- Email: [email protected]
Effects of changes in mean arterial pressure on SjO2 during cerebral aneurysm surgery
British Journal of Anaesthesia 1995; 75: 527–530
Effects of changes in mean arterial pressure on SjO2 during cerebral aneurysm surgery E. MOSS, N. M. DEARDEN AND J. C. BERRIDGE
Summary Twenty-six patients requiring clipping of cerebral aneurysms were anaesthetized with propofol, alfentanil and atracurium infusions and their lungs ventilated mechanically to hypocapnia (3.4– 4.5 kPa). $ M2 was measured continuously with an Oximetrix fibreoptic oximetry catheter. Normovolaemia was maintained by observing the response of mean arterial pressure (MAP) and central venous pressure (CVP) to fluid administration. The response of $ M2 to increased MAP was noted and the lactate oxygen index (LOI) calculated at regular intervals. $ M2 measurements indicated a critical MAP of between 80 and 110 mm Hg in nine patients, and one patient had a persistently low $ M2 value despite an MAP of 110 mm Hg. An increase in MAP was associated with an increase in $ M2 in 19 patients (P : 0.001). When the effects of changes in ! D &2 were eliminated, this change was still significant (P : 0.004) (n : 9). Patients with an LOI 9 0.08 at any time during the procedure had a worse initial outcome (within the first day) (P : 0.02) than patients who had a normal LOI throughout. Long-term outcome was similar to those with a normal LOI. Increasing MAP did not have a consistent effect on LOI. Jugular bulb cannulation to assess hypoperfusion in conjunction with lactate measurements and calculation of LOI provide useful information on which to base the intra- and postoperative management of patients with subarachnoid haemorrhage. (Br. J. Anaesth. 1995; 75: 527–530) Key words Surgery, neurological. Complications, aneurysm. Arterial pressure, measurement. Oxygen, saturation. Veins, jugular, cannulation.
Improvements in surgical techniques for cerebral aneurysms and an increasing tendency to operate early, that is within the first 2–3 days after rupture, have removed the need for induced hypotension in most cases [1]. Attention has now focused on the need to maintain cerebral perfusion pressure (CPP) in order to improve cerebral perfusion and reduce cerebral ischaemia. Unless obvious from the preoperative nursing charts that a neurological deficit develops when mean arterial pressure (MAP) is less than a certain value, it is difficult to determine if
there is a critical MAP below which cerebral hypoperfusion, which may lead to cerebral ischaemia, occurs. The aim of this study was to determine if it is possible to identify this critical MAP in individual patients during anaesthesia and surgery by continuously monitoring jugular bulb oxygen saturation (SjO2 ) .
Patients and methods The study was approved by the local Ethics Committee and patients or their relatives gave informed consent. Twenty-six patients requiring clipping of a cerebral aneurysm, 20 following rupture and six unruptured, were anaesthetized with propofol, alfentanil and atracurium infusions and their lungs ventilated mechanically to hypocapnia (3.5–4.5 kPa). The inspired oxygen concentration was adjusted to maintain PaO at 13–18 kPa. All patients received mannitol 0.5 g kg91 followed by frusemide 0.5 mg kg91 to reduce brain bulk and minimize retraction pressures on the brain. In addition to standard monitoring, including a subclavian central venous pressure (CVP) catheter and radial artery cannula, an Oximetric fibreoptic oximetry catheter was inserted into the jugular bulb on the side showing predominant venous drainage on the cerebral angiogram. If venous phase cerebral angiography was not available, the catheter was placed in the right jugular bulb. MAP, CVP and SjO2 were recorded continuously throughout the procedure. Normovolaemia was maintained by observing the responses of MAP and CVP to fluid administration, and colloids were given as required to maintain a normal CVP. A haemoglobin concentration of greater than 10 g dl91 was maintained by giving blood if necessary. The patient’s temperature was maintained constant throughout the procedure using a warming blanket (Bair Hugger). Jugular venous and arterial blood samples were obtained for baseline readings and calibration of the Oximetric catheter, immediately after placement of the jugular bulb catheter and at appropriate intervals after changes in MAP. The samples were analysed for saturation and haemoglobin concentration (Corning co-oximeter), lactate (YSI 2300 stat), 2
E. MOSS, MD, FRCA, N. M. DEARDEN, MB, CHB, FRCA, J. C. BERRIDGE, MB, CHB, FRCA, Department of Anaesthetics, Leeds General Infirmary, Great George Street, Leeds LS1 3EX. Accepted for publication: April 25, 1995. This article is accompanied by Editorial II.
528
British Journal of Anaesthesia Table 1 SjO2 values before and after an increase in mean arterial pressure (MAP), critical MAP values (NI : not identified), maximum values of LOI and outcome in patients undergoing clipping of cerebral aneurysms (ND : no deficit, FD : focal neurological deficit, DC : depressed conscious level; G : good, MD : moderate, SD : severe disability, D : death). * Still hypoperfused, † cardiorespiratory failure, ‡ PaCO2 remained constant during measurements. E : elective (i.e. aneurysm had not ruptured)
Patient no.
Days after rupture
1‡ 2 3 4‡ 5 6 7 8 9‡ 10 11‡ 12 13 14 15 16‡ 17 18‡ 19 20‡ 21 22‡ 23 24 25 26‡
5 E 5 12 5 28 6 E 2 5 4 6 10 2 2 9 E 7 4 E E E 2 2 2 2
Initial SjO2
SjO2 after increase in MAP
Critical MAP
Maximum LOI
Initial
Long-term
54.8 91.3 59 62 55 49.1 47 56.3 53.4 58 36 71.5 50.6 52 74.2 42.1 87.4 65 88.4 55.9 40.7 65.9 84.9 40.5 54.4 52.7
69.4 82.1 66 82 74 62.7 59 65.3 58.7 86 63 — 69 — 79.4 48.5 83.8 76 89.8 62.7 56.7 64.4 69.3 61.6 60.4 72.5
NI NI NI NI NI 90 110 NI 90 NI 80 NI 80 110* NI 100 NI NI NI NI 100 NI NI 95–100 NI 85
0.043 0.226 0.15 0.222 0.115 0.095 0.073 0.075 0.003 0.169 0.218 0.027 0.081 0.007 0.020 0.024 0.251 0.03 0.195 0.004 0.002 0.028 0.044 0.014 0.055 0.079
ND FD FD FD/DC FD FD ND ND FD FD FD/DC ND ND FD/DC FD ND ND FD/DC FD/DC ND ND ND ND FD/DC ND ND
G SD D SD MD MD G G SD SD MD D† G D MD G G D SD G G G G MD G G
Outcome
blood-gas tension and potassium concentration (Corning). The lactate oxygen index (LOI) [2] was calculated using the following formula: LOI =
− AJDL(mmol litre −1 ) × 2.24 AJDO2 (ml dl −1 )
where AJDL : arterial to jugular difference for lactate and AJDO2 : arterial to jugular difference in oxygen content. AJDO2 = (SaO2 % - SjO2 %) × Hb ×1.34/100 + (PaO2 (kPa) - PjO2 (kPa))× 0.0225 The normal value for LOI is :0.03 and an LOI of 90.08 was taken to indicate significant cerebral ischaemia. MAP was manipulated deliberately only at the request of the surgeon or if the clinical condition or blood test results required. MAP was increased deliberately if it was, or decreased below, a level at which SjO2 was less than 54 %. I.v. fluids were used to increase MAP in the first instance. Dopamine or adrenaline were used only if fluid therapy was ineffective. A critical MAP was defined as MAP below which SjO2 was :54 %. This level of SjO2 indicates cerebral hypoperfusion [3]. The jugular bulb catheter was removed before the patient was returned to the neurosurgical ward. Initial outcome was judged from the Glasgow coma score and neurological observations in the first 24 h after surgery. Long-term outcome was determined from the discharge summary or, for patients with poorer
outcomes, follow-up at 1 month using the Glasgow outcome score [4]. Changes in SjO2 in response to increasing MAP were analysed using a paired t test. Outcomes were compared using chi-square with Yates’ correction factor.
Results MAP required elevation in 24 of 26 patients and SjO2 measurements indicated a critical MAP of between 80 and 1 10 mm Hg in nine patients (table 1). However, in only one of six patients undergoing elective aneurysm clipping was this a feature. Another patient (patient No. 14) had a persistently low SjO2 despite an MAP of 110 mm Hg. Increasing MAP was associated with an increase in SjO2 in 19 patients, no change in two and a reduction in three, two of whom had a simultaneous reduction in P aco2. The increase in SjO2 with increasing MAP was statistically significant (P : 0.001). However, in more than 50 % of these patients there was also an increase in P aco2 between SjO2 samples. If those patients with a hyperaemic cerebral circulation ( SjO2 975 %) and those with an increase in P aco2 between samples were excluded, the increase in SjO2 with increasing MAP was still significant (P 0.004) (n 9). There were three patients (Nos 2, 17 and 23) with SjO2 values 975 % in whom an increase in MAP caused an apparent reduction in CBF. In patient Nos 17 and 23, this was associated with a reduction in P aco2 but in patient No. 2 there was an increase in P aco2.
SjO2 during cerebral aneurysm surgery
529 ischaemia became worse in six and improved in four after an increase in MAP.
Discussion
Figure 1 Changes in SjO2 with increases in mean arterial pressure (MAP) in nine patients with a constant PaCO2 throughout the period of measurement. Each line represents one patient. The two patients with absent autoregulation show lines with almost vertical slopes. Table 2 Relationship between lactate oxygen index (LOI) and initial and long-term outcome. *P : 0.05 Normal LOI LOI 9 0.08 Initial outcome Good Poor Long-term outcome Good/moderate disability Severe disability/ vegetative/death
12 4
2* 8*
12
4
4
6
Table 3 Changes in lactate oxygen index (LOI) with increasing mean arterial pressure LOI decreased LOI increased LOI more ischaemic LOI less ischaemic Never ischaemic
11 13 6 4 14
There were eight patients in whom P aco2 remained constant and there were SjO2 measurements at three different MAP values. Four showed a threshold level of CPP above which SjO2 showed no further increase with increasing MAP. Three of the other four patients had SjO2 values 975 % and the other failed to demonstrate a critical threshold for CPP. In those patients with a constant P aco2 during the period of study, by observing the response of SjO2 to a change in MAP, it was possible to demonstrate absent autoregulation in two patients and impaired autoregulation in six others (fig. 1). Patients with an LOI 90.08 at any time during the procedure (n 10) had a worse initial outcome (P : 0.02) than patients who had a normal LOI throughout (n 16), but the long-term outcome was similar to those with a normal LOI (0.5 9 P 9 0.1) (table 2). Of patients with an LOI showing cerebral ischaemia, three patients had SjO2 values :54 %, four 54–75 % and three 975 %. Increasing MAP did not have a consistent effect on LOI (table 3). In patients with an LOI 90.08,
When the cerebral metabolic rate for oxygen remains constant, SjO2 measurements reflect CBF. In this study the depth of anaesthesia was maintained at a constant level by continuous infusion of propofol and alfentanil, and therefore the cerebral metabolic rate would have remained fairly constant. Measurement of SjO2 also indicates if CBF is sufficient to satisfy the oxygen demands of the tissues. SjO2 reflects the balance between oxygen supply and demand with values of less than 54 % indicating cerebral hypoperfusion and readings less than 40 % being associated with global cerebral ischaemia [3, 5, 6], although significant regional ischaemia may be present with a normal SjO2 , as evidenced by an increase in LOI. The results of this study showed that approximately 50 % of patients undergoing aneurysms clipped acutely after subarachnoid haemorrhage had a critical MAP below which there was cerebral hypoperfusion. In contrast, a critical MAP was identified in only one of six patients during clipping of aneurysms which had not previously ruptured. von Helden and colleagues [7] found that a similar percentage of patients in a coma after subarachnoid haemorrhage had periods when SjO2 was critical (50–55 %). LOI gave some indication of patients who were likely to develop neurological deficits in the first 24 h after surgery but had no predictive value for long-term outcome. This is not surprising because patients with deficits were treated aggressively with hypertensive, hypervolaemic and haemodilution therapy which reversed the ischaemia in many patients. Unfortunately, because this study was performed in patients undergoing surgery, it was not always possible to maintain P aco2 constant between measurements of SjO2. However, in those patients in whom P aco2 remained constant, SjO2 still increased when MAP was increased, confirming an association between MAP and SjO2. One of the most effective ways of increasing SjO2 is to increase P aco2. Hypocapnia is usually recommended during intracranial surgery to reduce brain bulk and retraction pressures. However, with the use of mannitol and frusemide to reduce brain bulk, this hypocapnia may not be necessary. This was not demonstrated in this study. The increase in SjO2 associated with increasing MAP could be the result of defective autoregulation after subarachnoid haemorrhage [8] and is similar to the minimal or absent autoregulatory response demonstrated in head injuries by Fortune and colleagues [9]. An alternative explanation is that the plateau of the autoregulation graph is incorrect. It is known that CBF increases slightly as arterial pressure increases within the range of autoregulation. It has been suggested that increasing cardiac output increases CBF, but Bouma and colleagues [10] have shown that CBF is not related to cardiac output even when autoregulation is impaired. Thus for a constant MAP, any increase in CBF associated with intra-
530 vascular expansion appears to be mediated by a decrease in blood viscosity rather than an increase in cardiac output. In two patients (fig. 1) cerebrovascular autoregulation was clearly absent and in several others it appeared to be impaired. There was some evidence of a critical threshold for CPP, similar to that reported in patients with head injury [11], above which SjO2 did not increase further as pressure increased. In this study, the decision as to which jugular bulb to cannulate was made on the basis of cerebral angiography if venous phase films had been taken. If these films were not available the right jugular bulb was cannulated because the majority of cortical blood flow drains down the right internal jugular vein in most patients [12]. It is possible that some changes in SjO2 and LOI were missed because blood from hypoperfused or ischaemic areas was draining down the side opposite to the jugular catheter. Therefore, it could be argued that the jugular bulb on the side of the lesion [2, 13] or both jugular bulbs should be catheterized [14] because cerebral AJDO2 and LOI can differ between the two jugular bulbs. The four patients with a normal LOI who had a poor initial outcome may have been examples of patients in whom the wrong jugular bulb had been cannulated so that ischaemia was not detected. Alternatively, there may have been cerebral infarction so that no lactate was produced. All had developed neurological deficits before the jugular catheter was removed. There were two patients with an LOI 90.08, indicating cerebral ischaemia, who had a good initial outcome. These patients had a period of incomplete ischaemia which was probably caused by inadequate cerebral perfusion pressure or temporary arterial occlusion. Clearly the problem had been corrected before the patient recovered from the anaesthetic. It is now accepted that CPP should usually be maintained at preoperative levels during cerebral aneurysm surgery [1]. This helps to maintain flow through vessels affected by vasospasm and optimizes collateral flow. Induced hypotension does not alter outcome in patients undergoing early surgery for cerebral artery aneurysms compared with normotension, and the incidence of intraoperative rerupture is not reduced [15]. In fact, the outcome is worse when hypotension is used during intraoperative rupture of an aneurysm than when tamponade and temporary clipping with maintenance of MAP are used to control haemorrhage [16]. As the main objective of induced hypotension in these procedures is to reduce the incidence of intraoperative rupture of the aneurysm, it would not seem appropriate to use it in most patients. The present results support maintenance of CPP during surgery on cerebral aneurysms, particularly when surgery is performed within a few days of rupture, and show that measurement of SjO2 allows determination of the minimum MAP that should be allowed during
British Journal of Anaesthesia surgery. There is a risk of rupture if MAP is increased before the aneurysm is clipped, but the use of SjO2 monitoring and measurement of LOI allows determination of an MAP which gives the best balance between the risks of cerebral ischaemia and aneurysm rupture in individual patients.
References 1. Archer DP, Shaw DA, Leblanc RL, Tranmer BI. Haemodynamic considerations in the management of patients with subarachnoid haemorrhage. Canadian Journal of Anaesthesia 1991; 38: 454–470. 2. Robertson CS, Narayan RK, Gokaslan ZL, Pahwa R, Grossman RG, Caram P, Allen E. Cerebral arteriovenous oxygen difference as an estimate of cerebral blood flow in comatose patients. Journal of Neurosurgery 1989; 70: 222–230. 3. Dearden NM. SjO2 and critical perfusion pressure after severe brain injury. British Journal of Intensive Care 1992 (Suppl. 1): 7–11. 4. Jennett B, Bond M. Assessment of outcome after severe brain damage. A practical scale. Lancet 1975; 1: 480–484. 5. Sheinberg M, Kanter MJ, Robertson CS, Contant CF, Narayan RK, Grossman RG. Continuous monitoring of jugular venous oxygen saturation in head injured patients. Journal of Neurosurgery 1992; 76: 212–217. 6. Gopinath SP, Robertson CS, Contant CF, Hayes C, Feldman Z, Narayan RK, Grossman RG. Jugular venous desaturation and outcome after head injury. Journal of Neurology, Neurosurgery and Psychiatry 1994; 57: 717–723. 7. von Helden A, Schneider GH, Unterberg A, Lanksch WR. Monitoring of jugular venous oxygen saturation in comatose patients with subarachnoid haemorrhage and intracranial haematomas. Acta Neurochirurgica Supplementum 1993; 59: 102–106. 8. Pickard JD, Boisvert DPJ, Graham DI, Fitch W. Late effects of subarachnoid haemorrhage on the response of the primate cerebral circulation to drug-induced changes in arterial blood pressure. Journal of Neurology, Neurosurgery and Psychiatry 1979; 42: 899–903. 9. Fortune JB, Feustel PJ, Weigle CGM, Popp AJ. Continuous measurement of jugular venous oxygen saturation in response to transient elevations of blood pressure in head-injured patients. Journal of Neurosurgery 1994; 80: 461–468. 10. Bouma GJ, Muizelaar PJ, Bandoh K, Marmarou A. Blood pressure and intracranial pressure–volume dynamics in severe head injury : relationships with cerebral blood flow. Journal of Neurosurgery 1992; 77: 15–19. 11. Chan K-H, Miller JD, Dearden NM, Andrews PJD, Midgley S. The effect of changes in cerebral perfusion pressure upon middle cerebral artery blood flow velocity and jugular bulb venous oxygen saturation after severe brain injury. Journal of Neurosurgery 1992; 77: 55–61. 12. Williams PL, Warwick R, Dyson M, Bannister LH. Gray’s Anatomy, 37th Edn. Edinburgh, London, Melbourne, New York: Churchill Livingstone, 1989; 793–805. 13. Robertson CS, Grossman RG, Goodman JC, Narayan RK. The predictive value of cerebral anaerobic metabolism with cerebral infarction after head injury. Journal of Neurosurgery 1987; 67: 361–368. 14. Chieregato A, Targa L, Mantovani G, Droghetti L, Zatelli R. Cerebral arteriovenous difference and lactate–oxygen index: a case of bilateral monitoring. Journal of Neurosurgical Anesthesiology 1994; 6: 43–47. 15. Innomata S, Mizuyama K, Sato S, Takahashi H, Saito S, Mizutani T, Yamaguchi H, Fukuda T, Naito H. The effect of deliberate hypotensive anaesthesia on the prognosis of patients who underwent early surgeries for ruptured cerebral aneurysm. Masui 1992; 41: 207–213. 16. Gianotta SL, Oppenheimer JH, Levy ML, Zelman V. Management of intraoperative rupture of aneurysm without hypotension. Neurosurgery 1991; 28: 531–535.
Effects of changes in mean arterial pressure on SjO2 during cerebral aneurysm surgery E. MOSS, N. M. DEARDEN AND J. C. BERRIDGE
Summary Twenty-six patients requiring clipping of cerebral aneurysms were anaesthetized with propofol, alfentanil and atracurium infusions and their lungs ventilated mechanically to hypocapnia (3.4– 4.5 kPa). $ M2 was measured continuously with an Oximetrix fibreoptic oximetry catheter. Normovolaemia was maintained by observing the response of mean arterial pressure (MAP) and central venous pressure (CVP) to fluid administration. The response of $ M2 to increased MAP was noted and the lactate oxygen index (LOI) calculated at regular intervals. $ M2 measurements indicated a critical MAP of between 80 and 110 mm Hg in nine patients, and one patient had a persistently low $ M2 value despite an MAP of 110 mm Hg. An increase in MAP was associated with an increase in $ M2 in 19 patients (P : 0.001). When the effects of changes in ! D &2 were eliminated, this change was still significant (P : 0.004) (n : 9). Patients with an LOI 9 0.08 at any time during the procedure had a worse initial outcome (within the first day) (P : 0.02) than patients who had a normal LOI throughout. Long-term outcome was similar to those with a normal LOI. Increasing MAP did not have a consistent effect on LOI. Jugular bulb cannulation to assess hypoperfusion in conjunction with lactate measurements and calculation of LOI provide useful information on which to base the intra- and postoperative management of patients with subarachnoid haemorrhage. (Br. J. Anaesth. 1995; 75: 527–530) Key words Surgery, neurological. Complications, aneurysm. Arterial pressure, measurement. Oxygen, saturation. Veins, jugular, cannulation.
Improvements in surgical techniques for cerebral aneurysms and an increasing tendency to operate early, that is within the first 2–3 days after rupture, have removed the need for induced hypotension in most cases [1]. Attention has now focused on the need to maintain cerebral perfusion pressure (CPP) in order to improve cerebral perfusion and reduce cerebral ischaemia. Unless obvious from the preoperative nursing charts that a neurological deficit develops when mean arterial pressure (MAP) is less than a certain value, it is difficult to determine if
there is a critical MAP below which cerebral hypoperfusion, which may lead to cerebral ischaemia, occurs. The aim of this study was to determine if it is possible to identify this critical MAP in individual patients during anaesthesia and surgery by continuously monitoring jugular bulb oxygen saturation (SjO2 ) .
Patients and methods The study was approved by the local Ethics Committee and patients or their relatives gave informed consent. Twenty-six patients requiring clipping of a cerebral aneurysm, 20 following rupture and six unruptured, were anaesthetized with propofol, alfentanil and atracurium infusions and their lungs ventilated mechanically to hypocapnia (3.5–4.5 kPa). The inspired oxygen concentration was adjusted to maintain PaO at 13–18 kPa. All patients received mannitol 0.5 g kg91 followed by frusemide 0.5 mg kg91 to reduce brain bulk and minimize retraction pressures on the brain. In addition to standard monitoring, including a subclavian central venous pressure (CVP) catheter and radial artery cannula, an Oximetric fibreoptic oximetry catheter was inserted into the jugular bulb on the side showing predominant venous drainage on the cerebral angiogram. If venous phase cerebral angiography was not available, the catheter was placed in the right jugular bulb. MAP, CVP and SjO2 were recorded continuously throughout the procedure. Normovolaemia was maintained by observing the responses of MAP and CVP to fluid administration, and colloids were given as required to maintain a normal CVP. A haemoglobin concentration of greater than 10 g dl91 was maintained by giving blood if necessary. The patient’s temperature was maintained constant throughout the procedure using a warming blanket (Bair Hugger). Jugular venous and arterial blood samples were obtained for baseline readings and calibration of the Oximetric catheter, immediately after placement of the jugular bulb catheter and at appropriate intervals after changes in MAP. The samples were analysed for saturation and haemoglobin concentration (Corning co-oximeter), lactate (YSI 2300 stat), 2
E. MOSS, MD, FRCA, N. M. DEARDEN, MB, CHB, FRCA, J. C. BERRIDGE, MB, CHB, FRCA, Department of Anaesthetics, Leeds General Infirmary, Great George Street, Leeds LS1 3EX. Accepted for publication: April 25, 1995. This article is accompanied by Editorial II.
528
British Journal of Anaesthesia Table 1 SjO2 values before and after an increase in mean arterial pressure (MAP), critical MAP values (NI : not identified), maximum values of LOI and outcome in patients undergoing clipping of cerebral aneurysms (ND : no deficit, FD : focal neurological deficit, DC : depressed conscious level; G : good, MD : moderate, SD : severe disability, D : death). * Still hypoperfused, † cardiorespiratory failure, ‡ PaCO2 remained constant during measurements. E : elective (i.e. aneurysm had not ruptured)
Patient no.
Days after rupture
1‡ 2 3 4‡ 5 6 7 8 9‡ 10 11‡ 12 13 14 15 16‡ 17 18‡ 19 20‡ 21 22‡ 23 24 25 26‡
5 E 5 12 5 28 6 E 2 5 4 6 10 2 2 9 E 7 4 E E E 2 2 2 2
Initial SjO2
SjO2 after increase in MAP
Critical MAP
Maximum LOI
Initial
Long-term
54.8 91.3 59 62 55 49.1 47 56.3 53.4 58 36 71.5 50.6 52 74.2 42.1 87.4 65 88.4 55.9 40.7 65.9 84.9 40.5 54.4 52.7
69.4 82.1 66 82 74 62.7 59 65.3 58.7 86 63 — 69 — 79.4 48.5 83.8 76 89.8 62.7 56.7 64.4 69.3 61.6 60.4 72.5
NI NI NI NI NI 90 110 NI 90 NI 80 NI 80 110* NI 100 NI NI NI NI 100 NI NI 95–100 NI 85
0.043 0.226 0.15 0.222 0.115 0.095 0.073 0.075 0.003 0.169 0.218 0.027 0.081 0.007 0.020 0.024 0.251 0.03 0.195 0.004 0.002 0.028 0.044 0.014 0.055 0.079
ND FD FD FD/DC FD FD ND ND FD FD FD/DC ND ND FD/DC FD ND ND FD/DC FD/DC ND ND ND ND FD/DC ND ND
G SD D SD MD MD G G SD SD MD D† G D MD G G D SD G G G G MD G G
Outcome
blood-gas tension and potassium concentration (Corning). The lactate oxygen index (LOI) [2] was calculated using the following formula: LOI =
− AJDL(mmol litre −1 ) × 2.24 AJDO2 (ml dl −1 )
where AJDL : arterial to jugular difference for lactate and AJDO2 : arterial to jugular difference in oxygen content. AJDO2 = (SaO2 % - SjO2 %) × Hb ×1.34/100 + (PaO2 (kPa) - PjO2 (kPa))× 0.0225 The normal value for LOI is :0.03 and an LOI of 90.08 was taken to indicate significant cerebral ischaemia. MAP was manipulated deliberately only at the request of the surgeon or if the clinical condition or blood test results required. MAP was increased deliberately if it was, or decreased below, a level at which SjO2 was less than 54 %. I.v. fluids were used to increase MAP in the first instance. Dopamine or adrenaline were used only if fluid therapy was ineffective. A critical MAP was defined as MAP below which SjO2 was :54 %. This level of SjO2 indicates cerebral hypoperfusion [3]. The jugular bulb catheter was removed before the patient was returned to the neurosurgical ward. Initial outcome was judged from the Glasgow coma score and neurological observations in the first 24 h after surgery. Long-term outcome was determined from the discharge summary or, for patients with poorer
outcomes, follow-up at 1 month using the Glasgow outcome score [4]. Changes in SjO2 in response to increasing MAP were analysed using a paired t test. Outcomes were compared using chi-square with Yates’ correction factor.
Results MAP required elevation in 24 of 26 patients and SjO2 measurements indicated a critical MAP of between 80 and 1 10 mm Hg in nine patients (table 1). However, in only one of six patients undergoing elective aneurysm clipping was this a feature. Another patient (patient No. 14) had a persistently low SjO2 despite an MAP of 110 mm Hg. Increasing MAP was associated with an increase in SjO2 in 19 patients, no change in two and a reduction in three, two of whom had a simultaneous reduction in P aco2. The increase in SjO2 with increasing MAP was statistically significant (P : 0.001). However, in more than 50 % of these patients there was also an increase in P aco2 between SjO2 samples. If those patients with a hyperaemic cerebral circulation ( SjO2 975 %) and those with an increase in P aco2 between samples were excluded, the increase in SjO2 with increasing MAP was still significant (P 0.004) (n 9). There were three patients (Nos 2, 17 and 23) with SjO2 values 975 % in whom an increase in MAP caused an apparent reduction in CBF. In patient Nos 17 and 23, this was associated with a reduction in P aco2 but in patient No. 2 there was an increase in P aco2.
SjO2 during cerebral aneurysm surgery
529 ischaemia became worse in six and improved in four after an increase in MAP.
Discussion
Figure 1 Changes in SjO2 with increases in mean arterial pressure (MAP) in nine patients with a constant PaCO2 throughout the period of measurement. Each line represents one patient. The two patients with absent autoregulation show lines with almost vertical slopes. Table 2 Relationship between lactate oxygen index (LOI) and initial and long-term outcome. *P : 0.05 Normal LOI LOI 9 0.08 Initial outcome Good Poor Long-term outcome Good/moderate disability Severe disability/ vegetative/death
12 4
2* 8*
12
4
4
6
Table 3 Changes in lactate oxygen index (LOI) with increasing mean arterial pressure LOI decreased LOI increased LOI more ischaemic LOI less ischaemic Never ischaemic
11 13 6 4 14
There were eight patients in whom P aco2 remained constant and there were SjO2 measurements at three different MAP values. Four showed a threshold level of CPP above which SjO2 showed no further increase with increasing MAP. Three of the other four patients had SjO2 values 975 % and the other failed to demonstrate a critical threshold for CPP. In those patients with a constant P aco2 during the period of study, by observing the response of SjO2 to a change in MAP, it was possible to demonstrate absent autoregulation in two patients and impaired autoregulation in six others (fig. 1). Patients with an LOI 90.08 at any time during the procedure (n 10) had a worse initial outcome (P : 0.02) than patients who had a normal LOI throughout (n 16), but the long-term outcome was similar to those with a normal LOI (0.5 9 P 9 0.1) (table 2). Of patients with an LOI showing cerebral ischaemia, three patients had SjO2 values :54 %, four 54–75 % and three 975 %. Increasing MAP did not have a consistent effect on LOI (table 3). In patients with an LOI 90.08,
When the cerebral metabolic rate for oxygen remains constant, SjO2 measurements reflect CBF. In this study the depth of anaesthesia was maintained at a constant level by continuous infusion of propofol and alfentanil, and therefore the cerebral metabolic rate would have remained fairly constant. Measurement of SjO2 also indicates if CBF is sufficient to satisfy the oxygen demands of the tissues. SjO2 reflects the balance between oxygen supply and demand with values of less than 54 % indicating cerebral hypoperfusion and readings less than 40 % being associated with global cerebral ischaemia [3, 5, 6], although significant regional ischaemia may be present with a normal SjO2 , as evidenced by an increase in LOI. The results of this study showed that approximately 50 % of patients undergoing aneurysms clipped acutely after subarachnoid haemorrhage had a critical MAP below which there was cerebral hypoperfusion. In contrast, a critical MAP was identified in only one of six patients during clipping of aneurysms which had not previously ruptured. von Helden and colleagues [7] found that a similar percentage of patients in a coma after subarachnoid haemorrhage had periods when SjO2 was critical (50–55 %). LOI gave some indication of patients who were likely to develop neurological deficits in the first 24 h after surgery but had no predictive value for long-term outcome. This is not surprising because patients with deficits were treated aggressively with hypertensive, hypervolaemic and haemodilution therapy which reversed the ischaemia in many patients. Unfortunately, because this study was performed in patients undergoing surgery, it was not always possible to maintain P aco2 constant between measurements of SjO2. However, in those patients in whom P aco2 remained constant, SjO2 still increased when MAP was increased, confirming an association between MAP and SjO2. One of the most effective ways of increasing SjO2 is to increase P aco2. Hypocapnia is usually recommended during intracranial surgery to reduce brain bulk and retraction pressures. However, with the use of mannitol and frusemide to reduce brain bulk, this hypocapnia may not be necessary. This was not demonstrated in this study. The increase in SjO2 associated with increasing MAP could be the result of defective autoregulation after subarachnoid haemorrhage [8] and is similar to the minimal or absent autoregulatory response demonstrated in head injuries by Fortune and colleagues [9]. An alternative explanation is that the plateau of the autoregulation graph is incorrect. It is known that CBF increases slightly as arterial pressure increases within the range of autoregulation. It has been suggested that increasing cardiac output increases CBF, but Bouma and colleagues [10] have shown that CBF is not related to cardiac output even when autoregulation is impaired. Thus for a constant MAP, any increase in CBF associated with intra-
530 vascular expansion appears to be mediated by a decrease in blood viscosity rather than an increase in cardiac output. In two patients (fig. 1) cerebrovascular autoregulation was clearly absent and in several others it appeared to be impaired. There was some evidence of a critical threshold for CPP, similar to that reported in patients with head injury [11], above which SjO2 did not increase further as pressure increased. In this study, the decision as to which jugular bulb to cannulate was made on the basis of cerebral angiography if venous phase films had been taken. If these films were not available the right jugular bulb was cannulated because the majority of cortical blood flow drains down the right internal jugular vein in most patients [12]. It is possible that some changes in SjO2 and LOI were missed because blood from hypoperfused or ischaemic areas was draining down the side opposite to the jugular catheter. Therefore, it could be argued that the jugular bulb on the side of the lesion [2, 13] or both jugular bulbs should be catheterized [14] because cerebral AJDO2 and LOI can differ between the two jugular bulbs. The four patients with a normal LOI who had a poor initial outcome may have been examples of patients in whom the wrong jugular bulb had been cannulated so that ischaemia was not detected. Alternatively, there may have been cerebral infarction so that no lactate was produced. All had developed neurological deficits before the jugular catheter was removed. There were two patients with an LOI 90.08, indicating cerebral ischaemia, who had a good initial outcome. These patients had a period of incomplete ischaemia which was probably caused by inadequate cerebral perfusion pressure or temporary arterial occlusion. Clearly the problem had been corrected before the patient recovered from the anaesthetic. It is now accepted that CPP should usually be maintained at preoperative levels during cerebral aneurysm surgery [1]. This helps to maintain flow through vessels affected by vasospasm and optimizes collateral flow. Induced hypotension does not alter outcome in patients undergoing early surgery for cerebral artery aneurysms compared with normotension, and the incidence of intraoperative rerupture is not reduced [15]. In fact, the outcome is worse when hypotension is used during intraoperative rupture of an aneurysm than when tamponade and temporary clipping with maintenance of MAP are used to control haemorrhage [16]. As the main objective of induced hypotension in these procedures is to reduce the incidence of intraoperative rupture of the aneurysm, it would not seem appropriate to use it in most patients. The present results support maintenance of CPP during surgery on cerebral aneurysms, particularly when surgery is performed within a few days of rupture, and show that measurement of SjO2 allows determination of the minimum MAP that should be allowed during
British Journal of Anaesthesia surgery. There is a risk of rupture if MAP is increased before the aneurysm is clipped, but the use of SjO2 monitoring and measurement of LOI allows determination of an MAP which gives the best balance between the risks of cerebral ischaemia and aneurysm rupture in individual patients.
References 1. Archer DP, Shaw DA, Leblanc RL, Tranmer BI. Haemodynamic considerations in the management of patients with subarachnoid haemorrhage. Canadian Journal of Anaesthesia 1991; 38: 454–470. 2. Robertson CS, Narayan RK, Gokaslan ZL, Pahwa R, Grossman RG, Caram P, Allen E. Cerebral arteriovenous oxygen difference as an estimate of cerebral blood flow in comatose patients. Journal of Neurosurgery 1989; 70: 222–230. 3. Dearden NM. SjO2 and critical perfusion pressure after severe brain injury. British Journal of Intensive Care 1992 (Suppl. 1): 7–11. 4. Jennett B, Bond M. Assessment of outcome after severe brain damage. A practical scale. Lancet 1975; 1: 480–484. 5. Sheinberg M, Kanter MJ, Robertson CS, Contant CF, Narayan RK, Grossman RG. Continuous monitoring of jugular venous oxygen saturation in head injured patients. Journal of Neurosurgery 1992; 76: 212–217. 6. Gopinath SP, Robertson CS, Contant CF, Hayes C, Feldman Z, Narayan RK, Grossman RG. Jugular venous desaturation and outcome after head injury. Journal of Neurology, Neurosurgery and Psychiatry 1994; 57: 717–723. 7. von Helden A, Schneider GH, Unterberg A, Lanksch WR. Monitoring of jugular venous oxygen saturation in comatose patients with subarachnoid haemorrhage and intracranial haematomas. Acta Neurochirurgica Supplementum 1993; 59: 102–106. 8. Pickard JD, Boisvert DPJ, Graham DI, Fitch W. Late effects of subarachnoid haemorrhage on the response of the primate cerebral circulation to drug-induced changes in arterial blood pressure. Journal of Neurology, Neurosurgery and Psychiatry 1979; 42: 899–903. 9. Fortune JB, Feustel PJ, Weigle CGM, Popp AJ. Continuous measurement of jugular venous oxygen saturation in response to transient elevations of blood pressure in head-injured patients. Journal of Neurosurgery 1994; 80: 461–468. 10. Bouma GJ, Muizelaar PJ, Bandoh K, Marmarou A. Blood pressure and intracranial pressure–volume dynamics in severe head injury : relationships with cerebral blood flow. Journal of Neurosurgery 1992; 77: 15–19. 11. Chan K-H, Miller JD, Dearden NM, Andrews PJD, Midgley S. The effect of changes in cerebral perfusion pressure upon middle cerebral artery blood flow velocity and jugular bulb venous oxygen saturation after severe brain injury. Journal of Neurosurgery 1992; 77: 55–61. 12. Williams PL, Warwick R, Dyson M, Bannister LH. Gray’s Anatomy, 37th Edn. Edinburgh, London, Melbourne, New York: Churchill Livingstone, 1989; 793–805. 13. Robertson CS, Grossman RG, Goodman JC, Narayan RK. The predictive value of cerebral anaerobic metabolism with cerebral infarction after head injury. Journal of Neurosurgery 1987; 67: 361–368. 14. Chieregato A, Targa L, Mantovani G, Droghetti L, Zatelli R. Cerebral arteriovenous difference and lactate–oxygen index: a case of bilateral monitoring. Journal of Neurosurgical Anesthesiology 1994; 6: 43–47. 15. Innomata S, Mizuyama K, Sato S, Takahashi H, Saito S, Mizutani T, Yamaguchi H, Fukuda T, Naito H. The effect of deliberate hypotensive anaesthesia on the prognosis of patients who underwent early surgeries for ruptured cerebral aneurysm. Masui 1992; 41: 207–213. 16. Gianotta SL, Oppenheimer JH, Levy ML, Zelman V. Management of intraoperative rupture of aneurysm without hypotension. Neurosurgery 1991; 28: 531–535.