Acta Neurol Scand DOI: 10.1111/ane.12258
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd ACTA NEUROLOGICA SCANDINAVICA
Bedside diagnosis of mitochondrial dysfunction in aneurysmal subarachnoid hemorrhage Jacobsen A, Nielsen TH, Nilsson O, Schalen W, Nordstr€ om CH. Bedside diagnosis of mitochondrial dysfunction in aneurysmal subarachnoid hemorrhage. Acta Neurol Scand: DOI: 10.1111/ane.12258. © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd. Objectives – Aneurysmal subarachnoid hemorrhage (SAH) is frequently associated with delayed neurological deterioration (DND). Several studies have shown that DND is not always related to vasospasm and ischemia. Experimental and clinical studies have recently documented that it is possible to diagnose and separate cerebral ischemia and mitochondrial dysfunction bedside. The study explores whether cerebral biochemical variables in SAH patients most frequently exhibit a pattern indicating ischemia or mitochondrial dysfunction. Methods – In 55 patients with severe SAH, intracerebral microdialysis was performed during neurocritical care with bedside analysis and display of glucose, pyruvate, lactate, glutamate, and glycerol. The biochemical patterns observed were compared to those previously described in animal studies of induced mitochondrial dysfunction as well as the pattern obtained in patients with recirculated cerebral infarcts. Results – In 29 patients, the biochemical pattern indicated mitochondrial dysfunction while 10 patients showed a pattern of cerebral ischemia, six of which also exhibited periods of mitochondrial dysfunction. Mitochondrial dysfunction was observed during 5162 h. An ischemic pattern was obtained during 688 h. Four of the patients (40%) with biochemical signs of ischemia died at the neurosurgical department as compared with three patients (10%) in the group of mitochondrial dysfunction. Conclusions – The study documents that mitochondrial dysfunction is a common cause of disturbed cerebral energy metabolism in patients with SAH. Mitochondrial dysfunction may increase tissue sensitivity to secondary adverse events such as vasospasm and decreased cerebral blood flow. The separation of ischemia and mitochondrial dysfunction bedside by utilizing microdialysis offers a possibility to evaluate new therapies.
Introduction
Aneurysmal subarachnoid hemorrhage (SAH) is frequently associated with delayed neurological deterioration (DND) due to delayed cerebral ischemia (DCI; 1, 2). As ischemia is associated with well-defined changes in energy metabolism, it might be expected that DCI would be detected early by intracerebral microdialysis provided the catheters are positioned in the areas of interest (3). However, several studies have shown that
A. Jacobsen1, T. H. Nielsen1, O. Nilsson2, W. Schalen2, C. H. Nordstr€om1 1 Department of Neurosurgery, Odense University Hospital, Odense, Denmark; 2Department of Neurosurgery, Lund University Hospital, Lund, Sweden
Key words: cerebral energy metabolism; microdialysis; mitochondrial dysfunction; ischemia; lactate; pyruvate; subarachnoid hemorrhage Carl-Henrik Nordstr€om Persikev€agen 77, SE 22355 Lund, Sweden Tel.: +46 70 5655785, +46 46 146974 e-mail: [email protected] Accepted for publication April 3, 2014
DND is not always related to vasospasm and ischemia which might explain why therapies directed toward DCI may be ineffective (4, 5). It has repeatedly been reported that ‘metabolic crisis’ may occur although brain tissue oxygenation (PbtO2) appears to be adequate (6): In one study, PbtO2 was normal in approximately half of the episodes of compromised cerebral energy metabolism (7). Accordingly it is of clinical interest to identify the mechanisms underlying the observed ‘metabolic crisis.’ 1
Jacobsen et al. In a series of experimental studies, we have characterized the changes in variables related to cerebral energy metabolism after induced mitochondrial dysfunction (8–10). Briefly, mitochondrial dysfunction is biochemically characterized by increased lactate/pyruvate (LP) ratio due to a marked increase in cerebral lactate at a normal or elevated pyruvate level. This pattern is different from that obtained during cerebral ischemia which is characterized by a marked increase in LP ratio at a pronounced decrease in intracerebral pyruvate level. We have recently identified a similar pattern of mitochondrial dysfunction in patients with recirculated large infarcts after prolonged cerebral ischemia (11). In this study of non-traumatic SAH, we have monitored variables related to cerebral energy metabolism utilizing routine microdialysis with bedside biochemical analysis. The study aims at exploring whether in these patients, the biochemical ‘metabolic crisis’ usually reflects DCI or mitochondrial dysfunction. Materials and methods Patients
The study includes all 55 SAH patients monitored with intracerebral microdialysis admitted to the Department of Neurosurgery, Lund University Hospital, Sweden, between January 2001 and August 2009. During the period, altogether 1240 patients with spontaneous SAH were admitted to the department. Patients with ruptured saccular aneurysms (n = 792) were treated either with endovascular (n = 642; 81%) or open surgical (n = 150; 19%) occlusion of the aneurysm. Patients included in the study were selected for intracerebral microdialysis due to the severity of the initial clinical picture including evaluation of the neurological state, the amount of subarachnoidal blood, and the presence of a focal hematoma. Accordingly, a high proportion of the patients (44%) were treated with surgical evacuation of large intracerebral hematomas. All patients were treated according to the routine procedures at the department (12, 13). The present report is a retrospective interpretation of the data obtained at the bedside initiated by the recently described possibility to diagnose mitochondrial dysfunction and separate it from DCI (8–11). Microdialysis technique
Microdialysis was performed utilizing CMA 70 catheters (CMA Microdialysis, Stockholm, Swe2
den). The microdialysis catheters were perfused (Perfusion Fluid; CMA Microdialysis) at a rate of 0.3 ll/min, and the perfusates were collected in capped microvials at a 1-h interval (14). The samples were immediately analyzed utilizing conventional enzymatic techniques (CMA 600 or ISCUS Microdialysis Analyzer; CMA Microdialysis), and the results were displayed on a bedside monitor. The perfusates were analyzed for glucose, pyruvate, lactate, glutamate, and glycerol. All data obtained from cerebral microdialysis were integrated with global biochemical and physiological data utilizing a specially developed computer program (ICUpilot; CMA Microdialysis). The study includes altogether 12,171 microdialysis samples and approximately 60,000 bedside biochemical analyses. All microdialysis catheters were inserted in the operation theater. In patients subjected to open surgery, the microdialysis catheters were inserted after clipping of the aneurysm and evacuation of intracerebral hematomas. In these cases, one or multiple catheters were placed in the exposed cerebral tissue. Fig. 1 shows a patient with two microdialysis catheters placed in the right frontal lobe after clipping of an aneurysm of the anterior communicating artery and partial evacuation of a large intracerebral hematoma in the left frontal lobe. In patients not subjected to open surgery, the microdialysis catheter was inserted when there was reason to insert a ventricular catheter for monitoring of the intracranial pressure (ICP) and/or drainage of intraventricular fluid. In these cases, a separate burr hole adjacent to that used for insertion of the intraventricular catheter was used and the microdialysis catheter was positioned in the frontal lobe. Altogether 78 microdialysis catheters were inserted: 33 patients received one catheter, 21 patients two catheters and one patient three catheters. The Ethical Committee of Lund University Medical Faculty approved the use of multiple intracerebral microdialysis catheters in the present study. Classification of ischemia and mitochondrial dysfunction
The biochemical classification of the patients was based on the reference values obtained in the non-sedated normal human brain (14). For the LP ratio, the upper normal level was set at 30 (normal mean + 2SD) and the lower normal level for pyruvate was set at 70 (normal mean 2SD). As a LP ratio of 40 has been used as an upper level in previous publications from other centers (7), our data are also presented with LP ratio 40 as an upper cutoff level. In accordance with our
Mitochondrial dysfunction in SAH
Figure 1. CT scanning after open surgery with clips occlusion (C) of an aneurysm of the anterior communicating artery and partial evacuation of bifrontal intracerebral hematomas. Two intracerebral microdialysis catheters (A and B) were inserted into the basal right frontal lobe. The tip of the ventricular catheter (D) is visualized.
previous experimental and clinical studies, patients with a LP ratio >30 and a pyruvate level 30 and >40, respectively) and pyruvate >70 lM were classified as mitochondrial dysfunction (8, 9, 11). Data from the patients were included into the groups when the criteria defined above were fulfilled during three subsequent measurements (3 h) or more. Statistics
All values are expressed as mean and standard deviation (SD) or median (interquartile rage) as appropriate. Statistical comparison of the mean levels of microdialysis variables between the group defined as ischemia (n = 12), and the group defined as mitochondrial dysfunction with cutoff levels LP ratio >30 (n = 37) and LP ratio >40 (n = 24), respectively, was performed by utilizing the non-parametric Wilcoxon rank-sum test. A P-value below 0.05 was considered significant. As the P-values are to be considered exploratory, no adjustment for generating multiple P-values was performed. STATA 11.1 statistical software package (StataCorp LP, College Station, TX, USA) was used for data analysis. Results Patient characteristics
Basic clinical and radiographic characteristics are given in Table 1. Mortality refers to the number of deaths at the department of neurosurgery. At the time of admission, the following neurological
state was noted according to Glasgow Coma Score (GCS; 15): GCS 15 16%; GCS 9–14 36%, GCS ≤8 47%. All patients were classified as Fisher grade 3 or 4. In six patients, no aneurysm treatment was performed. In three of these patients, no aneurysm was found on angiography and none of these patients developed biochemical signs of mitochondrial dysfunction or ischemia. In three patients, no angiography was performed as—due to the poor condition of the patients— surgical evacuation of a large hematoma and clipping of an middle cerebral artery (MCA) aneurysm was performed immediately after CTscanning. Variables during ischemia and mitochondrial dysfunction
The patients were classified into the two groups of cerebral ischemia and mitochondrial dysfunction, respectively, according to the principles defined previously. The two conditions may occur in the same patient at different points of time (see below). This was observed in altogether six patients. In Table 1, these six patients are included in the group of mitochondrial dysfunction (n = 29) as well as in the group of ischemia (n = 10). In 22 patients, the biochemical parameters did not fulfill the criteria for inclusion into the two groups. The clinical data for these patients are given in Table 1. Table 1 gives mortality for the three groups during treatment in the neurocritical care ward. One of the patients in the group of mitochondrial dysfunction died after having developed a biochemical pattern of ischemia. In the table, this patient was included into both groups. The 3
Jacobsen et al. Table 1 Characteristics for patients included in the study
Age years, mean (SD) Sex M/F, n Mortality, n (%) Aneurysm location, n (%) ACom A, Pericallosal A Carotid A, PostCom A MCA Posterior Circ No aneurysm found Angiography not performed Treatment, n (%) Endovascular Open surgical clipping No aneurysm treatment Evacuation of hematoma
Total n = 55
Mitochondrial dysfunction n = 29
Ischemia n = 10
No ischemia/mitoch. dysf. n = 22
55 (12) 15/40 8 (15)
57 (10) 8/21 3 (10)
53 (8) 2/8 4 (40)
51 (14) 7/15 2 (9)
15 8 24 2 3 3
(27) (15) (44) (4) (5) (5)
6 3 17 1 0 1
(21) (10) (81) (5)
21 28 6 24
(38) (51) (11) (44)
10 18 1 15
(10) (20) (50)
(5)
1 2 5 0 0 2
(36) (14) (32) (5) (14)
(20)
8 3 7 1 3 0
(34) (62) (3) (52)
3 5 2 7
(30) (50) (20) (70)
9 10 3 7
(41) (45) (14) (32)
n refers to the number of patients within each category. Six patients are included in the group of mitochondrial dysfunction as well as in the group of ischemia (see text). Ischemia is defined as lactate/pyruvate (LP) ratio >30 at a pyruvate level 30 at a normal or increased pyruvate level. AComA, anterior communicating artery; Pericallosal A, pericallosal artery; Carotid A, carotid artery; PostComA, posterior communicating artery; MCA, middle cerebral artery; Posterior Circ, posterior circulation.
pattern of cerebral ischemia was observed during altogether 688 h, the pattern of mitochondrial dysfunction with cutoff level LP ratio 30 was observed during 5162 h, and the pattern of mitochondrial dysfunction with cutoff level LP ratio 40 was observed during 2920 h. Table 2 gives median values (interquartile range) for all biochemical variables displayed beside. Data are given for all microdialysis catheters (n = 78), separately for those classified as cerebral ischemia (n = 12 catheters) and mitochondrial dysfunction with cutoff levels LP ratio >30 (n = 37 catheters) and LP ratio >40 (n = 24 catheters). As described above, the data were included into the diagnostic groups when they fulfilled the criteria for inclusion during three consequent measurements (3 h) or more. The table also gives corre-
sponding data for patients who neither exhibited a pattern of ischemia nor a pattern of mitochondrial dysfunction (n = 29). In addition, reference levels from normal human brain are given as mean (SD) (14). In the table, statistical comparisons between the two groups defined as mitochondrial dysfunction and the group defined as ischemia are included. According to the definitions of the present study, the pyruvate level was higher in patients in the mitochondrial dysfunction group and the LP ratio was significantly lower than in the ischemic group. The levels of glucose were significantly higher in the two groups of mitochondrial dysfunction. The levels of lactate were higher in the two groups of mitochondrial dysfunction, but the difference did not quite reach statistical significance (P = 0.156
Table 2 Median value (interquartile range) for biochemical variables in patients with subarachnoid hemorrhage and biochemical indications of cerebral ischemia, mitochondrial dysfunction, and no ischemia/mitochondrial dysfunction, respectively
Ischemia (n = 12) Mitoch. dysf. LP ratio >30 (n = 37) Mitoch. dysf. LP ratio >40 (n = 24) No ischemia/mitoch. dysf. (n = 29) Reference (normal) level
Glycerol (lM)
LP ratio
Glucose (mM)
Lactate (mM)
Pyruvate (lM)
Glutamate (lM)
166 (81–502)
0.3 (0.1–0.7)
4.7 (2.7–7.4)
34 (15–51)
96 (19–432)
536 (392–750)
41 P 53 P 21
1.3 (0.6–1.9) P = 0.0014 1.2 (0.5–1.9) P = 0.008 1.4 (0.9–2.0)
7.6 (6.1–9.4) P = 0.156 8.5 (6.8–11.0) P = 0.06 2.4 (1.9–3.5)
178 (133–232)
115 (89–156)
22 (7–72) P = 0.020 42 (7–97) P = 0.08 4 (3–6)
89 P 98 P 78
1.7 (0.9)
2.9 (0.9)
166 (47)
16 (16)
82 (44)
(34–53) < 0.001 (45–68) < 0.001 (18–25)
23 (4)
152 (114–203)
(51–188) < 0.001 (56–198) < 0.001 (44–152)
Data for mitochondrial dysfunction are given for two cutoff levels: lactate/pyruvate (LP) ratio >30 and >40, respectively. n refers to the number of microdialysis catheters within each category. Reference (normal) levels are given as mean values (SD) in accordance with the original data obtained from Reinstrup et al. (14). Statistical comparison performed between ischemia and mitochondrial dysfunction with LP cutoff levels >30 and >40, respectively. As the pyruvate level was used for the classification into the groups of ischemia and non-ischemic mitochondrial dysfunction, no statistical comparison was performed regarding this variable.
4
Mitochondrial dysfunction in SAH and P = 0.06 for LP ratio cutoff at 30 and 40, respectively). Glutamate was significantly lower in the group of mitochondrial dysfunction with a LP ratio cutoff level of 30 but not 40, while glycerol was significantly higher in the ischemic group compared to both groups of mitochondrial dysfunction. As discussed above, the patterns of ischemia and mitochondrial dysfunction were in six cases noted at different points of time in the same patient. We present three representative patterns: (i) a pattern typical of progressive cerebral ischemia; (ii) transfer of cerebral ischemia into a pattern of mitochondrial dysfunction; and (iii) change in mitochondrial dysfunction into cerebral ischemia. Biochemical pattern of cerebral ischemia
Fig. 2 illustrates the biochemical pattern of progressive cerebral ischemia. This 64-year-old woman was admitted in deep coma with a large hematoma in the left Sylvian fissure. She was taken to surgery without preceding angiography. A giant aneurysm of the MCA was occluded, and the hematoma was evacuated. One microdialysis catheter was placed in the basal frontal lobe, and one was placed in the tip of the temporal lobe. Fig. 2 shows the data from the catheter in the frontal lobe. A few hours after surgery, the LP ratio started to increase and intracerebral pyruvate decreased to 30–40 lM. Intracerebral glucose simultaneously decreased to a subnormal level. During the time period in Fig. 2, lactate remained very high (11.0 1.2 mmol). The illustrated increase in LP ratio preceded an increase in ICP (to 25–30 mmHg) by approximately 4 h.
Figure 2. Biochemical pattern of progressive cerebral ischemia. The lactate/pyruvate (LP) ratio increases to a very high level as pyruvate and glucose decrease to below normal. The ranges of the variables (mean SD) for normal human brain (14) are indicated on the respective y-axis.
Transfer from ischemia to mitochondrial dysfunction
Fig. 3 illustrates the transfer from cerebral ischemia to mitochondrial dysfunction. This 54-yearold man was operated with occlusion of an aneurysm of the left MCA and evacuation of a large hematoma. The microdialysis catheter was positioned in the frontal lobe. After surgery, the LP ratio was initially very high, while pyruvate and glucose were below normal limit. During the following 48 h, the LP ratio decreased to 50–60 due to a continuous increase in pyruvate to supranormal levels while the lactate level remained at a very high level (12.2 1.6 mM). As shown in the figure, the initially low level of glucose normalized within the initial 12 h. During the period illustrated in Fig. 3, ICP was below 10 mmHg and cerebral perfusion pressure was above 70 mmHg. Transfer of mitochondrial dysfunction to cerebral ischemia
Fig. 4 illustrates the transfer from mitochondrial dysfunction to a pattern typical of cerebral ischemia. This 58-year-old man was operated with occlusion of an aneurysm of the right MCA and evacuation of an intracerebral hematoma. During surgery, the MCA was temporarily occluded for 7 + 4 min. The microdialysis catheter was placed in the frontal lobe close to the Sylvian fissure. The figure illustrates the biochemical pattern 122 h after surgery: The LP ratio was 60, while
Figure 3. Biochemical pattern of transfer from ischemia to non-ischemic mitochondrial dysfunction. During the initial 4–6 h, the pattern indicates ischemia: The lactate/pyruvate (LP) ratio is very high, and pyruvate and glucose are very low. The subsequent increase in glucose is paralleled by an increase in pyruvate. As the interstitial lactate level remained very high (not shown in the figure), the LP ratio remained elevated although interstitial pyruvate increased above normal. The ranges of the variables (mean SD) for normal human brain (14) are indicated on the respective y-axis.
5
Jacobsen et al.
Figure 4. Biochemical pattern of transfer from non-ischemic mitochondrial dysfunction to ischemia. About 120–130 h after surgery lactate/pyruvate (LP), ratio remains elevated (60–70) although pyruvate level is within normal range. The late decrease in pyruvate causes a marked increase in LP ratio. The simultaneous increase in interstitial glycerol level indicates secondary cellular degradation. The ranges of the variables (mean SD) for normal human brain (14) are indicated on the respective y-axis.
pyruvate was within normal limits. Later than 140 h after surgery, the LP ratio increased due to a marked decrease in pyruvate level to below normal. The increase in LP ratio was associated with a marked increase in intracerebral glycerol concentration, while lactate remained stable at a high level (8.6 0.8 mM). Discussion
In this study, we utilize the recently described possibility to diagnose and separate cerebral ischemia and mitochondrial dysfunction to estimate the relative frequency of the two conditions in a group of patients with severe SAH (8, 9, 11). The interpretations are based on the biochemical patterns displayed at the bedside during routine intracerebral microdialysis. It has previously been noted that a pattern of cerebral ‘metabolic crisis’ obtained during microdialysis is not always associated with reduction in cerebral blood flow or tissue oxygenation (7, 16, 17). The interpretations of the biochemical data in the present report are based on experimental studies obtained after induced mitochondrial dysfunction (8–10, 18). This biochemical pattern and its association with mitochondrial dysfunction in vitro have also been described after transient experimental cerebral ischemia (19–21). The biochemical interpretations are based on the hypothesis that cerebral interstitial LP ratio reflects cytoplasmic redox state. Lactate and pyruvate pass cell membranes by proton-linked 6
monocarboxylate transporters (MCTs; 22). MCTs are found in various tissues, and three isoforms have been described in the rodent brain. The efficacy of the MTCs in transmembrane transport of lactate and pyruvate appears to be high, and it has been documented that cerebral ischemia causes an instantaneous increase in interstitial LP ratio that reflects the intracellular shift in redox state (23). As the capacity of the monocarboxylate transporters across cell membranes is high, the LP ratio obtained from microdialysis of interstitial fluid can be used to evaluate cytoplasmic redox state (24). In cerebral ischemia, tissue supply of oxygen and glucose is compromised and the LP ratio momentarily increases to a very high level due to the increase in lactate and the marked decrease in pyruvate (23–25). In experimentally induced mitochondrial dysfunction, the increase in LP ratio occurs at a normal or increased pyruvate level (8–10). This pattern is also observed after transient experimental brain ischemia (19, 20) and following recirculation of large cerebral infarcts in humans (11). In the present group of patients, mitochondrial dysfunction was observed more frequently than a pattern of cerebral ischemia (Table 1). The fact that cerebral ‘metabolic crisis’ may occur in the absence of reduction in cerebral blood flow and tissue oxygenation has previously been observed in SAH patients as well as in brain trauma (7, 16, 17). The present analysis offers an explanation of this phenomenon and indicates that the ‘metabolic crisis’ described in previous studies may reflect mitochondrial dysfunction. The mitochondrial dysfunction observed in these clinical conditions may have been caused by a previous period of transient ischemia (11, 19–21). This situation is illustrated in Fig. 3 and shows that an initial period of cerebral ischemia was succeeded by a lasting increase in LP ratio while the levels of pyruvate and glucose increased to normal or supra-normal levels. Mitochondrial dysfunction may deteriorate into a second period of ischemia as illustrated in Fig. 4. In this patient, a biochemical pattern of cerebral ischemia was obtained 140 h after the initial insult in an area previously exhibiting a pattern of mitochondrial dysfunction. The ischemic development was associated with a pronounced increase in glycerol indicating degradation of cellular membranes (26, 27). The interpretations in the present study are based on the reference levels for the biochemical variables obtained in normal human brain (14). According to these reference values, the upper
Mitochondrial dysfunction in SAH cutoff level for LP ratio was set at 30 (mean reference level + 2SD) and the lower cutoff level for pyruvate was set at 70 (mean 2SD). As several previous studies have used an upper cutoff level for LP ratio at 40 (7, 16, 17), this level was included in Table 2. As shown in the table, there were no apparent differences in the studied biochemical variables between these two chosen cutoff levels. Both cutoff levels exhibited significant differences regarding glucose and LP ratio compared with the ischemic group. However, lactate tended to be higher with a cutoff level of 40. A metabolic pattern similar to that observed in the present study was recently described in a study of brain trauma patients combining data from microdialysis, PbtO2, and computed tomography cerebral perfusion (28). It was concluded that increased cerebral lactate was predominantly associated with activated glycolysis rather than hypoxia. It is well established that cerebral ischemia may cause mitochondrial dysfunction (21). We hypothesize that in the present series of patients, mitochondrial dysfunction was caused by episodes of ischemia that had preceded microdialysis. An example of this phenomenon was illustrated in Fig. 3. Further, we assume that, as illustrated in Fig. 4, mitochondrial dysfunction may lead to a reduced tolerance to a decrease in cerebral blood flow and result in cerebral ischemia. Study limitations
The report is based on a retrospective interpretation of biochemical data obtained bedside from a subset of patients treated for SAH. The patients were selected for microdialysis by the attending neurosurgeon from a much larger group of SAH patients. The selection was based on the assumption that these patients were at increased risk for secondary complications. Accordingly, the group includes a high proportion of patients treated with open surgery and evacuation of large hematomas (Table 1). The analyzed group also includes a high number of large aneurysms of the MCA. The relative frequencies of ischemia and mitochondrial dysfunction may accordingly not be representative of an unselected group of SAH patients. Conclusions
This retrospective interpretation of microdialysis data from SAH patients documents that mitochondrial dysfunction is a common cause of increase in LP ratio and ‘metabolic crisis.’ The fact that mitochondrial dysfunction appears to be
more frequent than ischemia in patients with SAH is of clinical importance. It seems probable that mitochondrial dysfunction increases tissue sensitivity to secondary adverse events such as DCI. Bedside diagnosis and separation of the two conditions by utilizing microdialysis offer a possibility to evaluate new therapies developed to mitigate mitochondrial dysfunction (29–31). Acknowledgments We thank Katarina Nielsen, Department of Neurosurgery, Lund University Hospital, for help with the microdialysis equipment and biochemical analyses. The study was supported by grants from Lund University Hospital, Odense University Hospital, and University of Southern Denmark.
Conflict of interest All authors declare that they have no conflict of interest.
References 1. DIRINGER MN, BLECK TP, HEMPHILL CJ III et al. Critical care management of patients following aneurysmal subarachnoid hemorrhage: recommendations from the Neurocritical Care Society’s Multidisciplinary Consensus Conference. Neurocrit Care 2011;15:211–40. 2. WASHINGTON CW, ZIPFEL GJ. Detection and monitoring of vasospasm and delayed cerebral ischemia: a review and assessment of the literature. Neurocrit Care 2011; 15:312–7. 3. SKJØTH-RASMUSSEN J, SCHULZ M, KRISTENSEN SR, BJERRE P. Delayed neurological deficits detected by an ischemic pattern in the extracellular cerebral metabolites in patients with aneurysmal subarachnoid hemorrhage. J Neurosurg 2004;100:8–15. 4. PLUTA RM, HANSEN-SCHWARTZ J, DREIER J et al. Cerebral vasospasm following subarachnoid hemorrhage: time for a new world of thought. Neurol Res 2009;31: 151–8. 5. ETMINAN N, VERGOUWEN MD, ILODIGWE D, MACDONALD RL. Effect of pharmaceutical treatment on vasospasm, delayed cerebral ischemia, and clinical outcome in patients with aneurysmal subarachnoid hemorrhage: a systematic review and meta-analysis. J Cereb Blood Flow Metab 2011;31:1443–51. 6. SCHMIDT JM, KO SB, HELBOK R et al. Cerebral perfusion pressure thresholds for brain tissue hypoxia and metabolic crisis after poor-grade subarachnoid hemorrhage. Stroke 2011;42:1351–6. 7. CHEN HI, STIEFEL MF, ODDO M et al. Detection of cerebral compromise with multimodality monitoring in patients with subarachnoid hemorrhage. Neurosurgery 2011;69:53–63. 8. NIELSEN TH, BINDSLEV TT, PEDERSEN SM, TOFT P, OLSEN € NV, NORDSTROM CH. Cerebral energy metabolism during induced mitochondrial dysfunction. Acta Anaesthesiol Scand 2013;57:229–35. € CH. Cere9. NIELSEN TH, OLSEN NV, TOFT P, NORDSTROM bral energy metabolism during mitochondrial dysfunction induced by cyanide in piglets. Acta Anaesthesiol Scand 2013;57:793–801.
7
Jacobsen et al. € 10. GARDENFORS A, NILSSON F, SKAGERBERG G, UNGERSTEDT € U, NORDSTROM CH. Cerebral physiological and biochemical changes during vasogenic brain oedema induced by intrathecal injection of bacterial lipopolysaccharides in piglets. Acta Neurochir (Wien) 2002;144:601–8. W, STAHL N, TOFT P, REINSTRUP 11. NIELSEN TH, SCHALEN € P, NORDSTROM CH. Bedside diagnosis of mitochondrial dysfunction after malignant middle cerebral artery infarctions. Neurocrit Care 2013; doi: 10.1007/6 s12028013-9875-5 [Epub ahead of print]. 12. HILLMAN J, FRIDRIKSSON S, NILSSON O, YU Z, SAVELAND H, JAKOBSSON KE. Immediate administration of tranexamic acid and reduced incidence of early rebleeding after aneurysmal subarachnoid hemorrhage: a prospective randomized study. J Neurosurg 2002;97:771–8. € P, ROMNER B, SAVELAND H, 13. KRONVALL E, UNDREN CRONQVIST M, NILSSON OG. Nimodipine in aneurysmal subarachnoid hemorrhage: a randomized study of intravenous or peroral administration. J Neurosurg 2009;110: 58–63. N, MELLERGARD P, USKI T, UNGER14. REINSTRUP P, STAHL € CH. Intracerebral microdialysis in STEDT U, NORDSTROM clinical practice: baseline values for chemical markers during wakefulness, anesthesia, and neurosurgery. Neurosurgery 2000;47:701–9. 15. JENNETT B, TEASDALE G. Aspects of coma after severe head injury. Lancet 1977;23:878–81. 16. VESPA P, BERGSNEIDER M, HATTORI N. Metabolic crisis without brain ischemia is common after traumatic brain injury: a combined microdialysis and positron emission tomography study. J Cereb Blood Flow Metab 2005;25: 763–74. 17. VESPA PM, O’PHELAN K, MCARTHUR D. Pericontusional brain tissue exhibits persistent elevation of lactate/pyruvate ratio independent of cerebral perfusion pressure. Crit Care Med 2007;35:1153–60. 18. MACMILLAN VH. Cerebral energy metabolism in cyanide encephalopathy. J Cereb Blood Flow Metab 1989;9:156–62. € CH, REHNCRONA S, SIESJO€ BK. Restitution 19. NORDSTROM of cerebral energy state, as well as of glycolytic metabolites, citric acid cycle intermediates and associated amino acids after 30 minutes of complete ischemia in rats anaesthetized with nitrous oxide or phenobarbital. J Neurochem 1978;30:479–86. € CH, REHNCRONA S, SIESJO€ BK. Effects of 20. NORDSTROM phenobarbital in cerebral ischemia. Part II: restitution of cerebral energy state, as well as of glycolytic metabolites,
8
21.
22.
23.
24. 25.
26.
27.
28.
29.
30.
31.
citric acid cycle intermediates and associated amino acids after pronounced incomplete ischemia. Stroke 1978;9: 335–43. REHNCRONA S, MELA L, S IESJO€ BK. Recovery of brain mitochondrial function in the rat after complete and incomplete cerebral ischemia. Stroke 1979; 10:437–46. PIERRE K, PELLERIN L. Monocarboxylate transporters in the central nervous system: distribution, regulation and function. J Neurochem 2005;94:1–14. I, NORD A, BOTTALICO B et al. Fetal cereAMER-WAHLIN bral energy metabolism and electrocardiogram during experimental umbilical cord occlusion and resuscitation. J Matern Fetal Neonatal Med 2010;23:158–66. SIESJO€ BK. Brain energy metabolism. Chichester, New York, Brisbane, Toronto: Wiley, 1978. € N, MELLERGARD P, HALLSTROM A, UNGERSTEDT STAHL € U, NORDSTROM CH. Intracerebral microdialysis and bedside biochemical analysis in patients with fatal traumatic brain lesions. Acta Anaesthesiol Scand 2001;45: 977–85. GRANDE € € € € OM T, HALLSTROM A, UNGERSTEDT U, BACKSTR € PO, MELLERGARD P, NORDSTROM CH. Microdialysis in normal and injured human brain. In: Kinney JM, Tucker HN, eds. Physiology, stress, and malnutrition: functional correlates. Philadelphia, New York: Nutritional Intervention. Lippincott-Raven Publishers, 1997; 361–74. HILLERED L, VALTYSSON J, ENBLAD P, PERSSON L. Interstitial glycerol as a marker for membrane phospholipid degradation in the acutely injured human brain. J Neurol Neurosurg Psychiatry 1998;64:486–91. SALA N, SUYS T, ZERLAUTH JB et al. Cerebral extracellular lactate increase is predominantly nonischemic in patients with severe traumatic brain injury. J Cereb Blood Flow Metab 2013;33:1815–22. UCHINO H, ELMER E, UCHINO K et al. Amelioration by cyclosporin A of brain damage in transient forebrain ischemia in the rat. Brain Res 1998;812:216–26. AKDEMIR G, ERGUNGOR MF, SEZER M, ALBAYRAK L, DAGLIOGLU E, KILINC K. Therapeutic efficacy of intraventricular cyclosporine A and methylprednisolone on a global cerebral ischemia model in rats. Neurol Res 2005;27:827–34. HWANG JH, LEE JH, LEE KH et al. Cyclosporine A attenuates hypoxic-ischemic brain injury in newborn. Brain Res 2010;1359:208–15.
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd ACTA NEUROLOGICA SCANDINAVICA
Bedside diagnosis of mitochondrial dysfunction in aneurysmal subarachnoid hemorrhage Jacobsen A, Nielsen TH, Nilsson O, Schalen W, Nordstr€ om CH. Bedside diagnosis of mitochondrial dysfunction in aneurysmal subarachnoid hemorrhage. Acta Neurol Scand: DOI: 10.1111/ane.12258. © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd. Objectives – Aneurysmal subarachnoid hemorrhage (SAH) is frequently associated with delayed neurological deterioration (DND). Several studies have shown that DND is not always related to vasospasm and ischemia. Experimental and clinical studies have recently documented that it is possible to diagnose and separate cerebral ischemia and mitochondrial dysfunction bedside. The study explores whether cerebral biochemical variables in SAH patients most frequently exhibit a pattern indicating ischemia or mitochondrial dysfunction. Methods – In 55 patients with severe SAH, intracerebral microdialysis was performed during neurocritical care with bedside analysis and display of glucose, pyruvate, lactate, glutamate, and glycerol. The biochemical patterns observed were compared to those previously described in animal studies of induced mitochondrial dysfunction as well as the pattern obtained in patients with recirculated cerebral infarcts. Results – In 29 patients, the biochemical pattern indicated mitochondrial dysfunction while 10 patients showed a pattern of cerebral ischemia, six of which also exhibited periods of mitochondrial dysfunction. Mitochondrial dysfunction was observed during 5162 h. An ischemic pattern was obtained during 688 h. Four of the patients (40%) with biochemical signs of ischemia died at the neurosurgical department as compared with three patients (10%) in the group of mitochondrial dysfunction. Conclusions – The study documents that mitochondrial dysfunction is a common cause of disturbed cerebral energy metabolism in patients with SAH. Mitochondrial dysfunction may increase tissue sensitivity to secondary adverse events such as vasospasm and decreased cerebral blood flow. The separation of ischemia and mitochondrial dysfunction bedside by utilizing microdialysis offers a possibility to evaluate new therapies.
Introduction
Aneurysmal subarachnoid hemorrhage (SAH) is frequently associated with delayed neurological deterioration (DND) due to delayed cerebral ischemia (DCI; 1, 2). As ischemia is associated with well-defined changes in energy metabolism, it might be expected that DCI would be detected early by intracerebral microdialysis provided the catheters are positioned in the areas of interest (3). However, several studies have shown that
A. Jacobsen1, T. H. Nielsen1, O. Nilsson2, W. Schalen2, C. H. Nordstr€om1 1 Department of Neurosurgery, Odense University Hospital, Odense, Denmark; 2Department of Neurosurgery, Lund University Hospital, Lund, Sweden
Key words: cerebral energy metabolism; microdialysis; mitochondrial dysfunction; ischemia; lactate; pyruvate; subarachnoid hemorrhage Carl-Henrik Nordstr€om Persikev€agen 77, SE 22355 Lund, Sweden Tel.: +46 70 5655785, +46 46 146974 e-mail: [email protected] Accepted for publication April 3, 2014
DND is not always related to vasospasm and ischemia which might explain why therapies directed toward DCI may be ineffective (4, 5). It has repeatedly been reported that ‘metabolic crisis’ may occur although brain tissue oxygenation (PbtO2) appears to be adequate (6): In one study, PbtO2 was normal in approximately half of the episodes of compromised cerebral energy metabolism (7). Accordingly it is of clinical interest to identify the mechanisms underlying the observed ‘metabolic crisis.’ 1
Jacobsen et al. In a series of experimental studies, we have characterized the changes in variables related to cerebral energy metabolism after induced mitochondrial dysfunction (8–10). Briefly, mitochondrial dysfunction is biochemically characterized by increased lactate/pyruvate (LP) ratio due to a marked increase in cerebral lactate at a normal or elevated pyruvate level. This pattern is different from that obtained during cerebral ischemia which is characterized by a marked increase in LP ratio at a pronounced decrease in intracerebral pyruvate level. We have recently identified a similar pattern of mitochondrial dysfunction in patients with recirculated large infarcts after prolonged cerebral ischemia (11). In this study of non-traumatic SAH, we have monitored variables related to cerebral energy metabolism utilizing routine microdialysis with bedside biochemical analysis. The study aims at exploring whether in these patients, the biochemical ‘metabolic crisis’ usually reflects DCI or mitochondrial dysfunction. Materials and methods Patients
The study includes all 55 SAH patients monitored with intracerebral microdialysis admitted to the Department of Neurosurgery, Lund University Hospital, Sweden, between January 2001 and August 2009. During the period, altogether 1240 patients with spontaneous SAH were admitted to the department. Patients with ruptured saccular aneurysms (n = 792) were treated either with endovascular (n = 642; 81%) or open surgical (n = 150; 19%) occlusion of the aneurysm. Patients included in the study were selected for intracerebral microdialysis due to the severity of the initial clinical picture including evaluation of the neurological state, the amount of subarachnoidal blood, and the presence of a focal hematoma. Accordingly, a high proportion of the patients (44%) were treated with surgical evacuation of large intracerebral hematomas. All patients were treated according to the routine procedures at the department (12, 13). The present report is a retrospective interpretation of the data obtained at the bedside initiated by the recently described possibility to diagnose mitochondrial dysfunction and separate it from DCI (8–11). Microdialysis technique
Microdialysis was performed utilizing CMA 70 catheters (CMA Microdialysis, Stockholm, Swe2
den). The microdialysis catheters were perfused (Perfusion Fluid; CMA Microdialysis) at a rate of 0.3 ll/min, and the perfusates were collected in capped microvials at a 1-h interval (14). The samples were immediately analyzed utilizing conventional enzymatic techniques (CMA 600 or ISCUS Microdialysis Analyzer; CMA Microdialysis), and the results were displayed on a bedside monitor. The perfusates were analyzed for glucose, pyruvate, lactate, glutamate, and glycerol. All data obtained from cerebral microdialysis were integrated with global biochemical and physiological data utilizing a specially developed computer program (ICUpilot; CMA Microdialysis). The study includes altogether 12,171 microdialysis samples and approximately 60,000 bedside biochemical analyses. All microdialysis catheters were inserted in the operation theater. In patients subjected to open surgery, the microdialysis catheters were inserted after clipping of the aneurysm and evacuation of intracerebral hematomas. In these cases, one or multiple catheters were placed in the exposed cerebral tissue. Fig. 1 shows a patient with two microdialysis catheters placed in the right frontal lobe after clipping of an aneurysm of the anterior communicating artery and partial evacuation of a large intracerebral hematoma in the left frontal lobe. In patients not subjected to open surgery, the microdialysis catheter was inserted when there was reason to insert a ventricular catheter for monitoring of the intracranial pressure (ICP) and/or drainage of intraventricular fluid. In these cases, a separate burr hole adjacent to that used for insertion of the intraventricular catheter was used and the microdialysis catheter was positioned in the frontal lobe. Altogether 78 microdialysis catheters were inserted: 33 patients received one catheter, 21 patients two catheters and one patient three catheters. The Ethical Committee of Lund University Medical Faculty approved the use of multiple intracerebral microdialysis catheters in the present study. Classification of ischemia and mitochondrial dysfunction
The biochemical classification of the patients was based on the reference values obtained in the non-sedated normal human brain (14). For the LP ratio, the upper normal level was set at 30 (normal mean + 2SD) and the lower normal level for pyruvate was set at 70 (normal mean 2SD). As a LP ratio of 40 has been used as an upper level in previous publications from other centers (7), our data are also presented with LP ratio 40 as an upper cutoff level. In accordance with our
Mitochondrial dysfunction in SAH
Figure 1. CT scanning after open surgery with clips occlusion (C) of an aneurysm of the anterior communicating artery and partial evacuation of bifrontal intracerebral hematomas. Two intracerebral microdialysis catheters (A and B) were inserted into the basal right frontal lobe. The tip of the ventricular catheter (D) is visualized.
previous experimental and clinical studies, patients with a LP ratio >30 and a pyruvate level 30 and >40, respectively) and pyruvate >70 lM were classified as mitochondrial dysfunction (8, 9, 11). Data from the patients were included into the groups when the criteria defined above were fulfilled during three subsequent measurements (3 h) or more. Statistics
All values are expressed as mean and standard deviation (SD) or median (interquartile rage) as appropriate. Statistical comparison of the mean levels of microdialysis variables between the group defined as ischemia (n = 12), and the group defined as mitochondrial dysfunction with cutoff levels LP ratio >30 (n = 37) and LP ratio >40 (n = 24), respectively, was performed by utilizing the non-parametric Wilcoxon rank-sum test. A P-value below 0.05 was considered significant. As the P-values are to be considered exploratory, no adjustment for generating multiple P-values was performed. STATA 11.1 statistical software package (StataCorp LP, College Station, TX, USA) was used for data analysis. Results Patient characteristics
Basic clinical and radiographic characteristics are given in Table 1. Mortality refers to the number of deaths at the department of neurosurgery. At the time of admission, the following neurological
state was noted according to Glasgow Coma Score (GCS; 15): GCS 15 16%; GCS 9–14 36%, GCS ≤8 47%. All patients were classified as Fisher grade 3 or 4. In six patients, no aneurysm treatment was performed. In three of these patients, no aneurysm was found on angiography and none of these patients developed biochemical signs of mitochondrial dysfunction or ischemia. In three patients, no angiography was performed as—due to the poor condition of the patients— surgical evacuation of a large hematoma and clipping of an middle cerebral artery (MCA) aneurysm was performed immediately after CTscanning. Variables during ischemia and mitochondrial dysfunction
The patients were classified into the two groups of cerebral ischemia and mitochondrial dysfunction, respectively, according to the principles defined previously. The two conditions may occur in the same patient at different points of time (see below). This was observed in altogether six patients. In Table 1, these six patients are included in the group of mitochondrial dysfunction (n = 29) as well as in the group of ischemia (n = 10). In 22 patients, the biochemical parameters did not fulfill the criteria for inclusion into the two groups. The clinical data for these patients are given in Table 1. Table 1 gives mortality for the three groups during treatment in the neurocritical care ward. One of the patients in the group of mitochondrial dysfunction died after having developed a biochemical pattern of ischemia. In the table, this patient was included into both groups. The 3
Jacobsen et al. Table 1 Characteristics for patients included in the study
Age years, mean (SD) Sex M/F, n Mortality, n (%) Aneurysm location, n (%) ACom A, Pericallosal A Carotid A, PostCom A MCA Posterior Circ No aneurysm found Angiography not performed Treatment, n (%) Endovascular Open surgical clipping No aneurysm treatment Evacuation of hematoma
Total n = 55
Mitochondrial dysfunction n = 29
Ischemia n = 10
No ischemia/mitoch. dysf. n = 22
55 (12) 15/40 8 (15)
57 (10) 8/21 3 (10)
53 (8) 2/8 4 (40)
51 (14) 7/15 2 (9)
15 8 24 2 3 3
(27) (15) (44) (4) (5) (5)
6 3 17 1 0 1
(21) (10) (81) (5)
21 28 6 24
(38) (51) (11) (44)
10 18 1 15
(10) (20) (50)
(5)
1 2 5 0 0 2
(36) (14) (32) (5) (14)
(20)
8 3 7 1 3 0
(34) (62) (3) (52)
3 5 2 7
(30) (50) (20) (70)
9 10 3 7
(41) (45) (14) (32)
n refers to the number of patients within each category. Six patients are included in the group of mitochondrial dysfunction as well as in the group of ischemia (see text). Ischemia is defined as lactate/pyruvate (LP) ratio >30 at a pyruvate level 30 at a normal or increased pyruvate level. AComA, anterior communicating artery; Pericallosal A, pericallosal artery; Carotid A, carotid artery; PostComA, posterior communicating artery; MCA, middle cerebral artery; Posterior Circ, posterior circulation.
pattern of cerebral ischemia was observed during altogether 688 h, the pattern of mitochondrial dysfunction with cutoff level LP ratio 30 was observed during 5162 h, and the pattern of mitochondrial dysfunction with cutoff level LP ratio 40 was observed during 2920 h. Table 2 gives median values (interquartile range) for all biochemical variables displayed beside. Data are given for all microdialysis catheters (n = 78), separately for those classified as cerebral ischemia (n = 12 catheters) and mitochondrial dysfunction with cutoff levels LP ratio >30 (n = 37 catheters) and LP ratio >40 (n = 24 catheters). As described above, the data were included into the diagnostic groups when they fulfilled the criteria for inclusion during three consequent measurements (3 h) or more. The table also gives corre-
sponding data for patients who neither exhibited a pattern of ischemia nor a pattern of mitochondrial dysfunction (n = 29). In addition, reference levels from normal human brain are given as mean (SD) (14). In the table, statistical comparisons between the two groups defined as mitochondrial dysfunction and the group defined as ischemia are included. According to the definitions of the present study, the pyruvate level was higher in patients in the mitochondrial dysfunction group and the LP ratio was significantly lower than in the ischemic group. The levels of glucose were significantly higher in the two groups of mitochondrial dysfunction. The levels of lactate were higher in the two groups of mitochondrial dysfunction, but the difference did not quite reach statistical significance (P = 0.156
Table 2 Median value (interquartile range) for biochemical variables in patients with subarachnoid hemorrhage and biochemical indications of cerebral ischemia, mitochondrial dysfunction, and no ischemia/mitochondrial dysfunction, respectively
Ischemia (n = 12) Mitoch. dysf. LP ratio >30 (n = 37) Mitoch. dysf. LP ratio >40 (n = 24) No ischemia/mitoch. dysf. (n = 29) Reference (normal) level
Glycerol (lM)
LP ratio
Glucose (mM)
Lactate (mM)
Pyruvate (lM)
Glutamate (lM)
166 (81–502)
0.3 (0.1–0.7)
4.7 (2.7–7.4)
34 (15–51)
96 (19–432)
536 (392–750)
41 P 53 P 21
1.3 (0.6–1.9) P = 0.0014 1.2 (0.5–1.9) P = 0.008 1.4 (0.9–2.0)
7.6 (6.1–9.4) P = 0.156 8.5 (6.8–11.0) P = 0.06 2.4 (1.9–3.5)
178 (133–232)
115 (89–156)
22 (7–72) P = 0.020 42 (7–97) P = 0.08 4 (3–6)
89 P 98 P 78
1.7 (0.9)
2.9 (0.9)
166 (47)
16 (16)
82 (44)
(34–53) < 0.001 (45–68) < 0.001 (18–25)
23 (4)
152 (114–203)
(51–188) < 0.001 (56–198) < 0.001 (44–152)
Data for mitochondrial dysfunction are given for two cutoff levels: lactate/pyruvate (LP) ratio >30 and >40, respectively. n refers to the number of microdialysis catheters within each category. Reference (normal) levels are given as mean values (SD) in accordance with the original data obtained from Reinstrup et al. (14). Statistical comparison performed between ischemia and mitochondrial dysfunction with LP cutoff levels >30 and >40, respectively. As the pyruvate level was used for the classification into the groups of ischemia and non-ischemic mitochondrial dysfunction, no statistical comparison was performed regarding this variable.
4
Mitochondrial dysfunction in SAH and P = 0.06 for LP ratio cutoff at 30 and 40, respectively). Glutamate was significantly lower in the group of mitochondrial dysfunction with a LP ratio cutoff level of 30 but not 40, while glycerol was significantly higher in the ischemic group compared to both groups of mitochondrial dysfunction. As discussed above, the patterns of ischemia and mitochondrial dysfunction were in six cases noted at different points of time in the same patient. We present three representative patterns: (i) a pattern typical of progressive cerebral ischemia; (ii) transfer of cerebral ischemia into a pattern of mitochondrial dysfunction; and (iii) change in mitochondrial dysfunction into cerebral ischemia. Biochemical pattern of cerebral ischemia
Fig. 2 illustrates the biochemical pattern of progressive cerebral ischemia. This 64-year-old woman was admitted in deep coma with a large hematoma in the left Sylvian fissure. She was taken to surgery without preceding angiography. A giant aneurysm of the MCA was occluded, and the hematoma was evacuated. One microdialysis catheter was placed in the basal frontal lobe, and one was placed in the tip of the temporal lobe. Fig. 2 shows the data from the catheter in the frontal lobe. A few hours after surgery, the LP ratio started to increase and intracerebral pyruvate decreased to 30–40 lM. Intracerebral glucose simultaneously decreased to a subnormal level. During the time period in Fig. 2, lactate remained very high (11.0 1.2 mmol). The illustrated increase in LP ratio preceded an increase in ICP (to 25–30 mmHg) by approximately 4 h.
Figure 2. Biochemical pattern of progressive cerebral ischemia. The lactate/pyruvate (LP) ratio increases to a very high level as pyruvate and glucose decrease to below normal. The ranges of the variables (mean SD) for normal human brain (14) are indicated on the respective y-axis.
Transfer from ischemia to mitochondrial dysfunction
Fig. 3 illustrates the transfer from cerebral ischemia to mitochondrial dysfunction. This 54-yearold man was operated with occlusion of an aneurysm of the left MCA and evacuation of a large hematoma. The microdialysis catheter was positioned in the frontal lobe. After surgery, the LP ratio was initially very high, while pyruvate and glucose were below normal limit. During the following 48 h, the LP ratio decreased to 50–60 due to a continuous increase in pyruvate to supranormal levels while the lactate level remained at a very high level (12.2 1.6 mM). As shown in the figure, the initially low level of glucose normalized within the initial 12 h. During the period illustrated in Fig. 3, ICP was below 10 mmHg and cerebral perfusion pressure was above 70 mmHg. Transfer of mitochondrial dysfunction to cerebral ischemia
Fig. 4 illustrates the transfer from mitochondrial dysfunction to a pattern typical of cerebral ischemia. This 58-year-old man was operated with occlusion of an aneurysm of the right MCA and evacuation of an intracerebral hematoma. During surgery, the MCA was temporarily occluded for 7 + 4 min. The microdialysis catheter was placed in the frontal lobe close to the Sylvian fissure. The figure illustrates the biochemical pattern 122 h after surgery: The LP ratio was 60, while
Figure 3. Biochemical pattern of transfer from ischemia to non-ischemic mitochondrial dysfunction. During the initial 4–6 h, the pattern indicates ischemia: The lactate/pyruvate (LP) ratio is very high, and pyruvate and glucose are very low. The subsequent increase in glucose is paralleled by an increase in pyruvate. As the interstitial lactate level remained very high (not shown in the figure), the LP ratio remained elevated although interstitial pyruvate increased above normal. The ranges of the variables (mean SD) for normal human brain (14) are indicated on the respective y-axis.
5
Jacobsen et al.
Figure 4. Biochemical pattern of transfer from non-ischemic mitochondrial dysfunction to ischemia. About 120–130 h after surgery lactate/pyruvate (LP), ratio remains elevated (60–70) although pyruvate level is within normal range. The late decrease in pyruvate causes a marked increase in LP ratio. The simultaneous increase in interstitial glycerol level indicates secondary cellular degradation. The ranges of the variables (mean SD) for normal human brain (14) are indicated on the respective y-axis.
pyruvate was within normal limits. Later than 140 h after surgery, the LP ratio increased due to a marked decrease in pyruvate level to below normal. The increase in LP ratio was associated with a marked increase in intracerebral glycerol concentration, while lactate remained stable at a high level (8.6 0.8 mM). Discussion
In this study, we utilize the recently described possibility to diagnose and separate cerebral ischemia and mitochondrial dysfunction to estimate the relative frequency of the two conditions in a group of patients with severe SAH (8, 9, 11). The interpretations are based on the biochemical patterns displayed at the bedside during routine intracerebral microdialysis. It has previously been noted that a pattern of cerebral ‘metabolic crisis’ obtained during microdialysis is not always associated with reduction in cerebral blood flow or tissue oxygenation (7, 16, 17). The interpretations of the biochemical data in the present report are based on experimental studies obtained after induced mitochondrial dysfunction (8–10, 18). This biochemical pattern and its association with mitochondrial dysfunction in vitro have also been described after transient experimental cerebral ischemia (19–21). The biochemical interpretations are based on the hypothesis that cerebral interstitial LP ratio reflects cytoplasmic redox state. Lactate and pyruvate pass cell membranes by proton-linked 6
monocarboxylate transporters (MCTs; 22). MCTs are found in various tissues, and three isoforms have been described in the rodent brain. The efficacy of the MTCs in transmembrane transport of lactate and pyruvate appears to be high, and it has been documented that cerebral ischemia causes an instantaneous increase in interstitial LP ratio that reflects the intracellular shift in redox state (23). As the capacity of the monocarboxylate transporters across cell membranes is high, the LP ratio obtained from microdialysis of interstitial fluid can be used to evaluate cytoplasmic redox state (24). In cerebral ischemia, tissue supply of oxygen and glucose is compromised and the LP ratio momentarily increases to a very high level due to the increase in lactate and the marked decrease in pyruvate (23–25). In experimentally induced mitochondrial dysfunction, the increase in LP ratio occurs at a normal or increased pyruvate level (8–10). This pattern is also observed after transient experimental brain ischemia (19, 20) and following recirculation of large cerebral infarcts in humans (11). In the present group of patients, mitochondrial dysfunction was observed more frequently than a pattern of cerebral ischemia (Table 1). The fact that cerebral ‘metabolic crisis’ may occur in the absence of reduction in cerebral blood flow and tissue oxygenation has previously been observed in SAH patients as well as in brain trauma (7, 16, 17). The present analysis offers an explanation of this phenomenon and indicates that the ‘metabolic crisis’ described in previous studies may reflect mitochondrial dysfunction. The mitochondrial dysfunction observed in these clinical conditions may have been caused by a previous period of transient ischemia (11, 19–21). This situation is illustrated in Fig. 3 and shows that an initial period of cerebral ischemia was succeeded by a lasting increase in LP ratio while the levels of pyruvate and glucose increased to normal or supra-normal levels. Mitochondrial dysfunction may deteriorate into a second period of ischemia as illustrated in Fig. 4. In this patient, a biochemical pattern of cerebral ischemia was obtained 140 h after the initial insult in an area previously exhibiting a pattern of mitochondrial dysfunction. The ischemic development was associated with a pronounced increase in glycerol indicating degradation of cellular membranes (26, 27). The interpretations in the present study are based on the reference levels for the biochemical variables obtained in normal human brain (14). According to these reference values, the upper
Mitochondrial dysfunction in SAH cutoff level for LP ratio was set at 30 (mean reference level + 2SD) and the lower cutoff level for pyruvate was set at 70 (mean 2SD). As several previous studies have used an upper cutoff level for LP ratio at 40 (7, 16, 17), this level was included in Table 2. As shown in the table, there were no apparent differences in the studied biochemical variables between these two chosen cutoff levels. Both cutoff levels exhibited significant differences regarding glucose and LP ratio compared with the ischemic group. However, lactate tended to be higher with a cutoff level of 40. A metabolic pattern similar to that observed in the present study was recently described in a study of brain trauma patients combining data from microdialysis, PbtO2, and computed tomography cerebral perfusion (28). It was concluded that increased cerebral lactate was predominantly associated with activated glycolysis rather than hypoxia. It is well established that cerebral ischemia may cause mitochondrial dysfunction (21). We hypothesize that in the present series of patients, mitochondrial dysfunction was caused by episodes of ischemia that had preceded microdialysis. An example of this phenomenon was illustrated in Fig. 3. Further, we assume that, as illustrated in Fig. 4, mitochondrial dysfunction may lead to a reduced tolerance to a decrease in cerebral blood flow and result in cerebral ischemia. Study limitations
The report is based on a retrospective interpretation of biochemical data obtained bedside from a subset of patients treated for SAH. The patients were selected for microdialysis by the attending neurosurgeon from a much larger group of SAH patients. The selection was based on the assumption that these patients were at increased risk for secondary complications. Accordingly, the group includes a high proportion of patients treated with open surgery and evacuation of large hematomas (Table 1). The analyzed group also includes a high number of large aneurysms of the MCA. The relative frequencies of ischemia and mitochondrial dysfunction may accordingly not be representative of an unselected group of SAH patients. Conclusions
This retrospective interpretation of microdialysis data from SAH patients documents that mitochondrial dysfunction is a common cause of increase in LP ratio and ‘metabolic crisis.’ The fact that mitochondrial dysfunction appears to be
more frequent than ischemia in patients with SAH is of clinical importance. It seems probable that mitochondrial dysfunction increases tissue sensitivity to secondary adverse events such as DCI. Bedside diagnosis and separation of the two conditions by utilizing microdialysis offer a possibility to evaluate new therapies developed to mitigate mitochondrial dysfunction (29–31). Acknowledgments We thank Katarina Nielsen, Department of Neurosurgery, Lund University Hospital, for help with the microdialysis equipment and biochemical analyses. The study was supported by grants from Lund University Hospital, Odense University Hospital, and University of Southern Denmark.
Conflict of interest All authors declare that they have no conflict of interest.
References 1. DIRINGER MN, BLECK TP, HEMPHILL CJ III et al. Critical care management of patients following aneurysmal subarachnoid hemorrhage: recommendations from the Neurocritical Care Society’s Multidisciplinary Consensus Conference. Neurocrit Care 2011;15:211–40. 2. WASHINGTON CW, ZIPFEL GJ. Detection and monitoring of vasospasm and delayed cerebral ischemia: a review and assessment of the literature. Neurocrit Care 2011; 15:312–7. 3. SKJØTH-RASMUSSEN J, SCHULZ M, KRISTENSEN SR, BJERRE P. Delayed neurological deficits detected by an ischemic pattern in the extracellular cerebral metabolites in patients with aneurysmal subarachnoid hemorrhage. J Neurosurg 2004;100:8–15. 4. PLUTA RM, HANSEN-SCHWARTZ J, DREIER J et al. Cerebral vasospasm following subarachnoid hemorrhage: time for a new world of thought. Neurol Res 2009;31: 151–8. 5. ETMINAN N, VERGOUWEN MD, ILODIGWE D, MACDONALD RL. Effect of pharmaceutical treatment on vasospasm, delayed cerebral ischemia, and clinical outcome in patients with aneurysmal subarachnoid hemorrhage: a systematic review and meta-analysis. J Cereb Blood Flow Metab 2011;31:1443–51. 6. SCHMIDT JM, KO SB, HELBOK R et al. Cerebral perfusion pressure thresholds for brain tissue hypoxia and metabolic crisis after poor-grade subarachnoid hemorrhage. Stroke 2011;42:1351–6. 7. CHEN HI, STIEFEL MF, ODDO M et al. Detection of cerebral compromise with multimodality monitoring in patients with subarachnoid hemorrhage. Neurosurgery 2011;69:53–63. 8. NIELSEN TH, BINDSLEV TT, PEDERSEN SM, TOFT P, OLSEN € NV, NORDSTROM CH. Cerebral energy metabolism during induced mitochondrial dysfunction. Acta Anaesthesiol Scand 2013;57:229–35. € CH. Cere9. NIELSEN TH, OLSEN NV, TOFT P, NORDSTROM bral energy metabolism during mitochondrial dysfunction induced by cyanide in piglets. Acta Anaesthesiol Scand 2013;57:793–801.
7
Jacobsen et al. € 10. GARDENFORS A, NILSSON F, SKAGERBERG G, UNGERSTEDT € U, NORDSTROM CH. Cerebral physiological and biochemical changes during vasogenic brain oedema induced by intrathecal injection of bacterial lipopolysaccharides in piglets. Acta Neurochir (Wien) 2002;144:601–8. W, STAHL N, TOFT P, REINSTRUP 11. NIELSEN TH, SCHALEN € P, NORDSTROM CH. Bedside diagnosis of mitochondrial dysfunction after malignant middle cerebral artery infarctions. Neurocrit Care 2013; doi: 10.1007/6 s12028013-9875-5 [Epub ahead of print]. 12. HILLMAN J, FRIDRIKSSON S, NILSSON O, YU Z, SAVELAND H, JAKOBSSON KE. Immediate administration of tranexamic acid and reduced incidence of early rebleeding after aneurysmal subarachnoid hemorrhage: a prospective randomized study. J Neurosurg 2002;97:771–8. € P, ROMNER B, SAVELAND H, 13. KRONVALL E, UNDREN CRONQVIST M, NILSSON OG. Nimodipine in aneurysmal subarachnoid hemorrhage: a randomized study of intravenous or peroral administration. J Neurosurg 2009;110: 58–63. N, MELLERGARD P, USKI T, UNGER14. REINSTRUP P, STAHL € CH. Intracerebral microdialysis in STEDT U, NORDSTROM clinical practice: baseline values for chemical markers during wakefulness, anesthesia, and neurosurgery. Neurosurgery 2000;47:701–9. 15. JENNETT B, TEASDALE G. Aspects of coma after severe head injury. Lancet 1977;23:878–81. 16. VESPA P, BERGSNEIDER M, HATTORI N. Metabolic crisis without brain ischemia is common after traumatic brain injury: a combined microdialysis and positron emission tomography study. J Cereb Blood Flow Metab 2005;25: 763–74. 17. VESPA PM, O’PHELAN K, MCARTHUR D. Pericontusional brain tissue exhibits persistent elevation of lactate/pyruvate ratio independent of cerebral perfusion pressure. Crit Care Med 2007;35:1153–60. 18. MACMILLAN VH. Cerebral energy metabolism in cyanide encephalopathy. J Cereb Blood Flow Metab 1989;9:156–62. € CH, REHNCRONA S, SIESJO€ BK. Restitution 19. NORDSTROM of cerebral energy state, as well as of glycolytic metabolites, citric acid cycle intermediates and associated amino acids after 30 minutes of complete ischemia in rats anaesthetized with nitrous oxide or phenobarbital. J Neurochem 1978;30:479–86. € CH, REHNCRONA S, SIESJO€ BK. Effects of 20. NORDSTROM phenobarbital in cerebral ischemia. Part II: restitution of cerebral energy state, as well as of glycolytic metabolites,
8
21.
22.
23.
24. 25.
26.
27.
28.
29.
30.
31.
citric acid cycle intermediates and associated amino acids after pronounced incomplete ischemia. Stroke 1978;9: 335–43. REHNCRONA S, MELA L, S IESJO€ BK. Recovery of brain mitochondrial function in the rat after complete and incomplete cerebral ischemia. Stroke 1979; 10:437–46. PIERRE K, PELLERIN L. Monocarboxylate transporters in the central nervous system: distribution, regulation and function. J Neurochem 2005;94:1–14. I, NORD A, BOTTALICO B et al. Fetal cereAMER-WAHLIN bral energy metabolism and electrocardiogram during experimental umbilical cord occlusion and resuscitation. J Matern Fetal Neonatal Med 2010;23:158–66. SIESJO€ BK. Brain energy metabolism. Chichester, New York, Brisbane, Toronto: Wiley, 1978. € N, MELLERGARD P, HALLSTROM A, UNGERSTEDT STAHL € U, NORDSTROM CH. Intracerebral microdialysis and bedside biochemical analysis in patients with fatal traumatic brain lesions. Acta Anaesthesiol Scand 2001;45: 977–85. GRANDE € € € € OM T, HALLSTROM A, UNGERSTEDT U, BACKSTR € PO, MELLERGARD P, NORDSTROM CH. Microdialysis in normal and injured human brain. In: Kinney JM, Tucker HN, eds. Physiology, stress, and malnutrition: functional correlates. Philadelphia, New York: Nutritional Intervention. Lippincott-Raven Publishers, 1997; 361–74. HILLERED L, VALTYSSON J, ENBLAD P, PERSSON L. Interstitial glycerol as a marker for membrane phospholipid degradation in the acutely injured human brain. J Neurol Neurosurg Psychiatry 1998;64:486–91. SALA N, SUYS T, ZERLAUTH JB et al. Cerebral extracellular lactate increase is predominantly nonischemic in patients with severe traumatic brain injury. J Cereb Blood Flow Metab 2013;33:1815–22. UCHINO H, ELMER E, UCHINO K et al. Amelioration by cyclosporin A of brain damage in transient forebrain ischemia in the rat. Brain Res 1998;812:216–26. AKDEMIR G, ERGUNGOR MF, SEZER M, ALBAYRAK L, DAGLIOGLU E, KILINC K. Therapeutic efficacy of intraventricular cyclosporine A and methylprednisolone on a global cerebral ischemia model in rats. Neurol Res 2005;27:827–34. HWANG JH, LEE JH, LEE KH et al. Cyclosporine A attenuates hypoxic-ischemic brain injury in newborn. Brain Res 2010;1359:208–15.
