfusion. Bernard et al. <|[32]|> reported 5 patients out of 20 received ABT and in Hersey et al. <|[33]|> only one out of 20 received one allogenic unit of packed red cells.



In conclusion, the importance of blood loss during spinal fusion may justify the practise of CH for some surgical teams. But ATS is now a necessity to really reduce the exposure to ABT. The combination of CH to hemodilution remains to be evaluated because of the risk of medullary ischemia <|[8]|>.



NECK AND HEAD SURGERY <|[Table IV]|>



In the old series reported by Warner et al. <|[34]|> in ENT surgery for cancer there was a significant difference between allogenic whole blood transfusion in the control group (2750 mL) and the two CH groups (1700 and 1500 mL). In orthognatic surgery no information concerning blood transfusion was available in the study of Fromme et al. <|[35]|>.

In the same type of surgery, Lessard et al. <|[36]|> chosed ABT when blood loss overcame 20% of the estimated blood volume. Therefore 12 out of 27 patients in the control group and 3 out of 25 in the CH group received ABT. In one more recent study performed by Schindler et al. <|[37]|> the authors were not concerned by the ABT which was not reported. They confirmed the results of Lessard et al. <|[36]|> that CH is sufficient to reduce bleeding from the osseous venous plexus opened during osteotomy but not the bleeding from mucosal and muscles. Samman et al. <|[38]|> reported a retrospective non comparative study with a moderate CH. It is worth of interest because 360 patients were included of whom 84 (24%) were transfused. Only 6 patients after single jaw surgery and iliac bone harvest (9%) received ABT and 78 (27%) after bimaxillary osteotomies. Forty-seven patients received one unit of blood, 25 had 2 units and 12 had more than 2 units. They concluded that CH is not useful in the first type of orthognatic surgery for which no patients should have received ABT if triggering hemoglobin had been predefined. In case of bimaxillary osteotomy only 4% of patients received more than 2 units but unfortunately this subgroup was not predictable. This subgroup would constitute the group of patients for whom ABT is probably necessary. For Enlund et al. <|[39]|> blood loss was small in both groups with or without CH and no patients required ABT.



In conclusion, in orthognatic surgery acceptation of a 7 to 9 g/dL of hemoglobin level during operation and hospitalization, with combination of ATS techniques if cancer is not concerned, could be an alternative to CH. In some ENT or orthognatic surgery for cancer where ATS may be questionnable CH could be mandatory.



DIFFERENT TYPES OF SURGERY <|[Table IV]|>



For thoraco-abdominal dissection of retroperitoneal lymph nodes, Stirt et al. <|[40]|> compared normotensive anesthesia to CH. The comparison of total perioperative blood transfusion showed that 32% of patients in the CH group versus 50% in the control group required ABT. The average ABT was 0.5 unit in the CH group, versus 1.5 units in the control group. Portal hypertension due to extrahepatic obstruction and non-cirrhotic portal fibrosis justified splenectomy and lineorenal shunt which can lead to severe hemorrhage. Sood et al. <|[41]|> demonstrated that ABT was reduced from 31 units in the control group to 11 unit in the CH group.

In urology for radical cystectomy, Alhering et al. <|[42]|> showed a significant reduction in ABT, 1600220 mL in the control group versus 700100 mL in the CH group, but the number of patients who required ABT was only decreased from 90% of patients in the control group to 70% in the CH group. The more recent study performed by Boldt et al. <|[43]|> deserves careful attention because it is the only modern attempt to compare CH, ANH with a control group, with a medical as well as an economic evaluation. The authors compared a variety of outcomes, particularly the number of patients exposed to ABT and the cost of transfusion and volume replacement in three groups of patients undergoing radical prostatectomy. The triggering level of hemoglobin for transfusion was defined. The blood transfusion requirement was lower in the CH group, consequently with a lower cost than in the other two groups <|[Table V]|>. There were no severe side-effects, particularly there were no patients with neurologic deficits or myocardial ischemia detected by troponin TnT increase. Surprisingly this study demonstrated the interest of CH in comparison to ANH which was rather disappointing. Unfortunately the combination of ANH and CH was not evaluated.



COMPLICATIONS



Enderby <|[2]|> reported only 6 complications in direct relation to CH in 9107 cases of hypotensive anesthesia. In all the studies concerned by this review few complications were reported with only one death attibuted to the technique in the oldest study <|[34]|>. Therefore CH can be considered as a safe technique since severe hemodilution or other complications are not associated with CH. Potential risk is linked to the presence of atherosclerosis and patients with carotid bruits or signs of myocardial
ischemia are classically excluded for CH.

Although very uncommon, ischemic optic neuropathy is well-documented <|[44, 45]|>. It was reported particularly in cardiac surgery and attributed to hypotension and anemia in arteriosclerotic patients. This severe complication responsible for blindness was recently reported after spinal fusion surgery with combined CH and hemodilution in 5 patients with hypertension or ischemic medical history <|[46, 47]|>. The mean arterial pressure ranged from 50 to 60 mmHg and hemoglobin was no less than 8 g/dL. These severe complications deserve great attention and further evaluation of CH associated to hemodilution is recommended. Clearly this combined technique must be avoided in arteriosclerotic patients.



CONCLUSION



During the last 15 years, the quality of surgical management has improved and only some surgeries typically associated with significant bleeding may justify the use of CH. But, for a similar surgery bleeding can be decreased with some surgical teams. For spinal surgery of scoliosis, some total hip replacements (redo, sepsis), CH could be a valuable technique.

CH reduces blood loss by half which allows a large reduction of ABT exposure when ATS techniques are associated. The finding that blood loss and transfusion requirements are lower in the CH group in the study of Boldt et al. <|[43]|> with rather disappointing results in the ANH group as predicted by mathematical models suggests for an editorialist that the combination of CH with hypervolemic hemodilution might be the most effective blood conservation option for such patients <|[34]|>. It is the responsability of the anesthesiologist to choose acceptable lower limits of hemoglobin and arterial pressure for the individual patient, although it would be helpful if nomograms could be derived from clinical controlled studies <|[48]|>. Clearly with the potential risks described in this review, the combination of CH (MAP around 50 to 60 mmHg) with hemodilution needs further evaluation. In elderly patients, the anesthesiologist must focus his attention on monitoring early detection of early any cardiac impairment. In some patients monitoring of Svo2 allows detection of any decrease indicative of cardiac impairment and may also contribute to decreasing the risks of medullary complications during scoliosis surgery.

Finally, at the end of the 90s with the improvement in safety in ABT, it is questionnable to risk profound hypotension to reduce blood transfusion requirements in most patients over 60 years old. Further prospective, randomized studies are required to assess the safety of the combination of CH and normo or hypervolemic hemodilution and to establish safe combined lower limits of hemoglobin and arterial pressure in the oldest and sickest population of surgical patients.

Appendix: 

References


1. Gardner WJ. The control of bleeding by induced hypotension. JAMA, 1946; 132: 572-574.

2. Enderby GE. Hypotensive anesthesia. Edinburgh: Churchill-Livingstone, 1985.

3. Pinaud M, Blanloeil Y, Bernard JM. Hypotension contrle. Encyclopdie Mdico-Chirurgicale (Paris, France), Anesthsie-ranimation, 36393-A10, 1990, 14p.

4. Griffiths HW, Gillies J. Thoracolumbar splanchnicectomy and sympathectomy: anaesthetic procedure. Anaesthesia, 1948; 3: 134-138.

5. Goodnough LT, Brecher ME, Kanter MH, Aubuchon JP. Transfusion medicine and conservation. N Engl J Med, 1999; 34 (part II): 525-533.

6. Hbert PC, Wells G, Blajchman MA, Marshall J, Martin C, Pagliarello G, Tweeddale M, Schwietzer I, Yetisis E, Transfusion Requirements in Critical Care Investigators for the Canadian Critical Care Trials Group. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. N Engl J Med, 1999; 340: 409-417.

7. Goodnough LT, Brecher ME, Kanter MH, Aubuchon JP. Transfusion medicine and conservation. N Engl J Med, 1999; 34 (part I): 438-447.

8. Bernard JM, Passuti N, Pinaud M. Combination of acute normovolemic hemodilution to hypotensive anesthesia (letter). Spine, 1993; 18: 1560-1562.

9. Fahmy NM. Indications and contra-indications for deliberate hypotension with a review of its cardiovascular effects. Intern Anesthesiol Clin, 1979; 17: 175-187.

10. Sollevi A. Hypotensive anesthesia and blood loss. Acta Anaesthesiol Scand, 1982; 32 (Suppl 89): 39-43.

11. Davis NJ, Jennings JJ, Harris WH. Induced hypotensive anesthesia for total hip replacement. General Orthoped, 1974; 101: 93-98.

12. Amaranath L, Cascorbi HF, Singh-Amanarath AV, Frankmann DB. Relation of anesthesia to total hip replacement and control of operative blood loss. Anesth Analg, 1975; 54: 641-648.

13. North ER, Nelson CL, Lawson NW. Prevention of blood loss in total hip replacement surgery by hypotensive anesthesia and intra-operative autotransfusion. Surg Forum, 1975, 26: 517-518.

14. Lawson NW, Thompson DS, Nelson CL, Flacke JW, North ER. Sodium nitroprusside-induced hypotension for supine total hip replacement. Anesth Analg, 1976; 55: 654-662

15. Gross HP. Hypotensive anesthesia in total hip replacement. Orthop Surg, 1997; 30: 27-29.

16. Keith I. Anaesthesia and blood loss in total hip replacement. Anaesthesia, 1997; 32: 444-540.

17. Thompson GE, Miller RD, Stevens WC, Murray WR. Hypotensive anesthesia for total hip arthroplasty: a study of blood loss and organ function (brain, heart, liver, and kidney). Anesthesiology, 1978; 48: 91-96.

18. Eerola R, Eerola M, Kaukinen L, Kaukinen S. Controlled hypotension and moderate haemodilution in major hip surgery. Ann Chir Gynaecol, 1979; 68: 109-113.

19. Vazeery AK, Lunde O. Controlled hypotension in hip joint surgery. Acta Orthop Scand, 1979; 50: 433-441.

20. Barbier-Bhm G, Desmonts JM, Couderc E, Moulin D, Prokocimer P, Olivier H. Comparative effects of induced hypotension and normovolemic haemodilution on blood loss in total hip arthroplasty. Br J Anaesth, 1980; 52: 1039-1043.

21. Rosberg B, Fredin H, Gustafson C. Anesthetic techniques and surgical blood loss in total hip arthroplasty. Acta Anaesthesiol Scand, 1982, 26: 189-193.

22. Qvist TF, Skovsted P, Bredgaard Srensen M. Moderate hypotensive anesthesia for reduction of blood loss during total hip replacement. Acta Anaesthesiol Scand, 1982; 26: 351-353.

23. Fredin H, Gustafsson G, Rosberg B. Hypotensive anesthesia, thromboprophylaxis and postoperative thromboembolism in total hip arthroplasty. Acta Anaesthesiol Scand, 1984; 28: 503-507.

24. Bernard JM, Pinaud M, Franois T, Babin M, Macquin-Mavier I, Letenneur J. Deliberate hypotension with nicardipine or nitroprusside during total hip arthroplasty. Anesth Analg, 1991; 73: 341-345.

25. Aragou M, Dufieux PL, Carrat C, Gory C, Olivier H, Goursot G. Prvention du saignement dans l'arthoplastie totale de hanche par hmodilution, autotransfusion et hypotension contrle. Rev Chir Orthop, 1987; 87: 216-217.

26. Mc Neil TW, de Wald RL, Kuo KN, Bennett EJ,
Salem MR.
Controlled hypotensive anesthesia in scoliosis surgery. J Bone Joint Surg 1974; 56-A: 1167-1172.

27. Grundy BL, Nash CL, Brown RH. Deliberate hypotension for spinal fusion: prospective randomized study with evoked potential monitoring. Can Anaesth Soc J, 1982; 29: 452-462.

28. Malcom-Smith NA, McMaster MJ. The use of induced hypotension to control bleeding during posterior fusion for scoliosis. J Bone Joint Surg, 1983; 65-B: 255-258.

29. Patel NJ, Patel BS, Paskin S, Laufer S. Induced moderate hypotensive anesthesia for spinal fusion and Harrington-rod instrumentation. J Bone Joint Surg, 1985; 67-A: 1384-1387.

30. Broisin F, Tauty-Spiers C, Dardet R, Moskovtchenko JF. Techniques d'hmodilution et d'hypotension contrle dans la chirurgie du rachis scoliotique. Cah Anesth, 1987; 35: 281-282.

31. Brodsky JW, Dickson JH, Erwin WD, Rossi CD, Hypotensive anesthesia for scoliosis surgery in Jehovahs witnesses. Spine, 1991; 16: 304-306.

32. Bernard JM, Passuti N, Pinaud M. Long-term hypotensive technique with nicardipine and nitroprusside during isoflurane anesthesia for spinal surgery. Anesth Analg, 1992; 75: 179-185.

33. Hersey SL, ODell NE, Lowe S, Rasmussen G,
Tobias JD, Deshpande JK, Mencio G, Green N.
Nicardipine versus nitroprusside for controlled hypotension during spinal surgery in adolescents. Anesth Analg, 1997; 84: 1239-1244.

34. Warner WA, Shumrick DA, Caffrey JA. Clinical investigation of prolonged induced hypotension in head and neck surgery. Br J Anaesth, 1970; 42: 39-44.

35. Fromme GA, Mac Kenzie RA, Gould AB, Lund BA, Offord KP. Controlled hypotension for orthognatic surgery Anesth Analg, 1986; 65: 683-686.

36. Lessard MR, Trpanier CA, Baribault JP, Brochu JG, Brousseau CA, Cot JJ, Denault PH. Isoflurane-induced hypotension in orthognatic surgery. Anesth Analg, 1989; 69: 379-383.

37. Schindler I, Andel H, Leber J, Kimla T. Moderate induced hypotension provides satisfactory operating conditions in maxillofacial surgery. Acta Anaesthesiol Scand, 1994; 38: 384-387.

38. Samman N, Cheung LK, Tong AC, Tiderman H. Blood loss and transfusion requirements in orthognatic surgery. J Oral Maxillofacial Surg, 1994; 54: 21-24.

39. Enlund MG, Ahistedt BL, Andersson LG, Kierkmanov LI. Induced hypotension may influence blood loss in orthognatic surgery, but it is not crucial. Scand J Plastic Reconstr hand Surg, 1997; 31: 311-317.

40. Stirt JA, Korn DE, Reynolds RC. Sodium nitroprusside-induced hypotension in radical thoraco-abdominal dissection of retroperitoneal lymph nodes. Br J Anaesth, 1980; 52: 1045-1048.

41. Sood S, Jayalami TS, Vijayaraghavan S, Nundy S. Use of sodium nitroprusside induced-hypotensive anaesthesia for reducing blood loss in patients undergoing lienorenal shunts for portal hypertension. Br J Surg, 1987; 74: 1036-1038.

42. Ahlering TE, Henderson JB, Skinner DG. Controlled hypotension anesthesia to reduce blood loss in radical cystectomy for bladder cancer. J Urol, 1983: 129: 953-954.

43. Boldt J, Weber A, Mailer K, Papsdorf M, Schuster P. Acute normovolemic haemodilution vs controlled hypotension for reducing the use of allogenic blood in patients undergoing radical prostatectomy. Br J Anaesth, 1999; 82: 170-174.

44. Gild WM, Posner KL, Caplan RA, Cheney FW. Eye injuries associated with anesthesia. A closed claims analysis. Anesthesiology, 1992; 76: 204-208.

45. Williams EL, Hart WH, Tempelhoff R. Postoperative ischemic optic neuropathy. Anesth Analg, 1995; 80: 1018-1029.

46. Katz DM, Trobe JD, Cornblath WT, Kline LB, Ischemic optic neuropathy after lumbar spine surgery. Arch Ophalmol, 1994; 112: 925-931.

47. Roth S, Thisted RA, Erickson JP, Black S,
Schreider BD.
Eye injuries after non ocular surgery; A study of 60, 965 anesthetics from 1988 to 1992. Anesthesiology, 1996; 85: 1020-1027.

48. Levack ID, Gillon J. Intraoperative conservation of red cell mass: controlled hypotension or haemodilution: not necessarily mutually exclusive? Br J Anaesth, 1999; 82: 161-163.